Nickel chloride was purchased from Sigma Aldrich/Merck KGaA (catalog no. N6136). Antioxidant treatments were performed with (−)-epigallocatechin gallate (EGCG) (Sigma Aldrich/Merck KGaA; catalog no. E4143), L-ascorbic acid (Asc; Thermo Fisher Scientific, Inc.; catalog no. A-61), and a-tocopherol/vitamin E (vit. E; Sigma Aldrich/Merck KGaA; catalog no. T3251). Hydrogen peroxide was purchased from Sigma Aldrich/Merck KGaA (catalog no. 216763). Immortalized human bronchial epithelial cells (BEAS-2B) were obtained from ATCC (ATCC^® CRL-9609), adapted to serum growth immediately after purchase, carefully maintained at below confluent density, and were authenticated by short tandem repeats (STR) analysis. Experiments were performed within approximately six passages (e.g. two weeks) from the time of thawing, and replicates were performed simultaneously. Ni-transformed BEAS-2B cells were previously generated and characterized (16). In brief, BEAS-2B cells were exposed to soluble Ni for 30 days, after which transformed clonal populations were selected based upon anchorage-independent growth of single cells in soft agar accompanied by a corresponding footprint of cancer-related gene expression changes (16). The 293 cell line was obtained from ATCC (ATCC^® CRL-1573) and authenticated by STR analysis. BEAS-2B, Ni-transformed BEAS-2B, and 293 cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Atlanta Biologicals), and 1% penicillin-streptomycin (Gibco; Thermo Fisher Scientific, Inc.; catalog no. 15140-122). Cells were incubated at 37°C in a humidified atmosphere of 5% carbon dioxide.

Stable transfections in BEAS-2B cells were performed using Lipofectamine LTX reagent with PLUS reagent (Thermo Fisher Scientific, Inc.; catalog no. 15338030). Control shRNA (Santa Cruz Biotechnology, Inc.; catalog no. sc-108060) and NUPR1 shRNA (Santa Cruz Biotechnology, Inc.; catalog no. sc-40792-SH) were used for stable knockdown. Transient transfections in BEAS-2B cells were performed using Lipofectamine RNAiMAX transfection reagent (Thermo Fisher Scientific, Inc.; catalog no. 13778030). Control siRNA (Invitrogen; Thermo Fisher Scientific, Inc.; catalog no. 12935112), siRNA against JUN (Invitrogen; Thermo Fisher Scientific, Inc.), and siRNA against FOS (Invitrogen; Thermo Fisher Scientific, Inc.) were used for transient knockdowns.

Cells were washed with ice-cold PBS 2X, lysed using RIPA buffer (150 mM NaCl, 1.0% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris, pH 8.0, 2 mM EDTA, 1 mM PMSF, and protease inhibitors; Roche Diagnostics; catalog no. 11 836 170 001) or boiling buffer (1% SDS, 4 mM Na₃VO₄, and 10 mM Tris-HCl (pH 7.4). Cells lysed with RIPA buffer were collected and incubated at 4°C for 30 min with constant agitation. Lysates were then centrifuged for 20 min at 4°C, 20,000 × g, transferred, and frozen at −80°C until use. Cells lysed with boiling buffer were collected, denatured at 100°C for 5–10 min, sonicated for 10 min, centrifuged at 20,000 × g for 15 min, transferred, and stored at −80°C until use. Protein was quantified using Pierce™ BCA protein assay kit (Thermo Fisher Scientific, Inc.; catalog no. 23225) according to the manufacturer's instructions. Approximately 30–70 µg of total protein was separated on 10–18% SDS-PAGE gels by electrophoresis, and proteins were transferred to a 0.2 µM nitrocellulose membrane (Bio-Rad) for 2 h at 100 V or overnight at 20 V. Membranes were blocked for non-specific binding sites in Tris-buffered saline containing 0.1% Tween-20 and 5% non-fat dry milk at room temperature. The membranes were immunoblotted with primary antibodies overnight at 4°C (JUN, 1:400, Santa Cruz Biotechnology, Inc., catalog no. sc-1694; FOS, 1:500, Cell Signaling Technology, Inc., catalog no. 2250; β-tubulin, 1:20,000, Proteintech, catalog no. 66240-1-Ig; β-actin, 1:15,000, Proteintech catalog no. 66009-1-Ig; and NUPR1, 1:200, Sigma Aldrich/Merck KGaA, catalog no. SAB1104559). Membranes were then incubated with HRP- or AP-conjugated secondary antibodies (HRP, goat anti-rabbit IgG, 1:5,000, Cell Signaling Technology, Inc., catalog no. 7074; HRP, goat anti-mouse IgG, 1:5,000, Santa Cruz Biotechnology, Inc., catalog no. sc-2005; AP, goat anti-rabbit, 1:5,000, Promega Corp., catalog no. S3731). Protein detection was performed using chemiluminescence (Thermo Fisher Scientific, Inc., catalog no. 32106) or chemifluorescence (Cytiva, catalog no. RPN5785) and developed using autoradiography or imaged using a Typhoon imager (GE Healthcare, model no. FLA 7000).

