*Article* **Hepatic Transcriptome Analysis Provides New Insight into the Lipid-Reducing Effect of Dietary Taurine in High–Fat Fed Groupers (***Epinephelus coioides***)**

**Mingfan Chen 1,†, Fakai Bai 1,†, Tao Song <sup>1</sup> , Xingjian Niu <sup>1</sup> , Xuexi Wang <sup>2</sup> , Kun Wang <sup>1</sup> and Jidan Ye 1,\***


**Abstract:** A transcriptome analysis was conducted to provide the first detailed overview of dietary taurine intervention on liver lipid accumulation caused by high–fat in groupers. After an eight-week feeding, the fish fed 15% fat diet (High–fat diet) had higher liver lipid contents vs. fish fed 10% fat diet (Control diet). 15% fat diet with 1% taurine (Taurine diet) improved weight gain and feed utilization, and decreased hepatosomatic index and liver lipid contents vs. the High–fat diet. In the comparison of the Control vs. High–fat groups, a total of 160 differentially expressed genes (DEGs) were identified, of which up- and down-regulated genes were 72 and 88, respectively. There were 49 identified DEGs with 26 and 23 of up- and down-regulated in the comparison to High–fat vs. Taurine. Several key genes, such as cysteine dioxygenase (*CDO1*), ADP–ribosylation factor 1/2 (*ARF1\_2*), sodium/potassium–transporting ATPase subunit alpha (*ATP1A*), carnitine/acylcarnitine translocase (*CACT*), and calcium/calmodulin–dependent protein kinase II (*CAMK*) were obtained by enrichment for the above DEGs. These genes were enriched in taurine and hypotaurine metabolism, bile secretion, insulin secretion, phospholipase D signaling pathway, and thermogenesis pathways, respectively. The present study will also provide a new insight into the nutritional physiological function of taurine in farmed fish.

**Keywords:** taurine; fat metabolism; liver fat accumulation; RNA–Seq; *Epinephelus coioides*

#### **1. Introduction**

Taurine is a sulfur–containing amino acid which is the most abundant free amino acid in animals [1]. It is clear that dietary taurine administration can reduce peripheral cholesterol and visceral lipid accumulation of rats and humans by enhancing rate–limiting enzyme activity of cholesterol 7α–hydroxylase in the liver, promote the synthesis of cholic acid from cholesterol, and increase the excretion of fecal cholesterol [2–4]. Taurine is a conditionally essential amino acid for most cultured fish [5]. It plays a range of key roles in fish physiology, including functions in bile acid conjugation, immune regulation, osmoregulation, antioxidation, nervous system development, and regeneration [6,7]. Moreover, the lipid-reducing effect of dietary taurine also occurred in different tissues of many farmed fish [8–11].

At present, one of the particularly concerning issues is fatty liver syndrome caused by widespread use of high–fat feed in intensive aquaculture for the purpose of proteinsparing and feed utilization [12,13]. Despite these benefits, excessive fat intake does cause growth retardation [14,15] and other undesirable effects, such as visceral lipid accumulation

**Citation:** Chen, M.; Bai, F.; Song, T.; Niu, X.; Wang, X.; Wang, K.; Ye, J. Hepatic Transcriptome Analysis Provides New Insight into the Lipid-Reducing Effect of Dietary Taurine in High–Fat Fed Groupers (*Epinephelus coioides*). *Metabolites* **2022**, *12*, 670. https://doi.org/10.3390/ metabo12070670

Academic Editors: Teruo Miyazaki, Takashi Ito, Alessia Baseggio Conrado and Shigeru Murakami

Received: 5 July 2022 Accepted: 19 July 2022 Published: 20 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

and fatty liver [16–19], accompanied by apoptosis and declined immune function [19–21]. Therefore, fatty liver induced by high–fat feeding has become a typical chronic liver disease which is closely associated with nutritional metabolic syndrome in intensive fish farming. There is an urgent need to find a suitable way to solve the problem of fatty livers, which represents a threat to aquaculture production. In the light of the lipid–reducing effect of taurine mentioned above, dietary taurine administration may therefore be a promising way to attenuate the adverse effects caused by high–fat diets.

The orange–spotted grouper (*Epinephelus coioides*) has become an economically important mariculture carnivorous fish species in Southeast Asian countries, including China [22,23]. Previous studies showed that dietary taurine supplementation can attenuate the tissue lipid accumulation of groupers [11,24]. However, the underlying mechanism involved in the regulation of lipid metabolism still remains unclear. Recent studies have shown that dietary taurine supplementation in 15% fat feed could reduce lipid accumulation through reducing the contents of triglyceride molecules containing 18:2n–6 at the *sn*–*2* and *sn*–*3* positions [25] and through accelerating lipid absorption of taurine−conjugated bile acids and fatty acid β-oxidation, and inhibiting lipogenesis [26]. However, determining how dietary taurine modulates lipid metabolism of fish at the transcriptional level has not yet been investigated. In the present study, three experimental diets (10% fat, 15% fat and 15% fat with 1% taurine) were formulated to investigate the lipid–reducing effect of dietary taurine supplementation in regard to taurine-mediated changes of key genes.

#### **2. Results**

#### *2.1. Growth Performance and Tissue Lipid Contents*

The growth performance of groupers is shown in Table 1. When the dietary lipid level increased from 10% to 15%, weight gain, feed conversion ratio and hepatosomatic index of the Control and High–fat groups were not different (*p* > 0.05). However, when 15% lipid diet was added with 1% taurine, weight gain was significantly enhanced, whereas feed conversion ratio and hepatosomatic index were markedly reduced compared with High–fat group (*p* < 0.05).

**Table 1.** Effects of experimental diets on growth performance and the contents of liver and muscle fat of groupers (*E. coioides*).


Weight gain (%) = 100 × (final body weight − initial body weight)/initial body weight; Feed conversion ratio = feed intake/wet weight gain; Hepatosomatic index (%) = 100 × (liver weight/wet body weight); Data are presented as the means ± SEM (*n* = 3 tanks); Data was presented as the means ± SEM (*n* = 30 fish); Values in the same row with different lowercase letter superscripts indicate significant differences (*p* < 0.05); Statistical analysis was performed by one way ANOVA, followed by Student–Neuman–Keuls multiple comparison test.

The lipid contents in the liver and muscle samples in High–fat group were significantly higher (*p* < 0.05) than that in Control group. Fish of Taurine group had lower (*p* < 0.05) liver lipid contents vs those of High–fat group. However, there was no difference in muscle lipid content between High–fat and Taurine groups (*p* > 0.05).

#### *2.2. Illumina Sequencing and De Novo Assembly*

According to the sequencing results of the transcriptome assay, we obtained 399.25 million clean reads (Table S1), with Q20 at between 97.56% and 98.46%, Q30 at between 93.16% and 95.45%, and the GC content between 42.16% and 47.00%. From the assembly results, 99,634 unigenes were obtained with the N50 and N90 of unigenes being 653 and 243 bp respectively (Table 2). The length distribution of transcript and unigenes is shown in Figure S1. After short and low–quality sequences were excluded, 349,211 unigenes were identified and annotated by matching them against the five public databases (NR, SwissProt, PFAM, GO and KO), and 72,679 unigenes had at least one Blast hit against the public databases, yielding 72,125 annotated unigenes for NR (20.65%), 13,104 for SwissProt (3.75%), 22,814 for PFAM (6.53%), 6988 for GO (2.00%) and 5124 for KO (1.47%) (Table 3).

**Table 2.** Summary statistics of the de novo transcriptome assembly of liver samples of *E. coioides*.


Min length, the minimum sequence length in a unigene set; Max length, the maximum sequence length in a unigene set; Mean length, average sequence length of a unigene set; N50/N90, the unigenes were calculated by ordering all sequences, and the length of unigenes was then collected one by one from the longest to the shortest until 50%/90% of the total length was attained.


**Table 3.** The number of annotated genes in different databases.

NR, NCBI non-redundant protein sequence database; SwissProt, protein sequence database; PFAM, a large collection of protein multiple sequence alignments and profile hidden Markov models; GO, gene ontology database; KO, KEGG orthology database for representation of gene/protein functional orthologs in molecular networks; KEGG, kyoto encyclopedia of genes and genome database.

## *2.3. Identification of DEGs*

To identify DEGs in liver samples of grouper after feeding trial, three digital gene expression libraries from the Control, High–fat, and Taurine groups were constructed (Figure 1). There were 160 DEGs identified in the comparison between the High–fat and Control groups, of which 72 were upregulated and 88 were downregulated. In the comparison between the Taurine and High–fat groups, 49 DEGs were identified, of which 26 were upregulated and the rest were downregulated.

**Figure 1.** Comparison of DEGs among liver transcriptome of *E. coioides* of the Control and the High-fat groups, and of the High–fat and the Taurine groups. Control, control diet; High–fat, 15% fat diet; Taurine, 15% fat diet with 1% taurine. (**A**)—volcano plot of DEGs of High-fat vs. Control group; (**B**)—volcano plot of DEGs of Taurine vs. High–fat group.

#### *2.4. Function Annotation and Analysis for Unigenes*

Based upon gene ontology (GO) classification, 6988 (2.00%) unigenes were mapped and clustered into biological processes, cellular components and molecular function categories. In the classification of biological processes, the cellular process and metabolic process occur most frequently. In the cellular components, most unigenes were classified into a cell and cell part. With regard to molecular function, most unigenes were clustered into binding and catalytic activity categories. The GO annotation statistics of Unigenes is shown in Figure 2.

**Figure 2.** Classification of the unigenes and annotation of differentially expressed genes (DEGs) of liver samples of *E. coioides* based on enrichment analysis of gene ontology. All DEGs were enriched into three categorizations: biological processes, cellular components and molecular function categories.

Upon KEGG enrichment analysis, 160 differentially expressed genes (DEGs) in the liver samples were identified in the comparison with the High–fat vs. the Control groups, of which 72 genes were upregulated and 88 genes were downregulated. The typical pathways of DEGs for enrichment were complement and coagulation cascades, respectively (Figure 3). The comparison of Taurine and High–fat groups had 49 DEGs (26 genes were upregulated and the rest were downregulated) for enrichment in the liver. The insulin secretion was the most enriched pathway of DEGs (Figure 3).

#### *2.5. Signaling Pathway Network Related to Lipid Metabolism*

According to the above analysis results, the DEGs were annotated as key genes such as cysteine dioxygenase (*CDO1*), ADP–ribosylation factor 1/2 (*ARF1\_2*), sodium/ potassium–transporting ATPase subunit alpha (ATP1α) in the carnitine/acylcarnitine translocase (CACT) and calcium/calmodulin-dependent protein kinase II (CAMK). These key genes were high–fat induced lipid metabolism genes regulated by dietary taurine in groupers. Several signaling pathways of taurine–mediated lipid metabolism were clustered in Table S2 and mainly included taurine and hypotaurine metabolism, primary bile secretion, insulin secretion, phospholipase D signaling pathway, phosphatidylinositol signaling system, inositol phosphate metabolism and calcium signaling pathway. Furthermore, the signaling pathway network related to taurine–mediated lipid metabolism in the liver was constructed by KEGG functional enrichment pathway conjoint analysis and is presented in Figure 4.

