Synthesis, Molecular Docking Analysis and Biological Evaluations of Saccharide-Modified Thiadiazole Sulfonamide Derivatives

A series of saccharide-modified thiadiazole sulfonamide derivatives has been designed and synthesized by the “tail approach” and evaluated for inhibitory activity against carbonic anhydrases II, IX, and XII. Most of the compounds showed high topological polar surface area (TPSA) values and excellent enzyme inhibitory activity. The impacts of some compounds on the viability of HT-29, MDA-MB-231, and MG-63 human cancer cell lines were examined under both normoxic and hypoxic conditions, and they showed certain inhibitory effects on cell viability. Moreover, it was found that the series of compounds had the ability to raise the pH of the tumor cell microenvironment. All the results proved that saccharide-modified thiadiazole sulfonamides have important research prospects for the development of CA IX inhibitors.


Introduction
Carbonic anhydrases (CAs, EC 4.2.1.1) are lyases that are widely distributed in all organisms from archaea to higher animals. According to the different coding genes, CAs can be divided into eight families: α-, β-, γ-, δ-, ζ-, η-, θ-, and ι-CAs. Among the eight genetically different CA families, human carbonic anhydrases (hCAs) belong to the αfamily [1][2][3][4]. hCAs are a class of zinc metalloenzymes with 15 subtypes. In the human body, their main physiological roles are as catalysts for the reversible interconversion of carbon dioxide and bicarbonate [5][6][7]. CA IX is a special isoform of the hCAs family. Its expression in normal tissues is limited to gastrointestinal epithelial cells but it is overexpressed in many solid tumors, contributing to the growth advantage of cancer cells [8][9][10][11]. Therefore, CA IX is considered an important target for tumor treatment, and hCA modulators are considered promising drugs in clinical applications [12]. At present, CA IX-targeted drugs have been studied in clinical research. The CA IX inhibitor SLC-0111 combined with gemcitabine for the treatment of metastatic pancreatic ductal cancer in patients overexpressing the CA IX isoform has entered clinical phase II [13][14][15].
X-ray crystallography studies showed that the catalytic domains of the hCA family are highly similar, which causes great difficulties for the search for selective CA inhibitors [6]. To alleviate the clinical side effects, carbonic anhydrase inhibitors such as brinzolamide, a drug for glaucoma, are administered locally [16]. Local administration is difficult for CA IX inhibitors, and the resulting side effects seriously hinder the development of selective CA IX inhibitors. Until the "tail approach" was proposed, it provided a favorable weapon for selective CA IX inhibitors [17]. The "tail approach" is a simple and effective method that refers to appending a "tail" with a wide spectrum of chemical functionalities to the Zn 2+ binding group through a simple chemical reaction to change the mode of action of provided a favorable weapon for selective CA IX inhibitors [17]. The "tail approach" is a simple and effective method that refers to appending a "tail" with a wide spectrum of chemical functionalities to the Zn 2+ binding group through a simple chemical reaction to change the mode of action of the inhibitor in different isoforms of CAs. A proper "tail" structure is of great significance for the research on selective CA IX inhibitors. One of the characteristics of CA IX is that its subcellular localization is outside the cell membrane, while its common off-target isoform CA II is located in the cytoplasm (Figure 1). The "tail" modification of appropriate fragments can improve the polarity of inhibitors to reduce cell membrane permeability, which is conducive to obtaining CA IX selectivity. Saccharides have the characteristics of low toxicity, high polarity, and high solubility. In the area of medicinal chemistry, glycol derivatives and glycosyl modification compounds have shown a wide range of pharmacological activities [18][19][20][21][22]. In the research on CA IX inhibitors, saccharide-modified compounds have achieved remarkable success, and various molecules of saccharide-modified CA IX inhibitors have been reported [23][24][25]. The high polarity of saccharide-modified compounds puts them at a disadvantage when they penetrate the cell membrane, which is conducive to the selectivity. At the same time, the hydrogen bonding between the saccharide fragment and the residues in the hydrophilic pocket at the rim of the enzymatic cavity is beneficial to improvement of the activity. To investigate the application of saccharide modifications in CA IX inhibitors, we designed and synthesized sixteen novel compounds in this work by linking glucose fragments and thiadiazole sulfonamides. These compounds were then tested for their inhibitory effects on the tumor-associated proteins hCA IX and XII and off-target hCA II subtypes. Their inhibitory activities were comparable to that of the positive control AZM in the low nanomolar region. The study of molecular docking studies and the summary of preliminary structure-activity relationships have provided a basis for further research.

