The study findings indicate that ARSB silencing replicates the effects of hypoxia and are consistent with the requirement for molecular oxygen to activate ARSB. Furthermore, the results suggest that ARSB may mediate intracellular oxygen signaling through effects on chondroitin-4- sulfation. Under normoxic conditions, ARSB silencing increased HIF-1α activation and expression, and ARSB overexpression reduced HIF-1α in human bronchial and colonic epithelial cells, demonstrating the ability of ARSB to regulate HIF-1α activation and expression. In addition to similar effects on HIF-1α activation and expression, hypoxia and ARSB silencing have similar effects, including 1) increase in cellular sulfated GAG and C4S content, 2) decline in GSH/GSSG ratio and total sulfhydryl level, 3) increase or decrease in mRNA expression of 84 hypoxia-associated genes in a PCR array, 4) increase in galectin-3 nuclear localization, and 5) AP-1 activation. These effects suggest a central role for ARSB in the regulation of vital cellular processes.
In addition to the findings that indicate which ARSB silencing can mediate and replicate the effects of hypoxia, the study results provide evidence to explain how the transcriptional effects of hypoxia and ARSB silencing might be mediated. Reduced binding of galectin-3 to C4S following reduction of ARSB activity was demonstrated by decline in the amount of galectin-3 that co-immunoprecipitated with C4S following hypoxia or ARSB knockdown. Decline in binding to more highly sulfated C4S was associated with increased nuclear galectin-3. Increases in nuclear AP-1 components c-Fos and c-Jun were also shown following ARSB silencing and hypoxia. Galectin-3 has been reported to induce transcriptional effects on MUC2 in association with AP-1 in human colon cancer cells 
, and we propose that the effects of hypoxia and ARSB silencing on the activation of HIF-1α may be achieved through the AP-1 binding sites present in the HIF-1α promoter in the presence of the increased nuclear galectin-3 
. Additional effects of ARSB, as well as of oxygen, may be mediated through expression of HIF-1α, which has a broad repertoire of known effects, including impact on cell proliferation and epithelial-mesenchymal transition 
. Galectin-3, like p53 and c-Myc, has a nuclear localization motif and interacts with importins α and β for nuclear translocation and has a wide range of effects on vital cellular processes 
. The binding of other galectins to C4S may also be modified by changes in chondroitin sulfation when ARSB activity is reduced 
is a schematic representation of the signaling in normoxic and hypoxic pathways, mediated through changes in ARSB activity. Transcriptional events following hypoxia may proceed from reduced ARSB activity, thereby modulating chondroitin-4-sulfation, galectin-3 binding to chondroitin-4-sulfate, galectin-3 nuclear translocation, and AP-1 activation. Variation in the chondroitin-4-sulfation interaction with galectin-3 affects the availability of galectin-3 for nuclear translocation and subsequent transcriptional events involving interaction of galectin-3 with AP-1. The promoters for IL-6, HMOX1, PLAU, COL1A1, genes significantly upregulated by both hypoxia and ARSB silencing in the hypoxia PCR array (), have AP-1 binding sites, indicating a transcriptional mechanism by which they also may be upregulated. The impact of changes in GAG sulfation on nuclear translocation of oncogenes and other transcription factors is a subject for further study which may have significant implications for how genes are regulated by extra-nuclear, and even extracellular, events.
Schematic illustration of oxygen→ARSB→C4S→galectin-3→AP-1→HIF-1α signaling pathway.
In protists, yeasts, and plants, a sulfate assimilation pathway (SAP) has been described by which sulfate is progressively reduced, leading to production of glutathione, cysteine, and methionine 
. The SAP involves adenosine 5′-phosphosulfate (APS) and 3′-phosphoadenosine 5′-phosphosulfate (PAPS). In mammalian cells, PAPS synthetases (PAPSS1 and PAPSS2) have been identified that combine the APS and PAPS functions and activate sulfate 
. The current study findings demonstrate that diminished availability of sulfate due to reduced ARSB activity following hypoxia in the human epithelial cells leads to decline in the GSH/GSSG ratio and the sulfhydryl content, indicating a profound impact on the overall cellular redox status. This suggests the operation of a mechanism by which molecular oxygen can regulate sulfate assimilation through activation of ARSB in mammalian epithelial cells. The cascade of reactions from sulfate to reduced sulfur in the sulfate assimilation pathway (SAP) (also known as the sulfate activation complex) can be invoked to integrate oxygen signaling, sulfate metabolism, and redox status, since the cascade of reactions from sulfate to reduced sulfur would be impaired due to reduced availability of sulfate following inhibition of ARSB activity and reduced production of free sulfate. The bacterial SAP involves interaction with several critical enzymes, including thioredoxin and glutaredoxin that act as hydrogen donors for PAPS reductase 
, and is thermodynamically driven by GTP hydrolysis 
Effects on glutathione reduction impact on other complex redox processes, such as those of monothiol glutaredoxins that utilize reduced glutathione and a cysteine residue to bind to a bridging [2Fe-2S] cluster 
. Decreased ratio of reduced glutathione to glutathione disulfide is associated with selective promotion of disulfide bond formation within cytoplasmic proteins, leading to functional impairment 
. For normal cellular aerobic metabolism, coenzyme A (CoA), a sulfhydryl, is processed to the thioester acetyl CoA. The production of acetyl-CoA from pyruvate by the pyruvate dehydrogenase complex is inhibited by increase in the ratio of acetyl-CoA to CoA, reflecting the impact of a thiol on regulation of aerobic metabolism 
. Impairment of sulfate assimilation, due to reduced activity of ARSB, in association with reduced sulfhydryl content, would impair oxidative metabolism by inhibition of pyruvate dehydrogenase. This suggests a mechanism whereby impairment of aerobic metabolism, as hypothesized by Warburg 
, might follow from decline in ARSB activity, and is consistent with reported findings of reduced ARSB activity in malignant cells 
The extensive, unexpected replication between the effects of hypoxia and ARSB silencing in the BEC and NCM460 cells suggests that ARSB mediates some intracellular effects of oxygen. In turn, the effects of hypoxia on C4S and sulfated GAG content indicate that a pathway of sulfate metabolism may be regulated by oxygen through activation of ARSB. Effects of both hypoxia and ARSB silencing on the reduced glutathione-glutathione disulfide ratio and on the cellular sulfhydryl levels demonstrate ramifications of reduced availability of free sulfate on the overall cellular redox state. Since removal of the 4-sulfate residue of N-acetylgalactosamine-4-sulfate is the only known direct effect of ARSB, ARSB may act as a redox switch that regulates the sulfate assimilation pathway. Transcriptional effects may also be mediated by ARSB through chondroitin-4-sulfation and the associated changes in galectin binding to more or less highly sulfated GAGs. By integrating the effect of oxygen with 1) chondroitin sulfation, 2) sulfate assimilation and cellular redox status, and 3) transcriptional regulation through variation in the interaction of galectins with GAGs, ARSB is a critical link in our understanding of how fundamental cell chemistry can regulate vital cell processes.