Access Type

Open Access Dissertation

Date of Award

January 2019

Degree Type


Degree Name




First Advisor

Melissa Runge-Morris


SULTs are conjugation enzymes that can modify the activity of a myriad of foreign and endogenous molecules. SULT expression was detected in various human tissues, including liver, small intestine, and colon. There are 13 human SULT genes that are classified into 4 families, SULT1, SULT2, SULT4, and SULT6. In humans, SULT1 and SULT2 families include 11 genes that are further divided into 6 subfamilies. In addition to their role in xenobiotic detoxification and regulation of physiological processes, SULT enzymes were implicated in the bioactivation of procarcinogens. Previous studies detected the expression of most SULT1 and SULT2 enzymes during early development, as early as the embryonic stage. There is limited information about the developmental expression profiles and regulation of SULT1 and SULT2 enzymes in the liver and intestine. The objective of this study was to gain more insight into the roles of SULT1 and SULT2 enzymes during prenatal and postnatal periods in the two main metabolic organs, liver and intestine. To learn more about the regulation of SULT mRNA in differentiating liver cells, we first characterized their expression in primary cultures of human fetal hepatocytes and the HepaRG model of liver cell differentiation, and then examined the effect of treatment with activators of lipid- and xenobiotic-sensing receptors on SULT expression in these in vitro models. Using RT-qPCR analysis we demonstrated that SULT1A1 (transcript variants 1, TV1), SULT1C2, SULT1C4, SULT1E1, and SULT2A1 mRNA was the most abundant in human fetal hepatocytes. In HepaRG cells, SULT1C2 and SULT1E1 mRNA and protein increased during the transition from proliferation to confluency and then decreased as the cells underwent further differentiation whereas SULT2A1 mRNA and protein increased during differentiation. Like SULT1C2, SULT1C3, SULT1C4, and SULT1B1 mRNA levels were highest in the confluency stage. SULT1A1 and SULT2B1 mRNA levels remined relatively constant. Treatment of fetal hepatocytes as well as confluent and differentiating HepaRG cells with activators of aryl hydrocarbon receptor, constitutive androstane receptor, liver X receptor, peroxisome proliferator-activated receptors (PPARs), pregnane X receptor, and vitamin D receptor indicated that SULT1 and SULT2 mRNA is regulated by xenobiotic stimuli.

We also determined the developmental expression profiles of SULT expression in libraries of human liver specimens and cytosols that were collected from prenatal and postnatal (i.e. infants, children 1-18 years-old, and adults) donors using RT-PCR and RNA-seq analysis to measure SULT mRNA and multiple monitoring reaction (MRM) analysis for SULT protein quantification. In this dissertation we reported that SULT1A1 expression did not vary substantially during development; SULT1A3, SULT1C2, SULT1C4, and SULT1E1 expression was highest in prenatal and/or infant specimens; SULT1A2 and SULT2A1 expression was highest postnatally; and SULT1B1 mRNA, as determined by RT-qPCR analysis and protein appears to be highest in children and adults. SULT1A1 (TV5), SULT1C3, and SULT2B1 mRNA levels were low regardless of developmental stage. SULT1C4 mRNA was most abundant in the prenatal livers, but the protein levels were very low. To investigate the reason for this discrepancy we measured the mRNA levels of SULT1C4 TVs in the same human liver specimens described above and determined whether the individual variants can be translated into protein. Using RT-qPCR and RNA-seq analyses we detected at least four SULT1C4 transcript variants, including TV1, TV2, E3DEL, which were detected in the intestinal and hepatic cell lines we examined. These TVs were preferentially expressed in prenatal liver and TV2 was the most abundant of all. Using western blot analysis we found that only TV1 and TV2 are translated into protein, but TV2 protein was much lower than that of TV1. This finding suggests that TV2 is either less efficiently translated into protein than is TV1 or that the TV2 protein is more rapidly degraded, and thus could explain the lack of correlation between SULT1C4 mRNA and protein level. Therefore, we conclude that SULT1 and SULT2 expression is modulated by xenobiotics and that most of these enzymes play an important role in hepatic metabolism, especially during early life stages.

Lastly, we examined the mechanism underlying the transcriptional regulation of SULT1C3, which is one of the least studied SULTs, by PPARγ. While attempting to amplify a 2.8 Kb fragment from different sources of human genomic DNA, a 1.9 Kb fragment was sometimes co-amplified with the expected 2.8 Kb fragment. When aligning the 1.9 Kb fragment sequence to the published SULT1C3 5’-flanking sequence an 863 nt deletion (nt -146 to -1008 relative to the transcription start site) was revealed. Transfection of reporter plasmids containing the 2.8 and 1.9 Kb fragments into LS180 cells followed by treatment with PPARα, δ, and γ induced the luciferase expression of the 2.8 but not the 1.9 Kb construct and indicated that the 863 nt deletion region was sufficient to confer PPAR-inducible reporter expression. Three putative PPAR-response elements (PPRE) were identified by computational analysis. Serial deletions, site-directed mutations, and RNA interference analysis demonstrated that only the distal PPRE (at nt -769) was required to mediate PPARγ transcriptional activation of SULT1C3. Genotyping analysis revealed that a similar deletion exist in the human genome. These findings suggest that SULT1C3 play a role in the regulation of PPARγ-controlled pathways.