Cells were collected in Tri reagent (Molecular Research Center, catalog no. RT 111), and either stored at −80°C or processed immediately for RNA isolation. The quantity and purity of RNA extracted from each sample were determined by UV absorbance spectroscopy on a NanoDrop 2000 spectrophotometer system. Reverse transcription was performed using either LunaScript RT SuperMix (New England BioLabs, catalog no. E3010) or ProtoScript First Strand cDNA Synthesis (New England BioLabs, catalog no. E6300) with 500 ng of RNA in a final volume of 10 µl. Following denaturation of the first-strand cDNA product for 5 min, quantitative real-time PCR analysis was performed using Power SYBR Green PCR Master Mix (Thermo Fisher Scientific, Inc., catalog no. 4367659) on an Applied Biosystems QuantStudio 6 Flex system (Thermo Fisher Scientific, Inc.). Relative gene expression levels were normalized to an endogenous control and calculated using the DDCq method (17). The following primers were used: NUPR1: 5′-CTGGCCCATTCCTACCTCG-3′ (forward) and 5′-TCTCTTGGTGCGACCTTTC-3′ (reverse); JUN: 5′-GAGCTGGAGCGCCTGATAAT-3′ (forward) and 5′-CCCTCCTGCTCATCTGTCAC-3′ (reverse); FOS: 5′-GAATCCGAAGGGAAAGGAATAAG-3′ (forward) and 5′-TCCGCTTGGAGTGTATCAGTCA-3′ (reverse); ATF2: 5′-CATGGCCCACCAGCTAGAAA-3′ (forward) and 5′-GTATTGCCTGGCAGAATTCACA-3′ (reverse); FOSL1: 5′-CCTTGTGAACGAATCAGCCC-3′ (forward) and 5′-GTCGGTCAGTTCCTTCCTCC-3′ (reverse); TUBULIN: 5′-GCAAGGTATCCTAAG-3′ (forward) and 5′-CTCGTCCTGGTTGGGAAACA-3′ (reverse); and ACTIN: 5′-TGACGTGGACATCCGCAAAG-3′ (forward) and 5′-CTGGAAGGTGGACAGCGAGG-3′ (reverse), and GAPDH: 5′-TCAAGAAGGTGGTGAAGCAGG-3′ (forward) and 5′-AGCGTCAAAGGTGGAGGAGTG-3′ (reverse).