Chemistry
The general synthetic route for the synthesis of the target saccharide-modified thiadiazole sulfonamide derivatives is shown in Scheme 1. We adopted a convergent strategy to complete the syntheses of compounds, which mainly include two parts: saccharide fragments and small molecule fragments. In the saccharide fragment, glucose was chosen as the starting material, and intermediate 4 was synthesized from glucose by benzoylation (2), bromination (3) and azide substitution. After hydrogen reduction with Pd/C as the catalyst, the product underwent a one pot reaction with succinic anhydride or glutaric anhydride to generate saccharide intermediates 5a and 5b. In the small molecule fragment, different substituted aminobenzoic acids 6a-h were converted into To investigate the application of saccharide modifications in CA IX inhibitors, we designed and synthesized sixteen novel compounds in this work by linking glucose fragments and thiadiazole sulfonamides. These compounds were then tested for their inhibitory effects on the tumor-associated proteins hCA IX and XII and off-target hCA II subtypes. Their inhibitory activities were comparable to that of the positive control AZM in the low nanomolar region. The study of molecular docking studies and the summary of preliminary structure-activity relationships have provided a basis for further research.

Chemistry
The general synthetic route for the synthesis of the target saccharide-modified thiadiazole sulfonamide derivatives is shown in Scheme 1. We adopted a convergent strategy to complete the syntheses of compounds, which mainly include two parts: saccharide fragments and small molecule fragments. In the saccharide fragment, glucose was chosen as the starting material, and intermediate 4 was synthesized from glucose by benzoylation (2), bromination (3) and azide substitution. After hydrogen reduction with Pd/C as the catalyst, the product underwent a one pot reaction with succinic anhydride or glutaric anhydride to generate saccharide intermediates 5a and 5b. In the small molecule fragment, different substituted aminobenzoic acids 6a-h were converted into the corresponding acid chlorides after protection of the amino groups. Acid chlorides 8a-h reacted with intermediate 10 to give the corresponding intermediates 11a-h. Intermediates 11a-h were deprotected to obtain small molecule intermediates 12a-h. the corresponding acid chlorides after protection of the amino groups. Acid chlorides 8a-h reacted with intermediate 10 to give the corresponding intermediates 11a-h. Intermediates 11a-h were deprotected to obtain small molecule intermediates 12a-h. Scheme  In the presence of EDCI, saccharide intermediates (5a/5b) and small molecule intermediates (12a-h) were condensed to obtain intermediates 13a-p. Finally, the target compounds 14a-p were obtained by deprotection (some compounds contained a few amounts of stereoisomers).

Carbonic Anhydrase Inhibition
The inhibitory activities of the novel saccharide-modified thiadiazole sulfonamide derivatives 14a-p and small molecule intermediates 12a-h toward three physiologically relevant hCA isoforms, hCA II (common off-target cytoplasmic isoform), hCA IX, and XII (tumor-associated membrane-bound isoforms), were evaluated by monitoring the hydrolysis of 4-nitrophenyl acetate (4-NPA). AZM was chosen as the positive control drug. The inhibition data are summarized in Table 1. In the presence of EDCI, saccharide intermediates (5a/5b) and small molecule intermediates (12a-h) were condensed to obtain intermediates 13a-p. Finally, the target compounds 14a-p were obtained by deprotection (some compounds contained a few amounts of stereoisomers).