**Figure 3.** Bubble chart of significantly enriched KEGG pathways in the differentially expressed genes (DEGs) of liver samples of *E. coioides* between the Control and High–fat groups, and between High–fat and Taurine groups. (**A**)—pathway enrichment of DEGs of High–fat vs. Control groups; (**B**)—pathway enrichment of DEGs of High–fat vs. Taurine groups. The vertical axis represents the pathway categories, the horizontal axis shows the enrichment factor. The point size shows the number of DEGs enriched in the KEGG pathway. The point color shows different Q values as indicated on the right.

**Figure 4.** Signal pathway network diagram of taurine–mediated lipid metabolism in the liver of *E. coioides*. Abbreviations: AACT, carnitine/acylcarnitine translocase; ARF1\_2, ADP–ribosylation factor 1/2; ATP1*α*, sodium/potassium–transporting ATPase subunit alpha; CAMK, calmodulin–dependent protein kinase II; CDO1, cysteine dioxygenase; GAPN, glyceraldehyde-3-phosphate dehydrogenase; GK, glucokinase; PDK1, pyruvate dehydrogenase kinase isozyme 2; PLCD, phosphatidylinositol phospholipase C.

#### *2.6. qRT–PCR Validation of DEGs*

To validate the reliability of RNA–Seq data, eight randomly selected DEG expression profiles in taurine and high fat comparison groups samples were examined by qRT–PCR. As shown in Figure 5, the fold–changes obtained by qRT–PCR were consistent with the values obtained by RNA–seq for the six selected genes (*ARF1\_2*, *GK*, *ATP1α*, *CAMK*, *CDO1*, and *CACT*), suggesting our RNA-Seq data and the results based on RNA–Seq data analysis were reliable.

**Figure 5.** Validation of RNA–seq data using qRT–PCR in the liver of *E. coioides* in High–fat and Taurine groups. To validate the data from RNA–seq analysis, relative mRNA levels of eight selected differentially expressed genes (*ARF1\_2*, *GK*, *ATP1α*, *CAMK*, *CDO1*, *CACT*, *PDK1*, and *PLCD*) from the liver samples of Control, High–fat and Taurine groups were examined by RT–qPCR. mRNA levels are presented as –fold changes when compared with the Control group after normalization against β-actin. The relative mRNA levels from the RNA–seq analysis were calculated as FPKM values. High–fat, 15% fat diet; Taurine, 15% fat diet with 1% taurine. Student's *t*–test was applied for comparison of High–fat vs. Taurine diets. Values are means ± SEM (*n* = 3). Asterisks (\*, \*\* and \*\*\*) represent significant differences with *p* < 0.05, *p* < 0.01 and *p* < 0.001 respectively.

#### **3. Discussion**

The results of the present study showed that the growth and feed utilization of groupers did not differ when the dietary fat was increased from 10% to 15%, which indicates that the fish species has a certain degree of tolerance to a higher fat diet. Similar results were observed in previous studies of black sea bream (*Acanthopagrus schlegelii*) [27] and large yellow croaker (*Larimichthys crocea*) [28]. However, high–fat diets increased liver lipid contents vs. control diets in this study and previous studies on many farmed fishes [16–18,20,28–30], indicating that feeding high–fat diet led to an increase in liver lipid accumulation of fish.

Interestingly, taurine addition in high-fat diets not only promoted both fish growth and feed utilization in the study and previous studies on many farmed fishes [31–35], but also reduced liver lipid content and/or hepatosomatic index of farmed fishes [10,11,36–40]. All of these results indicate a significant effect of taurine on reducing liver lipid. In view of the lipid–reducing effect of taurine on fish above mentioned, the next objective of this study was to investigate how high–fat diet does affect liver lipid deposition and to explore the lipid-reducing effect of dietary taurine intervention in relation to taurine–mediated changes of gene expression at the transcriptome level.

Herein, we present the first transcriptomic analysis of taurine intervention on the liver lipid metabolism of high–fat fed groupers in an attempt to understand the regulatory mechanism of taurine on liver lipid deposition caused by high–fat feeding. The DEGs were then screened out after the unigenes were mapped to five public databases (NR, SwissProt, PFAM, GO and KO) by the NCBI, and had at least one Blast hit against the public databases. A total of 160 DEGs were identified in the comparison of High–fat and Control groups. In the comparison of High–fat and Taurine groups, a total of 49 DEGs were identified. GO annotation and KEGG pathway analysis of DEGs in the comparison of Taurine and High–fat groups showed that the DEGs were enriched in primary bile secretion, insulin

secretion, phospholipase D signaling pathway and thermogenesis pathways. The discovery of these genes and signaling pathways should contribute to a better understanding of the molecular mechanism of regulation of taurine in fish lipid metabolism.

The cysteine sulfite dependent pathway is the main route of taurine biosynthesis in mammals, in which CDO1, as the rate–limiting enzyme, can produce hypotaurine through oxidation and decarboxylation, then hypotaurine is oxidized to form taurine [41]. Fish also have the capability for biosynthesis of taurine by CDO1 [6]. The activity of CDO1 in fish liver is generally higher than that in other tissues [5]. Dietary taurine addition was reported to upregulate the gene expression of *CDO1* in tissues of Atlantic bluefin tuna (*Thunnus thynnus*) [42], but did not affect its expression in the liver of tiger puffer (*Takifugu rubripes*) [43] and European bass (*Dicentrarchus labrax*) [8]. The dietary taurine promotes liver *CDO1* expression level in a dose–dependent manner [37]. The above results show that the ability of fish to synthesize taurine varies between fish species [37]. In this study, there was no difference in expression level of *CDO1* between Control and High–fat diets, but the Taurine diet resulted in an upregulation in gene expression of *CDO1* compared with the High–fat diet. The *CDO1* gene was enriched in the taurine and hypotaurine metabolic pathway. One possible reason for this is the fact that feeding high–fat diet may increase the demand for taurine of groupers. This indicates that taurine exerts the effect of reducing liver fat by affecting the synthesis and metabolism of taurine.

In the present study, dietary taurine supplementation downregulated the expression of ADP–ribosylation factor (ARF) gene *ARF1\_2* compared with the High–fat group. The *ARF1\_2* gene was enriched into phospholipase D signaling pathway. The *ARF1\_2* gene is a member of the Ras superfamily, and participates in vesicular trafficking and regulating the activation of phospholipase D (PLD), which plays an important role in intracellular signal transduction and substance transport [44]. PLD catalyzes not only the hydrolysis of the phosphodiester bond of glycerophospholipids to generate phosphatidic acid (PA), but also a transphosphatidylation reaction to produce phosphatidylethanol [45]. The PLD activated by ARF1\_2 hydrolyzed phosphatidylcholine to produce more precursor PA required for triglyceride synthesis [46,47]. Therefore, dietary taurine supplementation may reduce the synthesis of triglycerides in the liver of fish through downregulating *ARF1\_2* gene expression.

As a signal molecule, bile acids are able to regulate their own enterohepatic circulation by affecting transcription of the genes critically involved in transport and metabolism [48]. The farnesoid X receptor (FXR) is activated by bile acids [49]. The activated FXR triggers the transcriptional synthesis of bile salt export pump (BSEP) [50]. Due to the high affinity of BSEP to bile acids, the bile acids in the form of complex bile acids and BSEP are transported from hepatocytes to the intestine with the help of the energy generated by ATP hydrolysis. BSEP is therefore considered to be the most important ATP dependent bile acid transporter in the liver [51]. The bile acid metabolism and transport are achieved through the chain reaction triggered by FXR and its downstream BSEP [50,52], so as to maintain the glycolipid metabolic homeostasis of fish [18,53]. Dietary taurine administration ameliorates Na+/K+ATPase impairment in the retina of diabetic rats [54] and promotes gill Na+/K+ATPase activity of rainbow trout (*Oncorhynchus mykiss*) [55]. In this study, after dietary taurine intervention in the High–fat group, *ATP1α* gene expression was upregulated and enriched into the bile secretion pathway. Therefore, dietary taurine administration accelerates bile acid transport and metabolism through upregulating *ATP1α* gene expression.

Lipid metabolic disorder is associated with abnormal energy metabolism, including gluconeogenesis, glycogenolysis, and the TCA cycle. GK is a rate–limiting enzyme required for glucose metabolism in the liver and the regulation of insulin secretion from islets [56]. Hepatic GK enzyme activity of glycolysis pathway is stimulated by glucose uptake [57], contributing an increase in liver glycogen content to maintain glucose homeostasis. In contrast, the downregulation of hepatic *GK* gene and protein expression was accompanied by reduced hepatic glycogen synthesis in type 2 diabetic rats [58]. The findings of the

present study showed that dietary taurine supplementation could inhibit the glycolysis pathway as well as the biosynthesis of pyruvate, which is the final metabolites of glycolysis and the substrate for lipid synthesis de novo. Therefore, the lipid–reducing effect of taurine on fish can also be indirectly achieved by inhibiting the biosynthesis of substrate of lipogenesis de novo via down–regulating the gene expression in the glycolysis pathway.

The activation of CAMK II was involved in the hypertonicity–induced upregulation of human taurine transporter [59], enhancing insulin secretion [60]. In this study, the expression of gene *CAMK* (Ca2+/Calmodulin kinase II) was up-regulated after dietary taurine intervention, and enriched into the insulin secretion signal pathway. Taurine can promote insulin secretion through the interaction between taurine and ATP sensitive K<sup>+</sup> (KATP) channels [61,62].

Insulin is an important endocrine hormone. It promotes free fatty acid and cholesterol uptake, reduces lipolysis, and increases lipogenesis in fish [63] through regulating several enzymes involved in lipogenesis and lipolysis, as well as transcription factors regulating the expression of such enzymes [64,65]. High–fat diets caused lipid metabolism disorder, insulin resistance, and liver steatosis in mammals [66], and liver steatosis was positively associated with insulin resistance in nonalcoholic fatty liver disease [67]. Dietary taurine administration can increase the insulin sensitivity of mammals [68,69]. Therefore, dietary taurine may promote insulin secretion through up regulating the expression of *CAMK* gene. On the one hand, it may inhibit insulin resistance caused by a high–fat diet. On the other hand, it is conducive to the decomposition of peripheral free fatty acids and cholesterol into the liver, thus reducing peripheral fat deposition. However, the relevant results here still need further study in this regard.