Two degenerate primers, 915NUPR1FD and 2394NUPR1RD (Table SI) were used initially to amplify BEAS-2B genomic DNA. A 1,480-base pair DNA fragment was restricted with BamHI (present in the primer sequences), and cloned into the BamHI site of pUC19 (pNUPR264). The presence of NUPR1 promoter fragment nucleotide 925 to nucleotide 2384 of NCBI GenBank (https://www.ncbi.nlm.nih.gov/genbank/) submission AF069074 was confirmed by sequencing. In order to include more NUPR1 promoter DNA sequences in this clone, the BamHI site was destroyed using Bal 31 on both ends to generate two separate clones (pNUPR287 for 3′ and pNUPR288 for 5′ deletion) (18). A 1.6-kb PCR product was amplified from BEAS-2B genomic DNA with primers 72NUPR1FD and 1724NUPR1NR, restricted with BamHI (present in 72NUPR1FD) and EcoNI (present in the 3′ end of the PCR product) and inserted into corresponding sites of pNUPR287. This new recombinant (pNUPR289) included an additional 842 base pairs of promoter sequences at the 5′ end of the initial clone (pNUPR264). Similarly, 1,125 base pairs of DNA were amplified with primers 2099NUPR1NF and 3223NUPR1NRD, restricted with EcoRV (present in the 5′ end of the PCR product) and BamHI (present in the primer 3223NUPR1NRD) and inserted into corresponding sites of pNUPR288. This new recombinant (pNUPR290) contained an additional 699 base pairs of promoter sequence at the 3′end.

For NUPR1-luciferase vector construction, a 2,054 base pair KpnI-EcoRV 5′ promoter fragment from pNUPR289 and 1,074-base pair EcoRV-HindIII 3′ fragment from pNUPR290 was inserted into a pGL4.17 vector (Promega Corp.) into their KpnI-HindIII sites to generate the full length NUPR1 promoter construct, pNUPR308 (Fig. S1A). The recombinant construct was sequenced in both directions and is identical to the published sequence, except for an additional 19 base pairs (CAAGTATCCTGTCTTCACT) after nucleotide 957 of GenBank submission of AF069074. To further investigate promoter activity in the cloned sequence using promoter bashing technique, the 5′ end of this sequence was deleted sequentially using Bal 31 as previously described (18). Six recombinants (NUPR309-314) with increasing amount of deletions at the 5′ end of the NUPR1 promoter were selected and sequenced to determine the extent of deletion (Table SII).

Cells were stably transfected with pGL4-basic reporter vector containing the full length NUPR1 promoter or the NUPR1 promoter with deletions varying in length and position located upstream of the luciferase gene (Fig. S1A). Approximately 13,500 cells were seeded overnight into each well of a 48-well plate. After seeding, the medium was refreshed, and the cells were unexposed or exposed to Ni for 24 h. Luciferase reporter system (Promega Corp., catalog. no. E1500) was used to detect the luminescence intensity, and cell lysate preparation and luminescence measurements were conducted according to the manufacturer's protocol. All measurements were adjusted for total protein, normalized to pGL4 control, and performed in triplicate.

ALGGEN PROMO software V 3.0.2, a virtual laboratory for the identification of putative transcription factor binding sites in DNA sequences, was used to determine the transcription factor binding sites within the NUPR1 promoter region (19). Cells were stably transfected with a pGL4 vector containing the full length NUPR1 promoter (pNUPR308) or NUPR1 full length promoter with JUN transcription factor binding site deleted (−2339 to −2333) in the NUPR1 promoter positioned upstream of the luciferase gene (Fig. S1A). Deletions were made using the Q5^® site-directed mutagenesis assay kit (New England BioLabs), and the resulting plasmid was sequenced (Genewiz, http://www.genewiz.com/) to confirm the deletion of the JUN binding site in question (Fig. S1B). The luciferase gene reporter assay was conducted as described herein. All luciferase measurements were adjusted for total protein, normalized to pGL4 vector control, and performed in triplicate.