Carbonic Anhydrase Inhibition
The inhibitory activities of the novel saccharide-modified thiadiazole sulfonamide derivatives 14a-p and small molecule intermediates 12a-h toward three physiologically relevant hCA isoforms, hCA II (common off-target cytoplasmic isoform), hCA IX, and XII (tumor-associated membrane-bound isoforms), were evaluated by monitoring the hydrolysis of 4-nitrophenyl acetate (4-NPA). AZM was chosen as the positive control drug. The inhibition data are summarized in Table 1. Referring to the data in Table 1, the structure-activity relationships (SAR) of intermediates 12a-h and compounds 14a-p are preliminarily summarized below.
(i). The IC 50 values of intermediates 12a-h against hCA II and IX were 79.1-701.3 and 36.4-65.9 nM, respectively. Among them, 12b showed the strongest inhibitory activity against hCA II, with an IC 50 of 79.1 nM. The substitution of halogen or electron withdrawing groups led to a decrease in hCA II inhibitory activity. Intermediate 12h showed the strongest inhibitory activity with hCA IX, with an IC 50 of 36.4 nM. The effect of different substitutions on hCA IX inhibitory activity was not obvious. In the in vitro enzyme inhibitory activity test, 12a-h generally showed selectivity for hCA IX, and 12f exhibited the best selectivity for hCA IX; its inhibitory activity for hCA IX was 10 times higher than that for hCA II; (ii). The IC 50 values of saccharide-modified compounds 14a-p against hCA II and IX were 80.3-1403.1 and 29.0-1082.0 nM, respectively; 14b exhibited the strongest inhibitory activity against hCA II, with an IC 50 of 80.3 nM, while 14a and 14b had the best inhibitory activity against hCA IX, with an IC 50 of 29.0 nM. Compared with 12a-h, the TPSA values of compounds 14a-p were significantly increased, which was beneficial to improving the selectivity through the differences in subcellular localization of isoforms. However, the inhibitory activity against hCA II and IX decreased to varying degrees, and the inhibitory activity against hCA II decreased significantly; (iii). Among the compounds 14a-p, 14a-b and 14i-j, which did not feature substitution on the benzene ring, were the compounds with the best hCA II inhibitory activities when the aliphatic chain lengths were n = 2 and 3, respectively. Whether the benzene ring was substituted by an electron-donating group or electron-donating group, the inhibitory activity of the compounds against hCA II decreased, especially for the substituents Cl, Br and NO 2 . The activity data showed that for hCA II, the inhibitory activities of compounds 14a-h with aliphatic chain length n = 3 were generally higher than those of compounds 14i-p with n = 2; (iv). Compared with hCA II, compounds 14a-p generally had higher inhibitory activity against hCA IX, among which 14a-c, 14g-j and 14p exhibited more desirable IC 50 values of less than 100 nM. When the benzene ring was substituted by F or an electron donating group, the inhibitory activity of the compounds toward hCA IX did not change significantly, but when the substitution was for Cl, Br, or -NO 2 , the inhibitory activity was obviously reduced. Similar to hCA II, the inhibitory activities toward hCA IX for the compounds with aliphatic chain length n = 3 were higher than those of the compounds with n = 2; (v). Compounds 14a-c, 14g-j, and 14p with good inhibitory activity with hCA IX were tested for their inhibitory activities with hCA XII. The values of IC 50 ranged from 40.6 to 140.8 nM. All tested compounds showed strong hCA XII inhibitory activity, of which that of 14j was the most prominent.
In summary, this series of compounds showed significant inhibition against the tumorrelated isoforms hCA IX and XII, with high TPSA and potential selective qualifications, which are worthy of further study.