Liver lipid homeostasis regulation involves a complex interaction of triglycerides present in hepatocytes including fatty acid uptake, de novo lipogenesis, fatty acid β-oxidation, and fatty acid export [66,70,71]. Abnormal liver fat accumulation is often accompanied by excessive production of reactive oxygen free radicals and intermediates with lipotoxicity such as diacylglycerol as well as endoplasmic reticulum stress, resulting in lipid metabolism disorder [71,72]. Mitochondria are recognized as the main organelles of fatty acid β-oxidation. The fatty acid β-oxidation is regulated by gene *CACT* [73]. The transcriptional level of *CACT* gene was up–regulated by fish oil both in rats and fish compared with beef tallow or lard oil [74,75]. In this study, dietary taurine intervention resulted in an upregulation of the *CACT* gene expression. The *CACT* gene was enriched into the thermogenesis pathway. This indicates that up–regulated *CACT* gene expression caused by dietary taurine intervention promotes fatty acid transport to mitochondria and fatty acid β-oxidation, thereby reducing lipid deposition in the liver of groupers.

#### **4. Materials and Methods**

#### *4.1. Experimental Diets*

The optimal dietary levels of lipid and taurine were at about 10% and 1%, respectively for the growth of *E. coioides* [76,77]. In this experiment, therefore, three isonitrogenous (46% crude protein) experimental diets were prepared using casein and gelatin without taurine (food grade) and shrimp meal as protein source, fish oil, soy oil and soy lecithin as lipid source, namely 10% fat det (control diet), 15% fat diet (high fat diet) and 15% fat + 1% taurine diet (Taurine diet). The ingredients and proximate composition of the three experimental diets are shown in Table 4. The experimental diets were produced according to the method described in detail previously [24].

#### *4.2. Growth Trial*

Juvenile groupers purchased from a local fish farm (Zhangpu county, Fujian province, China) were transported to the Aquaculture Experimental Center of Jimei University. They were fed a commercial feed for two weeks of acclimatization in two tanks (1000 L/tank) before the start of the experiment. A total of 270 fish with a similar size (an initial wet body weight of 10.5 ± 0.1 g) were randomly distributed into 9 fiberglass tanks at a stocking

density of 30 fish per tank (300 L/tank), within a recirculating water aquaculture system connected with a circulation pump, biological filters, and an automatic temperature control device. The fish in triplicate tanks of each group were fed one of the three experimental diets to apparent satiation twice daily (8:30 and 18:30), across a feeding period of 8 weeks. Excess feed was collected 30 min after each meal, dried in a ventilated oven at 65 ◦C, and weighed for the calculation of feed intake. During the feeding period, water temperature was maintained at about 28.0 ◦C, dissolved oxygen level ranged between 6.1 ± 0.2 mg/L, and the ammonia nitrogen level was less than 0.2 mg/L.


**Table 4.** Ingredients and proximate composition of experimental diets (on an as–fed basis, %) <sup>1</sup> .

<sup>1</sup> All feed ingredients are provided by Jiakang Feed Co., Ltd. (Xiamen, China).

#### *4.3. Sample Collection*

At the end of the 8-week feeding trial, the fish in each tank were caught and anesthetized with a dose of 100 mg/L solution of MS-222 (tricaine methane sulphonate, Sigma-Aldrich Shanghai Trading Co. Ltd., Shanghai, China), followed by fish count and batch–weighing, and recorded on a wet weight basis to determine percent weight gain and feed conversion ratio. Five fish were randomly caught from each tank and individually weighed. The liver was then removed after abdominal dissection, and weighed for the determination of hepatosomatic index. The liver samples were pooled by tank, frozen immediately in liquid nitrogen and stored at −20 ◦C for the analysis of fat content. For gene expression analysis, another three fish in each tank were randomly sampled and anesthetized. After drawing blood, liver samples were aseptically removed and quickly frozen with liquid nitrogen and stored at −80 ◦C.

#### *4.4. Composition Analysis*

Proximate composition of ingredients, diets, and tissue samples were determined according to standard methods [78]. Dry matter was determined by drying the samples in an oven at 105 ◦C to a constant weight. Crude protein was determined by the Kjeldahl method (N × 6.25) using Kjeltec TM 8400 Auto Sample Systems (Foss Teacher AB). The crude fat content was determined by the Soxtec extraction method by using Soxtec Avanti 2050 (Foss Teacher AB). Ash was measured in the residues of samples burned in a muffle furnace at 550 ◦C for 8 h.

Total fat of muscle and liver samples were extracted by homogenization in chloroform/methanol (2:1, *v*/*v*) solution and determined gravimetrically after drying a 5 mL aliquot under nitrogen [79]. For taurine determination [80], feed samples were hydrolysed in nitrogen–flushed glass vials using 6 mol/L HCl at 116 ◦C for 22 h, followed by centrifugation (1500 g, 4 ◦C, 15 min), and the supernatant was collected and analyzed using an automatic AA analyzer (Hitachi L8900, Tokyo, Japan).

#### *4.5. RNA Extraction and cDNA Library Construction*

Total RNA was extracted from the liver sample using TRIzol® reagent (Magen) according to the manufacturer's instructions. The A260/A280 absorbance ratios of RNA in the liver samples were detected to test the RNA purity by using NanoDrop ND–2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA). The RIN value of sample RNA was detected to test the RNA quality of the samples by using Agilent Bioanalyzer 4150 (Agilent Technologies, Santa Clara, CA, USA). The cDNA library was constructed using the Illumina HiseqTM 2000 system (Illumina, San Diego, CA, USA) by APTBIO Co., Ltd. (Shanghai, China). Raw reads were filtered to remove low-quality sequences using the program written by APTBIO Co., Ltd.

#### *4.6. Sequence Data Processing and Analysis*

RNA–Seq de novo assembly was carried out by using Trinity software (http:// trinityrnaseq.github.io/) (accessed on 27 July 2020). The longest transcripts were regarded as unigenes after removing repetitive assemblies. BLAST software (http://blast. ncbi.nlm.nih.gov/Blast.cgi/) (accessed on 27 July 2020) was used to align unigene sequences with the Non-Redundant Protein Sequence Database (NR, https://ftp.ncbi.nlm. nih.gov/blast/db/FASTA/) (accessed on 27 July 2020), Protein Families Database (Pfam, http://pfam.xfam.org/) (accessed on 27 July 2020), SwissProt protein Database (SwissProt, https://www.expasy.org/) (accessed on 27 July 2020), Kyoto Encyclopedia of Gene and Genomes Database (KEGG, http://www.kegg.jp/) (accessed on 27 July 2020), and Gene Ontology Database (GO, http://geneontology.org) (accessed on 27 July 2020). RESM software (http://deweylab.github.io/RSEM/) (accessed on 27 July 2020) was used to accurately map the sequencing reads to reference genomes. The expression level of each gene was calculated from the fragment per kilobase of exon model per million mapped read (FPKM) values [81].

#### *4.7. Identification and Enrichment Analysis of Differentially Expressed Genes*

DESeq2 software (https://bioconductor.org/packages/release/bioc/html/DESeq2 .html) (accessed on 27 July 2020) was used to determine differentially expressed genes (DEGs), and unigenes with *p*-value < 0.05 and |log2 foldchange| > 1 were defined as DEGs. In the process of DEGs analysis, the recognized and effective Benjamin Hochberg method was used to correct the significance *p*-value obtained from the original hypothesis test. The corrected *p*-value, FDR (false discovery rate), is then used as the key index of DEG screening to reduce the false positive caused by independent statistical hypothesis test on the expression value of a large number of genes. The number of up- and downregulated DEGs in the liver under different dietary treatments was obtained. Using Blast2go software (https://www.blast2go.com/) (accessed on 27 July 2020), the gene ontology (GO) annotation information of all DEGs was obtained, the GO function of DEGs was classified, and the molecular function, cell composition and biological process of target genes were described [82]. The pathway enrichment analysis was performed using online service tool KAAS (KEGG automatic annotation server). Fisher's precision probability test was used to calculate the significance of enrichment of each gene in the pathway, so as to determine the corresponding significant signal transduction and metabolic pathways. The enrichment results of DEGs are displayed by KEGG enrichment scatter diagram.

#### *4.8. Quantitative Real–Time PCR Analysis*

The relative expression levels of eight selected DEGs were verified by quantitative reverse transcription PCR (qRT–PCR), so as to validate the gene expression data obtained by RNA–Seq. The eight pairs of primers were designed with Primer v. 5.0, with the β-actin gene of the fish used as the internal reference for the qPCR analysis (Table S3) and sent to

jinweizhi Biotechnology Co., Ltd. (Suzhou, China) for synthesis. The qRT–PCR reactions were performed using ChamQ Universal SYBR qPCR Master Mix (Vazyme, Nanjing, China) on a Real–Time PCR Detection System (ABI 7500, Applied Biosystems, Waltham, MA, USA). The PCR cycling conditions were as follows: 95 ◦C for 30 s, followed by 40 cycles of 95 ◦C for 10 s and 60 ◦C for 30 s, and then cycles at 95 ◦C for 15 s, 60 ◦C for 60 s and 95 ◦C for 15 s. To check reproducibility, the qRT–PCR reaction for each sample was performed in four biological replicates. The relative expression of genes was calculated using the 2−∆∆Ct method [83].

## *4.9. Statistical Analysis*

Data are presented as mean and standard errors of the mean. Data were subjected to *t*-test and one-way ANOVA and Student–Neuman–Keuls multiple comparison tests in SPSS Statistics 22.0 (SPSS, Michigan Avenue, Chicago, IL, USA). *p* < 0.05 was considered statistically significant.

#### **5. Conclusions**

In summary, the results of the present study show that feeding 15% fat diets did not result in alterations in growth and feed utilization, but increased liver fat accumulation of groupers vs those subject to 10% fat diets. However, 1% taurine addition in a 15% fat diet not only improved its growth performance, but also reduced liver fat deposition. Liver transcriptome analysis showed that 49 DEGs were identified in the comparison of High–fat and Taurine groups, of which the expression of *CDO1*, *ATP1α*, *CAMK*, and *CACT* genes was significantly up-regulated and *ARF1\_2* gene expression was significantly down–regulated. The key genes were involved in the taurine and hypotaurine metabolic pathway, bile secretion, insulin secretion, thermogenic pathway, and phospholipase D signaling pathway. As a result, the effect of dietary taurine on reducing liver fat accumulation of the fish species may be achieved by enhancing the synthesis of endogenous taurine in the liver, accelerating bile acid transport and promoting insulin secretion and fatty acid β-oxidation efficiency. The next step is to investigate the roles and functions of the key genes mentioned above in the fat metabolism of fish in response to dietary taurine.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/metabo12070670/s1. Figure S1: The sequence length of transcript and unigenes; Table S1: Alignment statistics of the RNA–Seq analysis of the nine liver samples of *E. coioides* between Control and High–fat groups, and between High–fat and Taurine groups; Table S2: The DEGs with significantly changed KEGG pathways related to lipid metabolism in the liver of *E. coioides* in the comparison of High–fat and Taurine groups; Table S3: Primers sequences of lipid metabolism related genes used for real–time PCR for *E. coioides*.