Anchorage-independent growth was determined by the ability of cells to grow in soft agar. A bottom layer of 0.5% 2-hydroxyethylagarose (Sigma Aldrich/Merck KGaA, catalog no. A4018), and top layer containing 5,000 cells in 0.35% 2-hydroxyethylagarose was place in a 6-well plate. After two weeks, the wells were stained with 500 µl INT/BCIP (Roche Diagnostics, catalog no. 11 681 460 001), and prepared according to the manufacturer's instructions. Images of each stained well were acquired using a Bio-Rad Molecular Imager Gel-Doc XR⁺ system and Image Lab software (Bio-Rad Laboratories, Inc.). Colony numbers were calculated using ImageJ (NIH, V 1.52q). When seeding cells in soft agar, 200 cells were simultaneously seeded into a 100-mm dish in order to determine the plating efficiency in monolayer culture, which is defined as the ratio of the number of colonies (those formed in a cell culture dish) vs. the number of cells seeded. After a 12-day incubation, the plates were fixed and stained overnight with 5% Giemsa in a 5:6 methanol:glycerol solution. After destaining, all cell colony numbers were counted using ImageJ (NIH) and plating efficiencies were determined. All soft-agar assays were adjusted for plating efficiency and performed in triplicate.

The cells were rinsed with PBS, briefly trypsinized, and neutralized with complete medium. Following neutralization, the cell suspension was then passed through a 40-µm cell strainer (Celltreat, catalog no. 229482) to eliminate cell clumps. Two hundred cells were then reseeded into each of three 100-mm dishes, and grown for 3 weeks. Surviving colonies were stained with Giemsa and counted using ImageJ (NIH, V 1.52q). All colony formation assays were conducted in triplicate.

ImageJ (NIH) was used to quantify western blots, and to determine cell colony numbers. Western blot images were converted into 8-bit JPEG images, and the intensity of protein bands were normalized to the loading control. For colony formation assays, images were captured using a Bio-Rad Molecular Imager Gel-Doc XR⁺ system and Image Lab software (V 2.0.1., Bio-Rad Laboratories, Inc.). Images were converted into 8-bit JPEG images, and the background was reduced using a median filter to eliminate non-colonies. Graphical depictions and statistical analyses were performed using GraphPad Prism 9 (GraphPad Software, Inc.). Differences between groups were compared using either a Student's t-test or ANOVA followed by a Tukey post hoc multiple comparisons test. Differences were considered statistically significant at P<0.05. *P<0.05 and **P<0.01, as shown in the figures.

Cr(VI), an IARC group I human lung carcinogen, induces NUPR1 mRNA and protein levels in human bronchial epithelial (BEAS-2B) cells (10). Ni is also a group I human lung carcinogen; however, the mechanisms of carcinogenesis between Cr(VI) and Ni differ in many respects. Cr(VI) is a classical carcinogen in the sense that it strongly binds DNA and proteins, creating protein and/or DNA adducts and subsequently, mutations; however, it also produces a high degree of oxidative stress through intracellular reduction to Cr(III) and acts carcinogenically through epigenetic mechanisms (20,21). Ni, on the other hand, is often thought of as a nonclassical carcinogen, and acts carcinogenically by mimicking hypoxia, which is common in tumors, as well as by dysregulating the epigenetic program of cells (13). Both Ni and Cr have been shown to induce AP-1, and multiple AP-1 sites reside in the promoter region of NUPR1 [reviewed in ref. (21)] (22–24). Therefore, the possibility of whether Ni is capable of inducing NUPR1 expression in BEAS-2B cells was investigated.