Studies on Docking into the Active Site of hCA IX
We firstly re-docked the native ligands into CA IX (PDB ID: 5FL4) with optimized docking parameters, and successfully produced the binding modes observed in the crystal structures, with RMSD values of 0.619 Å for the docked and experimental poses. As compounds 14a, 14e, and 14i were CA IX inhibitors; their binding mode to the active site of CA IX was investigated by a docking simulation using Autodock 4(Zn). The results are shown in Figure 2. (v). Compounds 14a-c, 14g-j, and 14p with good inhibitory activity with hCA IX were tested for their inhibitory activities with hCA XII. The values of IC50 ranged from 40.6 to 140.8 nM. All tested compounds showed strong hCA XII inhibitory activity, of which that of 14j was the most prominent.
In summary, this series of compounds showed significant inhibition against the tumor-related isoforms hCA IX and XII, with high TPSA and potential selective qualifications, which are worthy of further study.

Studies on Docking into the Active Site of hCA IX
We firstly re-docked the native ligands into CA IX (PDB ID: 5FL4) with optimized docking parameters, and successfully produced the binding modes observed in the crystal structures, with RMSD values of 0.619 Å for the docked and experimental poses. As compounds 14a, 14e, and 14i were CA IX inhibitors; their binding mode to the active site of CA IX was investigated by a docking simulation using Autodock 4(Zn). The results are shown in Figure 2. By analyzing the docking results of 14a, the oxygen atom of benzamide formed two hydrogen bonds interaction with Gln71 and Gln92 in CA IX (Figure 2A,D). The sulfonamide group was positioned in the bottom of the cavity, interacting strongly H-bonding interactions with His119 and Thr200, respectively. The ring of benzamide formed strong π-π stacking interaction with His68. Meanwhile, the region of the thiazole sulfonamide formed hydrophobic interactions with His94, His96, Leu199, and Trp210. In the docking results of 14e ( Figure 2B,E), the oxygen atom of benzamide formed two hydrogen bonds interaction with Gln71 in CA IX. The thiazolesulfonamide group was positioned in the bottom of the cavity, having strong H-bonding interactions with His119, Thr200, and Thr201, respectively. Meanwhile, the thiazolesulfonamide group also formed hydrophobic interactions with His94, His96 and Leu199. In the docking By analyzing the docking results of 14a, the oxygen atom of benzamide formed two hydrogen bonds interaction with Gln71 and Gln92 in CA IX (Figure 2A,D). The sulfonamide group was positioned in the bottom of the cavity, interacting strongly H-bonding interactions with His119 and Thr200, respectively. The ring of benzamide formed strong π-π stacking interaction with His68. Meanwhile, the region of the thiazole sulfonamide formed hydrophobic interactions with His94, His96, Leu199, and Trp210. In the docking results of 14e ( Figure 2B,E), the oxygen atom of benzamide formed two hydrogen bonds interaction with Gln71 in CA IX. The thiazolesulfonamide group was positioned in the bottom of the cavity, having strong H-bonding interactions with His119, Thr200, and Thr201, respectively. Meanwhile, the thiazolesulfonamide group also formed hydrophobic interactions with His94, His96 and Leu199. In the docking results of 14i ( Figure 2C,F), the oxygen atom of hydroxyl formed hydrogen bonds interaction with Asp131. The oxygen atom of ben-zamide interacted H-bonding interactions with Gln71 and Gln92. The thiadiazole group locating in the middle of the cavity formed hydrogen bonds interaction with Gln92 and Thr201. The sulfonamide group was positioned in the bottom of the cavity, interacting strongly H-bonding interactions with Thr200. The ring of thiadiazole formed strong π-π stacking interaction with His94. At the same time, the region of the thiazolesulfonamide formed hydrophobic interactions with His94, His96, Val121, and Leu199. Importantly, the sulfonamide group formed a coordination bond with the zinc atom, which was reflected in the three compounds playing a crucial role in the activity. Afterwards, in order to verify the accuracy of the docking, we conducted dynamic simulations on them.