**Author Contributions:** Conceptualization, J.Y. and F.B.; methodology, M.C. and F.B.; software, M.C.; validation, F.B., T.S. and X.N.; formal analysis, X.W. and X.N.; investigation, M.C. and F.B.; data curation, M.C.; writing—original draft preparation, M.C.; writing—review and editing, J.Y; visualization, F.B.; supervision, J.Y.; project administration, K.W.; funding acquisition, J.Y. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Natural Science Foundation of China, grant numbers 31772861 and 31372546, and the Science and Technology Project of Fujian Province, China, grant number 2020N0012.

**Institutional Review Board Statement:** The animal study protocol was approved by the Ethics Committee of Jimei University, Xiamen, China (protocol code 2011–58 and approved by 20 December 2011).

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data that support the findings of this study are available from the corresponding author upon reasonable request; further inquiries can be directed to NCBI Database at https://www.ncbi.nlm.nih.gov/bioproject/PRJNA853943 (accessed on 29 June 2022).

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


## *Article* **Signaling Pathway of Taurine-Induced Upregulation of TXNIP**

**Hideo Satsu 1,\*, Yusuke Gondo <sup>2</sup> , Hana Shimanaka <sup>2</sup> , Masato Imae <sup>3</sup> , Shigeru Murakami <sup>4</sup> , Kenji Watari <sup>1</sup> , Shunichi Wakabayashi <sup>5</sup> , Sung-Joon Park <sup>5</sup> , Kenta Nakai <sup>5</sup> and Makoto Shimizu <sup>6</sup>**


**Abstract:** Taurine, a sulfur-containing β-amino acid, is present at high concentrations in mammalian tissues and plays an important role in several essential biological processes. However, the genetic mechanisms involved in these physiological processes associated with taurine remain unclear. In this study, we investigated the regulatory mechanism underlying the taurine-induced transcriptional enhancement of the thioredoxin-interacting protein (TXNIP). The results showed that taurine significantly increased the luciferase activity of the human TXNIP promoter. Further, deletion analysis of the TXNIP promoter showed that taurine induced luciferase activity only in the TXNIP promoter region (+200 to +218). Furthermore, by employing a bioinformatic analysis using the TRANSFAC database, we focused on Tst-1 and Ets-1 as candidates involved in taurine-induced transcription and found that the mutation in the Ets-1 sequence did not enhance transcriptional activity by taurine. Additionally, chromatin immunoprecipitation assays indicated that the binding of Ets-1 to the TXNIP promoter region was enhanced by taurine. Taurine also increased the levels of phosphorylated Ets-1, indicating activation of Ets-1 pathway by taurine. Moreover, an ERK cascade inhibitor significantly suppressed the taurine-induced increase in TXNIP mRNA levels and transcriptional enhancement of TXNIP. These results suggest that taurine enhances TXNIP expression by activating transcription factor Ets-1 via the ERK cascade.

**Keywords:** taurine; TXNIP; Ets-1; ERK

## **1. Introduction**

Taurine (2-aminoethanesulfonic acid) is a free β-amino acid abundant in several tissues, such as muscles, the heart, brain, and eyes. Previous research indicates that taurine performs various functions, including as an antioxidant, osmoregulator, and bile acid conjugate, as well as in detoxification [1,2]. Further, taurine is reported to be essential for the development of fetuses and newborns [3,4].

Taurine is obtained directly from diet and is also synthesized endogenously from cysteine, which in turn is formed from methionine. Dietary taurine is absorbed by intestinal epithelial cells via the taurine transporter (TAUT, SLC6A6). We previously reported that TAUT is regulated by various extracellular conditions, such as adaptive regulation and hyperosmolarity, as well as by tumor necrosis factor-α (TNF-α) [5–7]. Further, considering that the intestinal TAUT was regulated by inflammatory cytokines [7], we previously

**Citation:** Satsu, H.; Gondo, Y.; Shimanaka, H.; Imae, M.; Murakami, S.; Watari, K.; Wakabayashi, S.; Park, S.-J.; Nakai, K.; Shimizu, M. Signaling Pathway of Taurine-Induced Upregulation of TXNIP. *Metabolites* **2022**, *12*, 636. https://doi.org/ 10.3390/metabo12070636

Academic Editor: Cholsoon Jang

Received: 21 May 2022 Accepted: 8 July 2022 Published: 11 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

investigated the relationship between taurine and inflammation and reported that taurine reduced colitis symptoms in a mouse model of DSS-induced colitis [8]. Furthermore, using DNA microarrays, we comprehensively analyzed the effects of taurine on overall gene expression in intestinal epithelial cells and found that taurine markedly enhanced the mRNA expression and transcriptional activity of the thioredoxin-interacting protein (TXNIP) [9].

TXNIP suppresses its activity by interacting with thioredoxin [10,11], and various physiological functions have been reported for TXNIP [11]. It is reported that TXNIP knockout mice have significantly reduced adaptability to energy deprivation [12–16] because they develop metabolic abnormalities, such as reduced efficiency of fatty acid metabolism [15], development of hyperlipidemia [13,17], and hypoglycemic states [14,16]. Furthermore, TXNIP contributes to glycogenesis in the liver [18], and the relationship between TXNIP mutations and the development of diabetes and hypertension has been suggested [17,19]. TXNIP has also been shown to regulate cardiac hypertrophy [20]. Furthermore, immune enhancement by moderately strong expression of TXNIP is also considered [11]. Physiological functions in the intestinal tract have been reported by Takahashi et al. [21] that TXNIP mRNA expression in the site of inflammation in patients with ulcerative colitis, is decreased, which is thought to contribute to the development of ulcerative colitis. Additionally, a previous study has shown that taurine regulates the functions of human intestinal Caco-2 cells via TXNIP induction [22]. Thus, taurine regulates the mRNA expression of TXNIP; however, the mechanisms underlying its regulation remain unclear.

In the present study, we analyzed the transcription factors and signaling molecules involved in the taurine-induced increase in the transcriptional activity of TXNIP.

#### **2. Results**

#### *2.1. Effect of Taurine on Luciferase Activity Involving the TXNIP Promoter Region*

Using reporter analysis, we previously reported that taurine increased the promoter activity of TXNIP [9]. A reporter vector containing the promoter region of TXNIP (−1299/+256) was used. Therefore, we examined the effect of taurine on promoter activity in various regions of the TXNIP promoter, including (−1299/+256), (−109/+256), and (−39/+256), with truncation of the 50 -flanking region. As shown in Figure 1A, taurine increased the promoter activity of all three reporter vectors, suggesting that the promoter region (−39/+256) is essential for the taurine-induced increase in TXNIP promoter activity.

We also constructed a reporter vector containing the (−39/+256), (−39/+142), and (−39/+65) sequences of the promoter region of TXNIP. Taurine markedly increased reporter activity in the presence of a reporter vector containing the promoter region between (−39/+256), but did not increase activity in the case of (−39/+142) and (−39/+65) (Figure 1B). This result suggests that a taurine response element is in the TXNIP promoter (+143/+256).

Based on the findings that the taurine response element is contained in the TXNIP promoter region between +143 and +256 (Figure 1), we next constructed a reporter vector containing the promoter region of (+122/+256) and examined the effect of taurine on the promoter activity of TXNIP (+122/+256). A reporter vector containing the promoter region (−1299/+142) was used as a negative control. The promoter activity of TXNIP (+122/+256) was significantly increased by taurine, whereas that of TXNIP (−1299/+142) did not change (Figure 2). This result confirmed that the taurine response element exists in the TXNIP promoter (+122/+256).

*Metabolites* **2022**, *12*, x FOR PEER REVIEW 3 of 15

**Figure 1.** Effect of taurine on the transcriptional activity of the TXNIP promoter between −1299/+256. Caco-2 cells were transfected with a reporter vector containing partial promoter regions of TXNIP and then replaced with a medium containing 100 mM of taurine. After 24 h, the cells were subjected to a luciferase assay, as described in Materials and Methods Section 4.4. (**A**) TXNIP promoter partial sequences (−1299/+256, −109/+256, −39/+256). Results are expressed as relative values with the control value of (−1299/+256) as 1. (**B**) TXNIP promoter partial sequences (−29/+256, −29/+142, −39/+65). Results are expressed as relative values with the control value of −39/+256 as 1. Each value represents the mean ± S.E. (*n* = 3), \* *p* < 0.05, vs. the control value (Student's *t*-test). **Figure 1.** Effect of taurine on the transcriptional activity of the TXNIP promoter between −1299/+256. Caco-2 cells were transfected with a reporter vector containing partial promoter regions of TXNIP and then replaced with a medium containing 100 mM of taurine. After 24 h, the cells were subjected to a luciferase assay, as described in Materials and Methods Section 4.4. (**A**) TXNIP promoter partial sequences (−1299/+256, −109/+256, −39/+256). Results are expressed as relative values with the control value of (−1299/+256) as 1. (**B**) TXNIP promoter partial sequences (−29/+256, −29/+142, −39/+65). Results are expressed as relative values with the control value of −39/+256 as 1. Each value represents the mean ± S.E. (*n* = 3), \* *p* < 0.05, vs. the control value (Student's *t*-test). Based on the findings that the taurine response element is contained in the TXNIP promoter region between +143 and +256 (Figure 1), we next constructed a reporter vector containing the promoter region of (+122/+256) and examined the effect of taurine on the promoter activity of TXNIP (+122/+256). A reporter vector containing the promoter region (−1299/+142) was used as a negative control. The promoter activity of TXNIP (+122/+256) was significantly increased by taurine, whereas that of TXNIP (−1299/+142) did not change (Figure 2). This result confirmed that the taurine response element exists in the TXNIP promoter (+122/+256).

**Figure 2.** Effect of taurine on the transcriptional activity of the TXNIP promoter containing −1299/+142 and +122/+256. Caco-2 cells were transfected with a reporter vector containing partial promoter region of TXNIP (−1299/+142, +122/+256) and then replaced with a medium containing 100 mM of taurine. After 24 h, the cells were subjected to luciferase assay. The results are expressed as relative values, with a control value of +122/+256 as 1. Each value represents the mean ± S.E. (*n* = 3), \* *p* < 0.05, vs. the control value (Student's *t*-test). **Figure 2.** Effect of taurine on the transcriptional activity of the TXNIP promoter containing −1299/+142 and +122/+256. Caco-2 cells were transfected with a reporter vector containing partial promoter region of TXNIP (−1299/+142, +122/+256) and then replaced with a medium containing 100 mM of taurine. After 24 h, the cells were subjected to luciferase assay. The results are expressed as relative values, with a control value of +122/+256 as 1. Each value represents the mean ± S.E. (*n* = 3), \* *p* < 0.05, vs. the control value (Student's *t*-test).

**Figure 2.** Effect of taurine on the transcriptional activity of the TXNIP promoter containing −1299/+142 and +122/+256. Caco-2 cells were transfected with a reporter vector containing partial promoter region of TXNIP (−1299/+142, +122/+256) and then replaced with a medium containing 100 Then, the TXNIP promoter region (+122/+256) was divided into three regions (+122/+178, +162/+218, and +211/+256) and luciferase vectors each containing one region were constructed. Taurine significantly increased the promoter activity of TXNIP (+162/+218) (Figure 3), suggesting that the taurine response element is contained in the TXNIP promoter region (+162/+218).

mM of taurine. After 24 h, the cells were subjected to luciferase assay. The results are expressed as relative values, with a control value of +122/+256 as 1. Each value represents the mean ± S.E. (*n* = 3),

\* *p* < 0.05, vs. the control value (Student's *t*-test).