BEAS-2B cells were acutely exposed to 0, 0.25, 0.5, 0.75, and 1.0 mM Ni for 12 h, and the mRNA levels of NUPR1 were measured by real-time PCR and protein expression was determined by western blot analysis. The levels of NUPR1 mRNA were dose-dependently increased following exposure of BEAS-2B cells to Ni (Fig. 1A) as were the protein levels of NUPR1 (Fig. 1B). Dose-dependent induction of NUPR1 protein and mRNA was also evident in BEAS-2B cells after 6 h of Ni exposure (data not shown). BEAS-2B cells were then exposed to 0, 0.05, and 0.1 mM Ni to explore whether extended exposure to Ni can induce NUPR1 expression. As shown in Fig. 1C, extended exposure to 0.1 mM Ni was capable of inducing NUPR1 protein levels by 3.1-fold after 7 days. NUPR1 mRNA levels were unchanged, however (data not shown). In order to determine if Ni induces NUPR1 in a time-dependent manner, BEAS-2B cells were exposed to 1.0 mM Ni for various time points up to 12 h. After 3 h there was only a 2.9-fold increase in NUPR1 mRNA expression (Fig. 1D). Surprisingly, however, NUPR1 mRNA was highly induced at 9 h (17.3-fold), and decreased through 12 h, after having peaked at 9 h. There was no significant time-dependent difference in NUPR1 mRNA expression in unexposed BEAS-2B cells over 12 h (data not shown). Collectively, these results show that NUPR1 is induced at both the protein and mRNA levels following dose- and time-dependent exposure to Ni. Furthermore, this suggests that NUPR1 is quickly and transiently induced by Ni, which is characteristic of early stress-response genes (25).

BEAS-2B cells were exposed to 0, 0.125, 0.25, 0.5, and 0.75 mM Ni for 24 h, and the protein expression of AP-1 subunits, JUN, FOS, and ATF2, was investigated. Both JUN and ATF2 protein expression were found to be dose-dependently increased following Ni exposure (Fig. 2A). However, unlike JUN and ATF2, FOS protein expression was undetected after 24 h of exposure to Ni at increasing doses. Instead, FOS protein expression was found to be quickly and transiently induced by 1.0 mM Ni after only 1 h, and its expression proceeded to increase until approximately 6 h when its expression then started to decrease (Fig. 2B). This was not determined to be the case for both JUN and ATF2, as their protein expression was unaltered over the course of 9 h in response to 1.0 mM Ni (Fig. 2B).

Transcription factor AP-1 is a heterodimer complex consisting of subunits from numerous protein subfamilies [reviewed in ref. (26)]. In lieu of this, the expression of additional AP-1 subunits was investigated following exposure to Ni. BEAS-2B cells were exposed to a single dose of 0.5 mM Ni for 24 h. FOSL1, a FOS-related member of the AP-1 complex, was found to be transcriptionally induced by Ni (Fig. 2C). However, no other AP-1 subunits investigated (i.e. JUNB, JUND, FOSL2, FOSB, and JDP2) were found to be highly elevated (data not shown). Collectively, these results showed that following Ni exposure, AP-1 subunits, JUN and ATF2, were dose-dependently induced and remained induced through 24 h in response to Ni exposure and likely contribute to NUPR1 induction over this time period. FOS, however, was quickly and transiently induced in response to Ni exposure, which indicates that FOS may be an early-response gene responsible for increased NUPR1 expression during the first 12 h of exposure.

NUPR1 was previously demonstrated to be important in Cr(VI)-induced cell transformation (10). Therefore, NUPR1 expression was measured in Ni-transformed cells. Real-time PCR was used to measure the mRNA expression of NUPR1 in Ni-transformed clones, and we observed increased expressions of NUPR1 mRNA in 4 out of 5 Ni-transformed clones, with extremely high levels in Ni-2 clones (Fig. 2D). Overall, the ability for Ni to induce NUPR1 in BEAS-2B cells and elevated expression of NUPR1 in BEAS-2B cells transformed by Ni suggests that NUPR1 likely plays a role in Ni-mediated carcinogenesis.

Since both AP-1 and NUPR1 were determined to be induced by Ni (Figs. 1 and 2) and AP-1 binding sites are present in the NUPR1 promoter, transcriptional induction of NUPR1 by AP-1 was investigated. Both JUN and FOS were transiently knocked down in BEAS-2B and 293 cells, and were exposed to 0.5 mM Ni for 24 h. JUN mRNA expression was reduced by at least 70% in both BEAS-2B and 293 cells (Fig. 3A). In both BEAS-2B and 293 cells silencing of JUN attenuated NUPR1 induction by Ni from 8.8 to 5.1-fold and 10 to 0.4-fold, respectively (Fig. 3B). FOS silencing was found to be highly lethal in both BEAS-2B and 293 cells, and knockdown efficiencies of only approximately 20% were achievable (Fig. 3C). Despite the low knockdown efficiency, FOS knockdown in both BEAS-2B and 293 cells also reduced NUPR1 induction from 14.1 to 2.8-fold and from 7.0 to 1.9-fold, respectively (Fig. 3D). These data support the notion that AP-1 is important in NUPR1 mRNA induction in response to Ni exposure in both BEAS-2B and 293 cells.