Molecular Dynamics Simulations
The 100 ns Molecular dynamics (MD) simulations were performed with three selected CA IX-compound complexes (14a, 14e, and 14i) to examine their interactions between receptors and ligands and elucidate the contribution of key residues during the binding process.
As shown in Figure 3, the RMSD and RMSF values of three complexes were monitored during the whole MD simulations. In Figure 3A, the CA IX-compound 14a complex reached equilibrium after 65 ns, where the protein RMSD value fluctuated around 5.5-7.0 Å. The CA IX-compound 14i complex reached equilibrium after 60 ns, where the protein RMSD value fluctuated around 6.0-7.0 Å. Throughout the MD simulation, 14e showed obvious fluctuations and did not reach equilibrium, indicating that 14e had poor binding affinity and the result was consistent with enzyme inhibitory activity. The RMSF plots were displayed in Figure 3B, the fluctuations of majority residues were less than 2 Å, and several regions with huge fluctuations located in the active site. results of 14i ( Figure 2C,F), the oxygen atom of hydroxyl formed hydrogen bonds interaction with Asp131. The oxygen atom of benzamide interacted H-bonding interactions with Gln71 and Gln92. The thiadiazole group locating in the middle of the cavity formed hydrogen bonds interaction with Gln92 and Thr201. The sulfonamide group was positioned in the bottom of the cavity, interacting strongly H-bonding interactions with Thr200. The ring of thiadiazole formed strong π-π stacking interaction with His94. At the same time, the region of the thiazolesulfonamide formed hydrophobic interactions with His94, His96, Val121, and Leu199. Importantly, the sulfonamide group formed a coordination bond with the zinc atom, which was reflected in the three compounds playing a crucial role in the activity. Afterwards, in order to verify the accuracy of the docking, we conducted dynamic simulations on them.