TXNIP promoter region (+162/+218).

TXNIP promoter region (+162/+218).

*Metabolites* **2022**,*12*, x FOR PEER REVIEW4of 15

Then, the TXNIP promoter region (+122/+256) was divided into three regions (+122/+178, +162/+218, and +211/+256) and luciferase vectors each containing one region were constructed. Taurine significantly increased the promoter activity of TXNIP (+162/+218) (Figure 3), suggesting that the taurine response element is contained in the

Then, the TXNIP promoter region (+122/+256) was divided into three regions (+122/+178, +162/+218, and +211/+256) and luciferase vectors each containing one region were constructed. Taurine significantly increased the promoter activity of TXNIP (+162/+218) (Figure suggesting that the taurine element is the

**Figure 3.** Effect of taurine on the transcriptional activity of the TXNIP promoter between +122/+256. Caco-2 cells were transfected with a reporter vector containing partial promoter regions of TXNIP (+122/+178, +162/+218, +211/+256) and then replaced with a medium containing 100 mM of taurine. After 24 h, the cells were subjected to luciferase assay. The results are expressed as relative values, with a control value of +122/+178 as 1. Each value represents the mean ± S.E. (*n* = 3), \* *p* < 0.05, vs. the control value (Student's *t*-test). **Figure 3.** Effect of taurine on the transcriptional activity of the TXNIP promoter between +122/+256. Caco-2 cells were transfected with a reporter vector containing partial promoter regions of TXNIP (+122/+178, +162/+218, +211/+256) and then replaced with a medium containing 100 mM of taurine. After 24 h, the cells were subjected to luciferase assay. The results are expressed as relative values, with a control value of +122/+178 as 1. Each value represents the mean ± S.E. (*n* = 3), \* *p* < 0.05, vs. the control value (Student's *t*-test). **Figure 3.** Effect of taurine on the transcriptional activity of the TXNIP promoter between +122/+256. Caco-2 cells transfected with a reporter containing regions (+122/+178, +162/+218, +211/+256) and then replaced with a medium containing 100 mM of taurine. After 24 h, the cells were subjected to luciferase assay. The results are expressed as relative values, with a control value of +122/+178 as 1. Each value represents the mean ± S.E. (*n* =3), \* *p* <0.05, vs. the control value (Student's *t*-test).

Further, the TXNIP promoter region (+162/+218) was divided into three regions (+174/+191, +187/+204, and +200/+218), and a luciferase vector containing each region was constructed. Taurine significantly increased the promoter activity of TXNIP (+200/+218) (Figure 4), suggesting that the taurine response element is contained in the TXNIP promoter region (+200/+218). Taurine also increased the promoter activity of TXNIP (+200/+218) in a dose-dependent manner (Figure 5), confirming that a taurine response element exists in the promoter region (+200/+218). Further, the TXNIP promoter region (+162/+218) was divided into three regions (+174/+191, +187/+204, and +200/+218), and a luciferase vector containing each region was constructed. Taurine significantly increased the promoter activity of TXNIP (+200/+218) (Figure 4), suggesting that the taurine response element is contained in the TXNIP promoter region (+200/+218). Taurine also increased the promoter activity of TXNIP (+200/+218) in a dose-dependent manner (Figure 5), confirming that a taurine response element exists in the promoter region (+200/+218). Further, the promoter was divided (+174/+191, +187/+204, and +200/+218), and a luciferase vector containing each region was constructed. Taurine significantly increased the promoter activity of TXNIP (+200/+218) (Figure 4), suggesting that the taurine response element is contained in the TXNIP promoter region (+200/+218). Taurine also increased the promoter activity of TXNIP (+200/+218) in a dose-dependent manner (Figure 5), confirming that a taurine response element exists in the promoter region (+200/+218).

**Figure 4.** Effect of taurine on the transcriptional activity of the TXNIP promoter between +174/+218. Caco-2 cells were transfected with a reporter vector containing partial promoter regions of TXNIP (+174/+191, +187/+204, +200/+218) and then replaced with a medium containing 100 mM of taurine. After 24 h, the cells were subjected to a luciferase assay. The results are expressed as relative values, with the control value as 1. Each value is the mean ± S.E. (*n* = 3) and the abc values indicated by different characters are significantly different from each other, and the abc values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05). **Figure 4.** Effect of taurine on the transcriptional activity of the TXNIP promoter between +174/+218. Caco-2 cells transfected with a reporter containing regions (+174/+191, +187/+204, +200/+218) and then replaced with a medium containing 100 mM of taurine. h, the cells subjected are as with the control value as 1. Each value is the mean ± S.E. (*<sup>n</sup>* <sup>=</sup> 3) and the abc values indicated by different characters are significantly different from each other, and the abcvalues indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05). **Figure 4.** Effect of taurine on the transcriptional activity of the TXNIP promoter between +174/+218. Caco-2 cells were transfected with a reporter vector containing partial promoter regions of TXNIP (+174/+191, +187/+204, +200/+218) and then replaced with a medium containing 100 mM of taurine. After 24 h, the cells were subjected to a luciferase assay. The results are expressed as relative values, with the control value as 1. Each value is the mean <sup>±</sup> S.E. (*<sup>n</sup>* = 3) and the abc values indicated by different characters are significantly different from each other, and the abc values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05).

*Metabolites* **2022**, *12*, x FOR PEER REVIEW 5 of 15

**Figure 5.** Dose-dependence of taurine-induced TXNIP promoter activity (+200/+218). Caco-2 cells were transfected with a reporter vector containing the partial promoter region of TXNIP (+200/+218) and then replaced with a medium containing 0–200 mM of taurine. After 24 h, the cells were subjected to luciferase assay. The results are expressed as relative values, with a control value of 1. Each value represents the mean ± S.E. (*n* = 3), \* *p* < 0.05, \*\* *p* < 0.01, vs. the control value (Dunnett's test). The promoter region of human TXNIP (+162/+218) was analyzed using a bioinfor-**Figure 5.** Dose-dependence of taurine-induced TXNIP promoter activity (+200/+218). Caco-2 cells were transfected with a reporter vector containing the partial promoter region of TXNIP (+200/+218) and then replaced with a medium containing 0–200 mM of taurine. After 24 h, the cells were subjected to luciferase assay. The results are expressed as relative values, with a control value of 1. Each value represents the mean ± S.E. (*n* = 3), \* *p* < 0.05, \*\* *p* < 0.01, vs. the control value (Dunnett's test). **Figure 5.** Dose-dependence of taurine-induced TXNIP promoter activity (+200/+218). Caco-2 cells were transfected with a reporter vector containing the partial promoter region of TXNIP (+200/+218) and then replaced with a medium containing 0–200 mM of taurine. After 24 h, the cells were sub-

matics approach involving the TRANSFAC database. At the same time, deletion analysis

of human TXNIP promoter was also performed. Then, the result of Figure 4 was obtained and the taurine-response region was narrowed down to +200/+218. Therefore, we especially analyzed the TXNIP promoter region (+200/+218) using TRANSFAC database and focused on Tst-1 and Ets-1 as candidates involved in taurine-induced transcription. We then constructed the human TXNIP (+211/+256) promoter vector with site-directed mutagenesis of putative Tst-1 and Ets-1 response elements, respectively (Figure 6A). The results showed that the mutation of the Tst-1 response element had no effect on the increase in TXNIP promoter activity induced by taurine (Figure 6B), but the mutation of the Ets-1 response element abolished the induction of TXNIP promoter activity by taurine (Figure 6C). This result strongly suggests that Ets-1 is involved in taurine-induced upregulation of TXNIP promoter activity. The promoter region of human TXNIP (+162/+218) was analyzed using a bioinformatics approach involving the TRANSFAC database. At the same time, deletion analysis of human TXNIP promoter was also performed. Then, the result of Figure 4 was obtained and the taurine-response region was narrowed down to +200/+218. Therefore, we especially analyzed the TXNIP promoter region (+200/+218) using TRANSFAC database and focused on Tst-1 and Ets-1 as candidates involved in taurine-induced transcription. We then constructed the human TXNIP (+211/+256) promoter vector with site-directed mutagenesis of putative Tst-1 and Ets-1 response elements, respectively (Figure 6A). The results showed that the mutation of the Tst-1 response element had no effect on the increase in TXNIP promoter activity induced by taurine (Figure 6B), but the mutation of the Ets-1 response element abolished the induction of TXNIP promoter activity by taurine (Figure 6C). This result strongly suggests that Ets-1 is involved in taurine-induced upregulation of TXNIP promoter activity. jected to luciferase assay. The results are expressed as relative values, with a control value of 1. Each value represents the mean ± S.E. (*n* = 3), \* *p* < 0.05, \*\* *p* < 0.01, vs. the control value (Dunnett's test). The promoter region of human TXNIP (+162/+218) was analyzed using a bioinformatics approach involving the TRANSFAC database. At the same time, deletion analysis of human TXNIP promoter was also performed. Then, the result of Figure 4 was obtained and the taurine-response region was narrowed down to +200/+218. Therefore, we especially analyzed the TXNIP promoter region (+200/+218) using TRANSFAC database and focused on Tst-1 and Ets-1 as candidates involved in taurine-induced transcription. We then constructed the human TXNIP (+211/+256) promoter vector with site-directed mutagenesis of putative Tst-1 and Ets-1 response elements, respectively (Figure 6A). The results showed that the mutation of the Tst-1 response element had no effect on the increase in TXNIP promoter activity induced by taurine (Figure 6B), but the mutation of the Ets-1 response element abolished the induction of TXNIP promoter activity by taurine (Figure 6C). This result strongly suggests that Ets-1 is involved in taurine-induced upregulation of TXNIP promoter activity.

**Figure 6.** TXNIP promoter sequence (+162/+218) and the effect of taurine on TXNIP promoter activity by employing site-directed mutagenesis. (**A**) Human TXNIP promoter sequences between +162 and +218 are shown. The sequence of putative Tst-1 and Ets-1 response sequences and their mutants are shown, respectively. (**B**,**C**) Caco-2 cells were transfected with a reporter vector containing the **Figure 6.** TXNIP promoter sequence (+162/+218) and the effect of taurine on TXNIP promoter activity by employing site-directed mutagenesis. (**A**) Human TXNIP promoter sequences between +162 and +218 are shown. The sequence of putative Tst-1 and Ets-1 response sequences and their mutants are shown, respectively. (**B**,**C**) Caco-2 cells were transfected with a reporter vector containing the partial promoter region of TXNIP (−39/+256; wild type, Tst-1 mutant, Ets-1 mutant) and then replaced with the medium containing 100 mM of taurine. After 24 h, the cells were subjected to the luciferase assay. Results are expressed as relative values with the control value of mock as 1. Each value is the mean <sup>±</sup> S.E. (*<sup>n</sup>* = 4) and the abcd values indicated by different characters are significantly different from each other and the abcd values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05).