Molecular Dynamics Simulations
The 100 ns Molecular dynamics (MD) simulations were performed with three selected CA IX-compound complexes (14a, 14e, and 14i) to examine their interactions between receptors and ligands and elucidate the contribution of key residues during the binding process.
As shown in Figure 3, the RMSD and RMSF values of three complexes were monitored during the whole MD simulations. In Figure 3A, the CA IX-compound 14a complex reached equilibrium after 65 ns, where the protein RMSD value fluctuated around 5.5-7.0 Å. The CA IX-compound 14i complex reached equilibrium after 60 ns, where the protein RMSD value fluctuated around 6.0-7.0 Å. Throughout the MD simulation, 14e showed obvious fluctuations and did not reach equilibrium, indicating that 14e had poor binding affinity and the result was consistent with enzyme inhibitory activity. The RMSF plots were displayed in Figure 3B, the fluctuations of majority residues were less than 2 Å, and several regions with huge fluctuations located in the active site.  As shown in Figure 4, the histogram showed the type and ratio of interactions between receptors and ligands, including hydrogen bonds, hydrophobic contacts, ionic interactions, and water bridges. In Figure 4A, compound 14a formed hydrogen bonds with residues Arg64, Gln71, Gln92, Ala128, Arg129, Val130, Asp131, Glu173, Thr200, and Thr201, and formed water bridges with residues Arg64, Gln71, Gln92, Ala128, Arg129, Val130, Asp131, and Pro202. The hydrophobic interactions were with residues Val130 and Leu199. Among them, 14a formed strong hydrogen bonds with Thr200, which accounted for 99% of the simulation time, which played a crucial role in activities. In Figure 4B, compound 14e formed hydrogen bonds with residues Tyr11, Arg64, Gln71, Gln92, Glu173, Thr200, and Thr201, and formed water bridges with residues Tyr11, Arg64, Asn66, Asn68, Gln71, Gln92, Glu173, and Thr201. The hydrophobic interactions were with residues Trp9 and His94. Among them, 14e formed strong hydrogen bonds with Thr200, which accounted for 99% of the simulation time. In addition, water bridges were also significant for the interactions, such as the carbonyl oxygen formed water bridges with Tyr11, Arg64, Asn66, and Gln71, which accounted for 40%, 40%, 35% and 42% of the simulation time, respectively. In Figure 4C, compound 14i formed hydrogen bonds with residues Arg64, Asn66, Gln71, Gln92, Val130, Asp131, Glu173, Thr200, and Thr201, and formed water bridges with residues Tyr11, Arg64, Asn66, Asn68, Gln71, Gln92, Val130, Asp131, and Glu173. The hydrophobic interactions were with residues Trp9, His94, and Val130. Among them, 14i formed strong hydrogen bonds with Thr200, which accounted for 99% of the simulation time. Meanwhile, residue Tyr 11 formed water bridge with 14i accounting for 50% of the simulation time. As shown in Figure 4, the histogram showed the type and ratio of interactions between receptors and ligands, including hydrogen bonds, hydrophobic contacts, ionic interactions, and water bridges. In Figure 4A, compound 14a formed hydrogen bonds with residues Arg64, Gln71, Gln92, Ala128, Arg129, Val130, Asp131, Glu173, Thr200, and Thr201, and formed water bridges with residues Arg64, Gln71, Gln92, Ala128, Arg129, Val130, Asp131, and Pro202. The hydrophobic interactions were with residues Val130 and Leu199. Among them, 14a formed strong hydrogen bonds with Thr200, which accounted for 99% of the simulation time, which played a crucial role in activities. In Figure 4B, compound 14e formed hydrogen bonds with residues Tyr11, Arg64, Gln71, Gln92, Glu173, Thr200, and Thr201, and formed water bridges with residues Tyr11, Arg64, Asn66, Asn68, Gln71, Gln92, Glu173, and Thr201. The hydrophobic interactions were with residues Trp9 and His94. Among them, 14e formed strong hydrogen bonds with Thr200, which accounted for 99% of the simulation time. In addition, water bridges were also significant for the interactions, such as the carbonyl oxygen formed water bridges with Tyr11, Arg64, Asn66, and Gln71, which accounted for 40%, 40%, 35% and 42% of the simulation time, respectively. In Figure 4C, compound 14i formed hydrogen bonds with residues Arg64, Asn66, Gln71, Gln92, Val130, Asp131, Glu173, Thr200, and Thr201, and formed water bridges with residues Tyr11, Arg64, Asn66, Asn68, Gln71, Gln92, Val130, Asp131, and Glu173. The hydrophobic interactions were with residues Trp9, His94, and Val130. Among them, 14i formed strong hydrogen bonds with Thr200, which accounted for 99% of the simulation time. Meanwhile, residue Tyr 11 formed water bridge with 14i accounting for 50% of the simulation time.  The results of MD simulations indicated 14a and 14i were more stable during MD simulations and revealing that 14a and 14i might have stronger interactions with CA IX. In addition, compound 14e had the largest RMSD fluctuation, this might be due to its weak contact with CA IX which made the complex unstable, and could explain its lowest activity among three compounds.

In Vitro Cytotoxicity Studies on Cancer Cells
To evaluate the ability of novel saccharide-modified compounds to inhibit the viability of tumor cells, several compounds that exhibited the best inhibitory profiles were chosen to evaluate their effects against human colon cancer HT-29 cells and breast adenocarcinoma MDA-MB-231 cells via CCK-8 assay. The results are shown in Figure 5.
The results of MD simulations indicated 14a and 14i were more stable during MD simulations and revealing that 14a and 14i might have stronger interactions with CA IX. In addition, compound 14e had the largest RMSD fluctuation, this might be due to its weak contact with CA IX which made the complex unstable, and could explain its lowest activity among three compounds.

In Vitro Cytotoxicity Studies on Cancer Cells
To evaluate the ability of novel saccharide-modified compounds to inhibit the viability of tumor cells, several compounds that exhibited the best inhibitory profiles were chosen to evaluate their effects against human colon cancer HT-29 cells and breast adenocarcinoma MDA-MB-231 cells via CCK-8 assay. The results are shown in Figure 5. As shown in Figure 5, compounds 14a, 14b, 14h, 14p, and AZM were evaluated for their effects against two cancer cell lines that overexpress CA IX under normoxic and hypoxic conditions. Based on the results, we summarize the following conclusions: The inhibitory activities of compounds 14a, 14b, 14h, and 14p on the two tumor cell lines were better than those of the positive control drug AZM; (ii).
Under hypoxic conditions, the inhibitory activities of the tested compounds and AZM on the tumor cell lines were all higher than those under normoxia; (iii).
Compared with the breast adenocarcinoma MDA-MB-231, the tested compounds generally showed higher inhibitory efficacy with the colon cancer cell line HT-29.
Then, considering the high inhibitory activity of 14a on the two isoforms of CA IX and XII, we evaluated the inhibitory activity of 14a on human osteosarcoma MG-63 featuring high expression of CA IX and XII. The results are shown in Figure 6. As shown in Figure 5, compounds 14a, 14b, 14h, 14p, and AZM were evaluated for their effects against two cancer cell lines that overexpress CA IX under normoxic and hypoxic conditions. Based on the results, we summarize the following conclusions: (i). The inhibitory activities of compounds 14a, 14b, 14h, and 14p on the two tumor cell lines were better than those of the positive control drug AZM; (ii). Under hypoxic conditions, the inhibitory activities of the tested compounds and AZM on the tumor cell lines were all higher than those under normoxia; (iii). Compared with the breast adenocarcinoma MDA-MB-231, the tested compounds generally showed higher inhibitory efficacy with the colon cancer cell line HT-29.
Then, considering the high inhibitory activity of 14a on the two isoforms of CA IX and XII, we evaluated the inhibitory activity of 14a on human osteosarcoma MG-63 featuring high expression of CA IX and XII. The results are shown in Figure 6. Similar to the previous conclusions, the effect of 14a in inhibiting t MG-63 cells under hypoxic conditions was more obvious than that und conditions. Under hypoxic conditions, the administration of 800 μM 14a viability of MG-63 cells by approximately 30%, and the effect was signif than that of the positive control AZM.
Although these novel saccharide-modified CA IX inhibitors showe isoform inhibitory activities, their inhibitory effect on the cell viability of t cells was not obvious. This is a common problem in research on CA IX inhi This might indicate that CA IX inhibitors are not effective enough to be u cancer therapy [25].

Extracellular pH Measurement in the Presence of Compound 14b, 14h, and 14
The high expression of CA IX plays an important role in the mainte tumor cell microenvironment, and inhibiting its activity might have a potent the pH of the tumor microenvironment. To evaluate the effect of compoun of the microenvironment, the pH values under normoxic and hypoxic con determined in the presence of compounds 14b, 14h, and 14p. The results 14b, 14h, and 14p increased tumor pH to varying degrees at concentration mM (Figure 7). The tested compounds had no obvious effect on cells u conditions, but they significantly improved extracellular acidosis induced b addition, this improvement in pH was dose-dependent. Similar to the previous conclusions, the effect of 14a in inhibiting the vitality of MG-63 cells under hypoxic conditions was more obvious than that under normoxic conditions. Under hypoxic conditions, the administration of 800 µM 14a reduced the viability of MG-63 cells by approximately 30%, and the effect was significantly better than that of the positive control AZM.
Although these novel saccharide-modified CA IX inhibitors showed promising isoform inhibitory activities, their inhibitory effect on the cell viability of tumor-related cells was not obvious. This is a common problem in research on CA IX inhibitors [26,27]. This might indicate that CA IX inhibitors are not effective enough to be used alone in cancer therapy [25].