#### *2.2. Effect of Taurine on Ets-1 Binding to TXNIP Promoter (ChIP Assay) 2.2. Effect of Taurine on Ets-1 Binding to TXNIP Promoter (ChIP Assay) 2.2. Effect of Taurine on Ets-1 Binding to TXNIP Promoter (ChIP Assay)*

*Metabolites* **2022**, *12*, x FOR PEER REVIEW 6 of 15

*Metabolites* **2022**, *12*, x FOR PEER REVIEW 6 of 15

different (Tukey's test; *p* < 0.05).

different (Tukey's test; *p* < 0.05).

To confirm the enhancement of Ets-1 binding to the human TXNIP promoter region by taurine, a ChIP assay was performed. Caco-2 cells were cultured with 100 mM of taurine for 48 h and the cells were recovered. The lysate was immunoprecipitated using an anti-Ets-1 antibody. A real-time polymerase chain reaction (PCR) analysis was performed to determine whether the precipitated DNA contained the TXNIP promoter region with the Ets-1 response element. The results showed that taurine clearly increased Ets-1 binding to the TXNIP promoter region (Figure 7). We did not perform experimental verification of Tst-1 binding but only focused our attention on Ets-1. To confirm the enhancement of Ets-1 binding to the human TXNIP promoter region by taurine, a ChIP assay was performed. Caco-2 cells were cultured with 100 mM of taurine for 48 h and the cells were recovered. The lysate was immunoprecipitated using an anti-Ets-1 antibody. A real-time polymerase chain reaction (PCR) analysis was performed to determine whether the precipitated DNA contained the TXNIP promoter region with the Ets-1 response element. The results showed that taurine clearly increased Ets-1 binding to the TXNIP promoter region (Figure 7). We did not perform experimental verification of Tst-1 binding but only focused our attention on Ets-1. To confirm the enhancement of Ets-1 binding to the human TXNIP promoter region by taurine, a ChIP assay was performed. Caco-2 cells were cultured with 100 mM of taurine for 48 h and the cells were recovered. The lysate was immunoprecipitated using an anti-Ets-1 antibody. A real-time polymerase chain reaction (PCR) analysis was performed to determine whether the precipitated DNA contained the TXNIP promoter region with the Ets-1 response element. The results showed that taurine clearly increased Ets-1 binding to the TXNIP promoter region (Figure 7). We did not perform experimental verification of Tst-1 binding but only focused our attention on Ets-1.

partial promoter region of TXNIP (−39/+256; wild type, Tst-1 mutant, Ets-1 mutant) and then replaced with the medium containing 100 mM of taurine. After 24 h, the cells were subjected to the luciferase assay. Results are expressed as relative values with the control value of mock as 1. Each value is the mean ± S.E. (*n* = 4) and the abcd values indicated by different characters are significantly different from each other and the abcd values indicated by the same characters are not significantly

partial promoter region of TXNIP (−39/+256; wild type, Tst-1 mutant, Ets-1 mutant) and then replaced with the medium containing 100 mM of taurine. After 24 h, the cells were subjected to the luciferase assay. Results are expressed as relative values with the control value of mock as 1. Each value is the mean ± S.E. (*n* = 4) and the abcd values indicated by different characters are significantly different from each other and the abcd values indicated by the same characters are not significantly

**Figure 7.** Effect of taurine on the binding of Ets-1 to the TXNIP promoter. Caco-2 cells were cultured in medium containing 100 mM of taurine. After 48 h, a ChIP assay was performed as described in Materials and Methods. The results are expressed as relative values with 100% being the DNA value before immunoprecipitation. Each value represents the mean ± S.E. (*n* = 2). **Figure 7.** Effect of taurine on the binding of Ets-1 to the TXNIP promoter. Caco-2 cells were cultured in medium containing 100 mM of taurine. After 48 h, a ChIP assay was performed as described in Materials and Methods. The results are expressed as relative values with 100% being the DNA value before immunoprecipitation. Each value represents the mean ± S.E. (*n* = 2). **Figure 7.** Effect of taurine on the binding of Ets-1 to the TXNIP promoter. Caco-2 cells were cultured in medium containing 100 mM of taurine. After 48 h, a ChIP assay was performed as described in Materials and Methods. The results are expressed as relative values with 100% being the DNA value before immunoprecipitation. Each value represents the mean ± S.E. (*n* = 2).

#### *2.3. Effect of Taurine on Ets-1 Activation (Phosphorylation) 2.3. Effect of Taurine on Ets-1 Activation (Phosphorylation) 2.3. Effect of Taurine on Ets-1 Activation (Phosphorylation)*

We also examined whether taurine activates Ets-1 or not. Ets-1 is activated by Thr38 [23]. Caco-2 cells were incubated with 100 mM of taurine for 3 h and the cell lysate was recovered and used for western blot analysis. Figure 8 shows that taurine increased the protein expression of phosphorylated Ets-1, suggesting that taurine activates Ets-1. We also examined whether taurine activates Ets-1 or not. Ets-1 is activated by Thr38 [23]. Caco-2 cells were incubated with 100 mM of taurine for 3 h and the cell lysate was recovered and used for western blot analysis. Figure 8 shows that taurine increased the protein expression of phosphorylated Ets-1, suggesting that taurine activates Ets-1. We also examined whether taurine activates Ets-1 or not. Ets-1 is activated by Thr38 [23]. Caco-2 cells were incubated with 100 mM of taurine for 3 h and the cell lysate was recovered and used for western blot analysis. Figure 8 shows that taurine increased the protein expression of phosphorylated Ets-1, suggesting that taurine activates Ets-1.

**Figure 8.** Effect of taurine on the phosphorylation (activation) of Ets-1. Caco-2 cells were cultured in a medium containing 100 mM of taurine for 3 h. The cell lysate was recovered and used for the western blot analysis. The bands were quantified using an image gauge. **Figure 8.** Effect of taurine on the phosphorylation (activation) of Ets-1. Caco-2 cells were cultured in a medium containing 100 mM of taurine for 3 h. The cell lysate was recovered and used for the western blot analysis. The bands were quantified using an image gauge. **Figure 8.** Effect of taurine on the phosphorylation (activation) of Ets-1. Caco-2 cells were cultured in a medium containing 100 mM of taurine for 3 h. The cell lysate was recovered and used for the western blot analysis. The bands were quantified using an image gauge.

#### *2.4. The Involvement of MAP Kinase Family on Taurine-Induced Induction of TXNIP mRNA*

To further elucidate the signaling pathway involved in taurine-induced TXNIP induction, we examined the effect of MAPK inhibitors on taurine TXNIP mRNA expression. The results showed that PD98059, an ERK1/2 pathway inhibitor, significantly suppressed the taurine-induced increase in TXNIP mRNA, whereas the p38 (SB203580) and JNK (SP600125) inhibitors did not (Figure 9).

*Metabolites* **2022**, *12*, x FOR PEER REVIEW 7 of 15

(SP600125) inhibitors did not (Figure 9).

(SP600125) inhibitors did not (Figure 9).

**Figure 9.** Effect of MAP kinase inhibitors on the taurine-induced increase in TXNIP mRNA in Caco-2 cells. Caco-2 cells were pretreated with each of the three MAPK inhibitors for 2 h and then cultured in medium containing 100 mM of taurine and each inhibitor. After 48 h, RNA was extracted and used for real-time PCR. The results are expressed as relative values, with the control value without the inhibitor as 1. Each value is the mean ± S.E. (*n* = 3) and the abcd values indicated by different characters are significantly different from each other, and the abcd values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05). **Figure 9.** Effect of MAP kinase inhibitors on the taurine-induced increase in TXNIP mRNA in Caco-2 cells. Caco-2 cells were pretreated with each of the three MAPK inhibitors for 2 h and then cultured in medium containing 100 mM of taurine and each inhibitor. After 48 h, RNA was extracted and used for real-time PCR. The results are expressed as relative values, with the control value without the inhibitor as 1. Each value is the mean <sup>±</sup> S.E. (*<sup>n</sup>* = 3) and the abcd values indicated by different characters are significantly different from each other, and the abcd values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05). **Figure 9.** Effect of MAP kinase inhibitors on the taurine-induced increase in TXNIP mRNA in Caco-2 cells. Caco-2 cells were pretreated with each of the three MAPK inhibitors for 2 h and then cultured in medium containing 100 mM of taurine and each inhibitor. After 48 h, RNA was extracted and used for real-time PCR. The results are expressed as relative values, with the control value without the inhibitor as 1. Each value is the mean ± S.E. (*n* = 3) and the abcd values indicated by different characters are significantly different from each other, and the abcd values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05).

*2.4. The Involvement of MAP Kinase Family on Taurine-Induced Induction of TXNIP mRNA*

*2.4. The Involvement of MAP Kinase Family on Taurine-Induced Induction of TXNIP mRNA*

To further elucidate the signaling pathway involved in taurine-induced TXNIP induction, we examined the effect of MAPK inhibitors on taurine TXNIP mRNA expression. The results showed that PD98059, an ERK1/2 pathway inhibitor, significantly suppressed the taurine-induced increase in TXNIP mRNA, whereas the p38 (SB203580) and JNK

To further elucidate the signaling pathway involved in taurine-induced TXNIP induction, we examined the effect of MAPK inhibitors on taurine TXNIP mRNA expression. The results showed that PD98059, an ERK1/2 pathway inhibitor, significantly suppressed the taurine-induced increase in TXNIP mRNA, whereas the p38 (SB203580) and JNK

#### *2.5. Effect of Taurine on ERK1/2 Activation in Caco-2 Cells 2.5. Effect of Taurine on ERK1/2 Activation in Caco-2 Cells 2.5. Effect of Taurine on ERK1/2 Activation in Caco-2 Cells*

Next, to assess whether ERK1/2 was activated by taurine, the western blot analysis was performed. ERK1/2 was phosphorylated after 3 or 24 h of incubation with 100 mM of taurine (Figure 10). These results indicated that taurine activated ERK1/2 in Caco-2 cells. Next, to assess whether ERK1/2 was activated by taurine, the western blot analysis was performed. ERK1/2 was phosphorylated after 3 or 24 h of incubation with 100 mM of taurine (Figure 10). These results indicated that taurine activated ERK1/2 in Caco-2 cells. Next, to assess whether ERK1/2 was activated by taurine, the western blot analysis was performed. ERK1/2 was phosphorylated after 3 or 24 h of incubation with 100 mM of taurine (Figure 10). These results indicated that taurine activated ERK1/2 in Caco-2 cells.