Extracellular pH Measurement in the Presence of Compound 14b, 14h, and 14p
The high expression of CA IX plays an important role in the maintenance of the tumor cell microenvironment, and inhibiting its activity might have a potential impact on the pH of the tumor microenvironment. To evaluate the effect of compounds on the pH of the microenvironment, the pH values under normoxic and hypoxic conditions were determined in the presence of compounds 14b, 14h, and 14p. The results showed that 14b, 14h, and 14p increased tumor pH to varying degrees at concentrations of 0.1 and 1 mM (Figure 7). The tested compounds had no obvious effect on cells under normal conditions, but they significantly improved extracellular acidosis induced by hypoxia. In addition, this improvement in pH was dose-dependent.

Chemistry
All the reagents (Energy Chemical, Shanghai, China) were used without further purification unless otherwise specified. Solvents were dried and redistilled prior to use in the usual manner. Analytical TLC was per-formed using silica gel HF254 (Qingdao Haiyang Chemical, Qingdao, Shandong, China). Preparative column chromatography was performed with silica gel H. Melting points were obtained on a Büchi melting point B-540 apparatus (Büchi Labortechnik, Flawil, Switzerland). 1 H and 13 C NMR spectra (details of raw data for compounds, see Figures S1-S54) were recorded on a Bruker ARX 600 or 400 MHz spectrometer (Bruker, Zurich, Switzerland) HRMS were obtained on an Agilent ESI-QTOF instrument (Agilent, Santa Clara, CA, USA).

Chemistry
All the reagents (Energy Chemical, Shanghai, China) were used without further purification unless otherwise specified. Solvents were dried and redistilled prior to use in the usual manner. Analytical TLC was per-formed using silica gel HF 254 (Qingdao Haiyang Chemical, Qingdao, Shandong, China). Preparative column chromatography was performed with silica gel H. Melting points were obtained on a Büchi melting point B-540 apparatus (Büchi Labortechnik, Flawil, Switzerland). 1 H and 13 C NMR spectra (details of raw data for compounds, see Figures S1-S54) were recorded on a Bruker ARX 600 or 400 MHz spectrometer (Bruker, Zurich, Switzerland) HRMS were obtained on an Agilent ESI-QTOF instrument (Agilent, Santa Clara, CA, USA).

Measurement of Extracellular pH
The measurement of the change of extracellular pH referred to the reported literature [26]. In brief, cancer cells were incubated for 12 h. After removing the medium, fresh medium was added to the plates and then incubated for 48 h under hypoxia or normoxia. The pH of the medium was measured immediately. The tested compounds were dissolved in DMSO and diluted with culture medium to the desired concentration. After the addition of tested compounds, the same procedure was repeated.

Conclusions
Herein, a series of saccharide-modified thiadiazole sulfonamide derivatives were reported. Intermediates 12a-h and compounds 14a-p were assayed for their inhibitory activities against three hCA isoforms of pharmacological relevance: tumor-associated transmembrane isoforms hCA IX and XII and cytosolic off-target isoform hCA II. The results indicated that all the tested compounds showed activity against hCA II, IX, and XII. Among them, the IC 50 values for the saccharide-modified derivatives with high TPSA values for hCA II and IX ranged from 80.3-1403.1 nM and 29.0-1082.0 nM, respectively. Molecular docking results showed that the interaction with residue Gln92 and the distance between sulfanilamide and Zn 2+ had important impacts on the activity. Further cell experiments showed that compounds 14a, 14b, 14h, and 14p had inhibitory effects on the viability of cancer cell lines with high expression of CA IX, and this viability inhibition was enhanced under hypoxic conditions. Compound 14a had better inhibitory activity than the positive control AZM on MG-63 cells with high expression of CA IX and XII. In addition, compounds 14b, 14h, and 14p significantly modified the acidic microenvironments of cancer cells. Based on the above results and the known importance of the acidic microenvironment for tumor growth, we believe that saccharide-modified CA IX inhibitors have prospects for further development as combination drugs for tumor treatment.