**Figure 10.** Effect of taurine on the phosphorylation (activation) of ERK1/2. Caco-2 cells were cultured in medium containing 100 mM of taurine for 3, 24, and 48 h. The cell lysate was recovered and used for the western blot analysis. The bands were quantified using an image gauge. Changes in expression are presented as relative values, with the expression level in the control at each treatment time as 1. **Figure 10.** Effect of taurine on the phosphorylation (activation) of ERK1/2. Caco-2 cells were cultured in medium containing 100 mM of taurine for 3, 24, and 48 h. The cell lysate was recovered and used for the western blot analysis. The bands were quantified using an image gauge. Changes in expression are presented as relative values, with the expression level in the control at each treatment time as 1. **Figure 10.** Effect of taurine on the phosphorylation (activation) of ERK1/2. Caco-2 cells were cultured in medium containing 100 mM of taurine for 3, 24, and 48 h. The cell lysate was recovered and used for the western blot analysis. The bands were quantified using an image gauge. Changes in expression are presented as relative values, with the expression level in the control at each treatment time as 1.

#### *2.6. Involvement of ERK Signaling Pathway in Taurine-Induced Enhancement of Transcriptional Activity of TXNIP*

Finally, we examined the effect of PD98059 on the taurine-induced increase in TXNIP promoter activity (+200/+218). PD98059 significantly suppressed the taurine-induced increase in TXNIP promoter activity (Figure 11). This result suggests that taurine activates the TXNIP promoter activity via the ERK cascade.

the TXNIP promoter activity via the ERK cascade.

**Figure 11.** Effect of ERK1/2 cascade inhibitor on the taurine-induced increase in TXNIP promoter. After transfection with a reporter vector containing the promoter region of TXNIP (+200/+218), Caco-2 cells were pretreated with 50 µM PD98059 for 2 h and then cultured in medium containing 100 mM of taurine and the inhibitor. After 24 h, the cells were subjected to a luciferase assay. The results are expressed as relative values, with the control value without inhibitor as 1. Each value is the mean ± S.E. (*n* = 3) and the ab values indicated by different characters are significantly different from each other and the ab values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05). **Figure 11.** Effect of ERK1/2 cascade inhibitor on the taurine-induced increase in TXNIP promoter. After transfection with a reporter vector containing the promoter region of TXNIP (+200/+218), Caco-2 cells were pretreated with 50 µM PD98059 for 2 h and then cultured in medium containing 100 mM of taurine and the inhibitor. After 24 h, the cells were subjected to a luciferase assay. The results are expressed as relative values, with the control value without inhibitor as 1. Each value is the mean <sup>±</sup> S.E. (*<sup>n</sup>* = 3) and the ab values indicated by different characters are significantly different from each other and the ab values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05).

*2.6. Involvement of ERK Signaling Pathway in Taurine-Induced Enhancement of Transcrip-*

Finally, we examined the effect of PD98059 on the taurine-induced increase in TXNIP promoter activity (+200/+218). PD98059 significantly suppressed the taurine-induced increase in TXNIP promoter activity (Figure 11). This result suggests that taurine activates

#### **3. Discussion 3. Discussion**

*tional Activity of TXNIP*

In the present study, we performed a detailed analysis of the mechanisms underlying the taurine-induced enhancement of TXNIP transcriptional activity. Our results demonstrate that taurine activates (phosphorylates) the transcription factor Ets-1 via the ERK signaling pathway and that the activated Ets-1 enhances the transcriptional activity of TXNIP by binding to the response sequence at +200**/**+218 in the TXNIP promoter region. In the present study, we performed a detailed analysis of the mechanisms underlying the taurine-induced enhancement of TXNIP transcriptional activity. Our results demonstrate that taurine activates (phosphorylates) the transcription factor Ets-1 via the ERK signaling pathway and that the activated Ets-1 enhances the transcriptional activity of TXNIP by binding to the response sequence at +200/+218 in the TXNIP promoter region.

A total of 100 mM of taurine increases the TXNIP promoter activity (+200/+218) about two-fold in Figure 4, whereas taurine increases the promoter activity (+200/+218) about 1.3 fold in Figure 5. The difference in the induction ratio of luciferase activity by taurine between Figure 4 and Figure 5 is thought to be due to the difference in the condition of Caco-2 cells. Caco-2 cells are used as an intestinal epithelial model, but this cell line is heterogeneous and it is well known that the cell characteristics and cell response are often changed during passage [24,25]. Although the induction ratio is different, the significant increase of TXNIP promoter activity (+200/+218) by taurine is observed in both Figures 4 A total of 100 mM of taurine increases the TXNIP promoter activity (+200/+218) about two-fold in Figure 4, whereas taurine increases the promoter activity (+200/+218) about 1.3 fold in Figure 5. The difference in the induction ratio of luciferase activity by taurine between Figures 4 and 5 is thought to be due to the difference in the condition of Caco-2 cells. Caco-2 cells are used as an intestinal epithelial model, but this cell line is heterogeneous and it is well known that the cell characteristics and cell response are often changed during passage [24,25]. Although the induction ratio is different, the significant increase of TXNIP promoter activity (+200/+218) by taurine is observed in both Figures 4 and 5.

and 5. The deletion analysis of the TXNIP promoter suggests that the response sequences of the transcription factors Tst-1 and Ets-1 in the 200/+218 region of the TXNIP promoter were the candidates for taurine response sequences. However, there have been no reports of promoter analyses on TXNIP in region 3′ from −39, and it is unknown what kind of transcription factor response sequences exist in this region. Therefore, we analyzed the sequences using JASPAR, a database of predicted transcription factor-binding sequences. These results revealed the presence of a putative Tst-1 transcription response sequence The deletion analysis of the TXNIP promoter suggests that the response sequences of the transcription factors Tst-1 and Ets-1 in the 200/+218 region of the TXNIP promoter were the candidates for taurine response sequences. However, there have been no reports of promoter analyses on TXNIP in region 30 from −39, and it is unknown what kind of transcription factor response sequences exist in this region. Therefore, we analyzed the sequences using JASPAR, a database of predicted transcription factor-binding sequences. These results revealed the presence of a putative Tst-1 transcription response sequence and an Ets-1 transcription factor response sequence in +210/+215 of the TXNIP promoter.

and an Ets-1 transcription factor response sequence in +210/+215 of the TXNIP promoter. Tst-1 belongs to a gene family with a Pit-Oct-Unc (POU) domain, and is reported to be expressed in the brain [26], and represses the expression of the P0 gene, a cell surface adhesion molecule [26], and specifically induces gene expression of the nicotinic acetylcholine receptor (nAchR) subunit 3 [27]. However, Tst-1 has only been examined in the brain and not much information is available about its role in other tissues. Ets-1 is a member of the E26 transformation-specific (Ets) family, which contains a helix-turn-helix DNA-binding region and has been reported to bind to the GGAA/T motif of DNA [28]. The Ets family is reported to be a family of transcription factors involved in diverse

physiological functions, such as cell proliferation, differentiation, and apoptosis [29], and Ets-1 is known to be expressed in many tissues [30]. Therefore, we constructed luciferase vectors with mutations in the response sequences of each transcription factor (Tst-1 and Ets-1) and proceeded with the analysis.

Mutation analysis suggests the involvement of Ets-1 in taurine-induced enhancement of TXNIP transcription. Subsequently, we examined whether taurine activates Ets-1 and found that taurine addition activated (phosphorylated) Ets-1 (Figure 8). Considering that Ets-1 is reported to be activated by ERK [31], we hypothesized that taurine phosphorylates the transcription factor Ets-1 via ERK and enhances TXNIP transcriptional activity. A ChIP assay was performed on Caco-2 cells cultured in taurine-containing medium as shown in Figure 7; the results showed that taurine enhanced Ets-1 binding to the TXNIP promoter region. Notably, the binding of Ets-1 to the TXNIP promoter region was observed even in controls, suggesting that Ets-1 is involved in TXNIP transcriptional activity, even in the steady-state. This result is consistent with the results of the mutation analysis shown in Figure 6. A previous study reported that Ets-1 binds to the TXNIP promoter region at a steady-state in the regulation of TXNIP expression in pancreatic cells [32]. Further, Ets-1 has been shown to bind to the Flvcr1 promoter region in Caco-2 cells [33].

These findings suggest that taurine activates transcription factor Ets-1 via ERK and binds to the TXNIP promoter region, resulting in increased TXNIP transcriptional activity. The involvement of Ets-1 in the transcriptional activity of TXNIP has previously been reported in studies that observed induction of TXNIP by synthetic retinoids in osteosarcoma cells [34] and in the regulation of TXNIP expression in pancreatic cells [32]. However, considering that these two previous studies have shown that Ets-1 binds to the response sequence in the TXNIP promoter −300/−400 region, and our results that taurine-induced enhancement of TXNIP transcriptional activity occurs in the TXNIP promoter +200/+218 region, it appears that taurine may activate Ets-1 through multiple mechanisms. In addition to TXNIP, biomolecules that Ets-1 induces include connective tissue growth factor (CTGF/CCN2), a factor that promotes bone formation [35,36], vascular endothelial growth factor (VEGF) [37], parathyroid hormone-related peptide (PTHrP) [38], and p16 [39]. In addition, Ets-1 interacts with other transcription factors such as prox1, SP-1, and AP-1 to induce their expression [37,38]. Therefore, we cannot rule out the involvement of factors other than Ets-1 in the enhancement of TXNIP transcriptional activity by taurine.

Ets-1 is expressed in all tissues during embryonic development in mice and is reported to be crucial for morphogenesis [40]. In this process, Ets-1 plays a particularly important role in angiogenesis, and it has been reported that double-mutant mice with Ets-2 die in the early stages of angiogenesis [41]. As Ets-1 is involved in the formation of many tissues, it is likely that Ets-1 aids the key function of taurine during fetal, neonatal, and infant life. It has also been reported that natural killer cells do not mature in Ets-1-deficient mice [42]. Thus, Ets-1 appears to be involved in various physiological functions and likely regulates the diverse physiological effects of taurine. Our study elucidated the mechanism of taurineinduced enhancement of TXNIP expression and indicates that this occurs via the ERK pathway. However, the mechanism by which taurine activates ERK is not clear. Upstream signals of the ERK pathway include Ras/Raf/MEK/ERK [43], Ras/PKC/MEK/ERK [44], PKC/Ras/MEK/ERK [45], and PKC/MEK/ERK without Ras and Raf [46]. Since the PKC pathway is involved in three of these four pathways, future studies should examine the effect of the PKC inhibitor Ro318220 on taurine-induced increases in TXNIP mRNA expression. As PKC has also been reported to inhibit TAUT activity [47], it is possible that the cellular uptake of taurine is affected by it, though this possibility needs to be tested. Ras is involved in the activation of Ets-1 [48] and Ets-1 is activated by the Ras/Raf/MEK pathway, particularly in the induction of p16 expression [39]. Therefore, it is likely that taurine activates Ras, leading to the activation of ERK. Consequently, to clarify the upstream factors of the ERK pathway, it is necessary to first examine in detail whether taurine activates PKC and Ras. Further, it is reported that the inhibition of the ERK1/2 pathway by PD98059 in vitro leads to compensatory upregulation of the PI3K/Akt signaling pathway [49]. Therefore, it cannot be ruled out that PI3K/Akt activation is involved in this TXNIP induction. However, we recently found that taurine increases TXNIP mRNA expression in human hepatic HepG2 cells as well as in Caco-2 cells and further revealed that LY294002, a specific PI3K inhibitor, had no significant effect on taurine-induced increase in TXNIP mRNA (Figure A1). Therefore, it is not likely that PI3K/Akt signaling pathway is involved in this regulation.

Caco-2 cells are thought to recognize intracellular or extracellular taurine via unknown taurine receptors or binding proteins, leading to ERK-Ets-1 activation. In the brain, taurine binds to GABA<sup>A</sup> [50–52] and glycine [52,53] receptors and functions as an agonist. However, its binding affinity to the receptors is weaker than that of GABA and glycine. There have been no reports of binding to these receptors in tissues other than cranial nerves. Furthermore, taurine, but neither GABA nor glycine, contributes to the regulation of TAUT expression in intestinal epithelial cells [5]. TAUT receptors are expressed in the plasma membrane [54] and are activated by β-alanine [55]. Noticeably, in the present study, mRNA expression of TXNIP was not induced by β-alanine [9]; therefore, TAUT receptors are unlikely to be involved in the taurine-induced upregulation of TXNIP transcription. Furthermore, the expression of TAUT receptors has been reported only in neural tissues, and the DNA microarray results from our laboratory showed that their expression was absent from our cell cultures [9], suggesting that they are not expressed in Caco-2 cells. These findings suggest that there may be other receptors in cells that specifically recognize taurine; however, these receptors are yet to be identified.

In the present study, we revealed that taurine induced the transcriptional activity of TXNIP via the ERK and Ets-1 signaling pathways. These findings will hopefully contribute to the elucidation of the mechanisms underlying the diverse physiological effects of taurine and the discovery of new functions of taurine.

#### **4. Materials and Methods**

#### *4.1. Materials*

The Caco-2 cell line, derived from human colon cancer tissue, was purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). Dulbecco's Modified Eagle's Medium (DMEM) was purchased from Wako Pure Chemicals (Osaka, Japan). Fetal bovine serum (FBS) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Penicillinstreptomycin (10,000 U/mL and 10 mg/mL in 0.9% sodium chloride) and non-essential amino acids (NEAA) were purchased from Gibco (Gaithersburg, MD, USA). The lipofectamine reagent was purchased from Invitrogen (Carlsbad, CA, USA), and the ExScript RT reagent kit and SYBR Premix Ex Taq for real-time PCR were obtained from Takara Bio (Otsu, Japan). All other chemicals used were of reagent grade and commercially available.

## *4.2. Cell Culture*

Caco-2 cells were cultured in a medium consisting of DMEM, 10% FBS (*v*/*v*), 1% NEAA (*v*/*v*), 100 U/mL of penicillin, and 100 µg/mL of streptomycin. The cells were incubated at 37 ◦C in a humidified atmosphere containing 5% CO2. The culture medium was replaced every other day. After reaching confluence, the cells were trypsinized with 0.1% trypsin and 0.02% EDTA in PBS and then subcultured. Caco-2 cells used in this study were between passages 35 and 79.

#### *4.3. Plasmid Construct*

The human TXNIP reporter vector containing the human TXNIP promoter region (−1299/+256) was inserted into the pGL4-basic vector (Promega, Madison, WI, USA) as previously described [10]. The human TXNIP promoter region (−109/+256, −39/+256, −39/+142, −39/+65, −1299/+142, and +122/+256, respectively) was cloned from the pGL4 reporter vector containing the TXNIP promoter (−1299/+256) by PCR. The PCR product was inserted into the pGL4-basic vector or pGL4-Promoter vector by digesting *Kpn*I and *Hind*III, and the restriction enzyme sequence was introduced into the primers.

Primers used for each reporter vector are listed in Table A1. The human TXNIP reporter vector containing the human TXNIP promoter region (+122/+178, +162/+218, +211/+256, +174/+191, +187/+204, and +200/+218) was constructed by ligating the pGL4-Promoter vector with each pair of oligonucleotides and annealing as listed in Tables A2 and A3.

#### *4.4. Transfection and Reporter Assay*

Caco-2 cells grown in a 24-well plate to 80% confluency were transiently transfected with 1 µg of the reporter vector and 0.05 µg of pRL-CMV using a lipofectamine reagent. Cells treated with or without 100 mM of taurine for 24 h were washed with PBS and lysed with Passive Lysis Buffer (Promega, Madison, WI, USA). Luciferase activity was measured using the dual-luciferase reporter assay (Promega, Madison, WI, USA) and an LB9507 Lumet luminometer (Berthold Technologies, Bad Wildbad, Germany).

#### *4.5. Real-Time PCR Analysis*

Total RNA was extracted from Caco-2 cells cultured with or without taurine using Isogen (Nippon Gene, Tokyo, Japan), according to the manufacturer's instructions. Reverse transcription of the RNA was performed using the PrimeScript RT Reagent Kit (TAKARA, Shiga, Japan), and first-strand cDNA was prepared from 0.5 µg of total RNA) and amplified using a SYBR Green Kit (TAKARA). The real-time PCR denaturation temperature was at 95 ◦C for 15 min, followed by 40 cycles of denaturation at 95 ◦C, at 60 ◦C for 15 s, and an extension at 72 ◦C for 10 s. The primer sequences were as follows: human TXNIP, 5 0 -ACGCTTCTTCTGGAAGACCA-30 (forward), and 50 -AAGCTCAAAGCCGAACTTGT-30 (reverse); β-actin, 50 -CCAGCACAATGAAGATCAAGA-30 (forward) and 50 - AGAAAGGGT GTAACGCAACTAA-30 (reverse). Real-time PCR was run on a LightCycler (Roche Applied Sciences, Penzberg, Germany).

#### *4.6. Chip Assay*

Caco-2 cells were cultured in 6-well plates and incubated with taurine for 48 h. The cells were then homogenized, and the nuclear extract was prepared using the ChIP assay. The ChIP assay was performed using ChIP-IT Express (Active Motif, Carlsbad, CA, USA), according to the manufacturer's instructions. To quantify the number of DNA fragments containing the TXNIP promoter region bound by Ets-1 protein, real-time PCR was performed. The primer sequences were as follows: human TXNIP promoter, 50 - TCGGATCTTTCTCCAGCAAT-30 (forward), and 50 -AAATCGAGGAAACCCCTTTG-30 (reverse).

#### *4.7. Western Blot Analysis*

Caco-2 cells were cultured in 6-well plates with or without taurine for several hours. The cells were collected with a cell scraper and suspended in lysis buffer (50 mM HEPES (pH 7.5), 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 1 mM NaVO3, 50 mM NaF, 20 mM β-glycerophosphate, 0.1% inhibitor cocktail (Sigma, St. Louis, MO, USA). The cell homogenate was centrifuged at 20,000× *g* for 10 min at 4 ◦C. The supernatant was used for western blot analysis as described previously [11]. The protein assay was performed using the Bio-Rad Protein Assay Solution (Bio-Rad, Hercules, CA, USA). The primary antibodies used were rabbit anti-human p-Ets-1 (Thr38) (Abcam, Cambridge, UK) and antihuman Ets-1 (Santa Cruz, Dallas, TX, USA). Rabbit anti-human p-ERK1/2 (Thr202/Tyr204) antibody (Santa Cruz, Dallas, TX, USA) and rabbit anti-human ERK1/2 antibody (Santa Cruz, Dallas, TX, USA) were used as the primary antibodies. The secondary antibody was goat anti-rabbit IgG antibody conjugated to horseradish peroxidase (Amersham PLC, Bucks, UK). The bound antibodies were analyzed using an ECL plus western blotting detection system (GE Healthcare, Boston, MA, USA) and a Lumino Image Analyzer (ImageQuant LAS-4000 miniPR; Cytiva, Krefeld, Germany). Original, whole membrane images were provided for the review process and are available from the journal upon request.

#### *4.8. Statistical Analysis*

Data are expressed as mean ± SE. Statistical comparisons were performed using the Student's *t*-test, Dunnett's test, or Tukey's test.

**Author Contributions:** H.S. (Hideo Satsu), Y.G. and H.S. (Hana Shimanaka) conceived and designed the experiments; H.S. (Hideo Satsu), Y.G., H.S. (Hana Shimanaka) and K.W. performed the experiments and analyzed the data; S.W., S.-J.P. and K.N. conceived and performed the informatic analysis; H.S. (Hideo Satsu), Y.G., H.S. (Hana Shimanaka), M.I., S.M. and M.S. discussed the data; and H.S. (Hideo Satsu) wrote the paper. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was funded in part by a research grant from Taisho Pharmaceutical Co., Ltd.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data supporting the findings of this study are available from the corresponding author, H.S., upon reasonable request due to restrictions on privacy.

**Acknowledgments:** We acknowledge Mika Takahashi for the technical support.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Appendix A**

**Table A1.** Primers used for constructing each TXNIP promoter vector.


**Table A2.** Partial TXNIP promoter sequence (+122/+256) of oligonucleotide for annealing.



0 **)**

**Table A3.** Partial TXNIP promoter (+174/+218) sequence of oligonucleotide for annealing. reverse agcttctccaaatcgaggaaacccctttgcaaaaaattatttcactttaaggggtac

reverse agcttctttaaggaattaaggtattcttaagcagtttgagcttaaaaataaaataagatttaaggtac

*Metabolites* **2022**, *12*, x FOR PEER REVIEW 13 of 15

+211/+256 forward cccttaaagtgaaataattttttgcaaaggggtttcctcgatttggaga

**Figure A1.** Effect of the PI3K inhibitor on taurine-induced increase in TXNIP mRNA in human hepatic HepG2 cells. HepG2 cells were pretreated with 20 µM LY294002 for 2 h and then cultured in medium containing 100 mM of taurine and the inhibitor. After 48 h, RNA was extracted and used for real-time PCR. The results are expressed as relative values, with the control value without LY294002 as 1. Each value is the mean ± S.E. (*n* = 4), and the ab values indicated by different characters are significantly different from each other and the ab values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05). **Figure A1.** Effect of the PI3K inhibitor on taurine-induced increase in TXNIP mRNA in human hepatic HepG2 cells. HepG2 cells were pretreated with 20 µM LY294002 for 2 h and then cultured in medium containing 100 mM of taurine and the inhibitor. After 48 h, RNA was extracted and used for real-time PCR. The results are expressed as relative values, with the control value without LY294002 as 1. Each value is the mean <sup>±</sup> S.E. (*<sup>n</sup>* = 4), and the ab values indicated by different characters are significantly different from each other and the ab values indicated by the same characters are not significantly different (Tukey's test; *p* < 0.05).

#### **References References**


24. Yu, H.; Cook, T.J.; Sinko, P.J. Evidence for diminished functional expression of intestinal transporters in Caco-2 cell monolayers at high passages. *Pharm. Res.* **1997**, *14*, 757–762. [CrossRef] [PubMed]