ROS upregulates AP-1, specifically JUN and FOS at the transcriptional level [reviewed in ref. (26)]. In addition, Ni exposure produces time- and dose-dependent increases in ROS (26). Albeit, ROS generation by Ni is relatively low compared to that generated by other metals such as cobalt or chromate, and is generally not considered a major mechanism of Ni-mediated carcinogenesis (20,26). Therefore, due to the connection between ROS and AP-1 induction and the capacity for Ni to generate some degree of ROS, the possibility that ROS induces NUPR1 was investigated. BEAS-2B cells were treated with 0.1 mM of hydrogen peroxide (H₂O₂) for 6 h, and NUPR1 mRNA expression was measured. As shown in Fig. 4A, H₂O₂ was capable of inducing NUPR1 in BEAS-2B cells; however, compared to 8-fold increase on NUPR1 mRNA by 6 h Ni exposure, NUPR1 induction by H₂O₂ (2.5-fold) was considerably lower. In order to further investigate a possible connection between ROS production by Ni and NUPR1 induction, BEAS-2B cells were co-treated with antioxidant and ROS scavenger, epigallocatechin gallate (EGCG; 0.025 mM), and Ni (1.0 mM) for 6 h. Co-treatment with EGCG was not able to significantly prevent NUPR1 induction by Ni (Fig. 4B). In addition, BEAS-2B cells were pre-treated for 2 h with 0.1 mM (Asc) or 0.1 mM vitamin E (vit. E) and then exposed to 1.0 mM Ni for 6 h. Pre-treatment with either antioxidant was also unable to prevent NUPR1 induction by Ni (Fig. 4C and D). On the contrary, pre-treatment with ascorbate followed by Ni exposure actually significantly increased NUPR1 expression compared to Ni alone (Fig. 4C). This may be due to the fact that tissue culture media lacks the amount of ascorbate found in vivo, and usually only contains approximately 0.05 mM ascorbate, which is the amount in 10% fetal bovine serum. High levels of ascorbate can be toxic to tissue culture cells because ascorbate can also be a pro-oxidant. Collectively, these results suggest that NUPR1 induction by Ni primarily occurs through a mechanism other than ROS generation.

The NUPR1 promoter contains binding sites for numerous transcription factors involved in cellular stress including xenobiotic response elements, antioxidant response elements, and binding sites for x-box-binding protein 1, which classically functions in the unfolded protein response. In order to fully elucidate the mechanism for NUPR1 transcriptional activation and induction following exposure to Ni, deleted segments of varying length of the NUPR1 promoter were made and BEAS-2B cells were stably transfected with luciferase gene reporter constructs containing either the full-length NUPR1 promoter (construct pNUPR308), or a series of luciferase gene reporter constructs containing the NUPR1 promoter with deleted segments, which ranged in size from 299–2074 base pairs in length (Fig. S1A). Stable transfectants were treated with 1.0 mM Ni for 24 h to determine the regions responsible for NUPR1 transactivation. Fig. 5 shows NUPR1 promoter activity of the full-length NUPR1 promoter (pNUPR308) with and without Ni exposure. Luciferase activity in Ni-treated transfectants with the full-length promoter (pNUPR308) was significantly higher than unexposed full-length promoter transfectants (128-fold vs. 58-fold, respectively), which demonstrates that the NUPR1 promoter is highly active following exposure to Ni and is consistent with increased NUPR1 expression.

Luciferase activity in stable transfectants containing the NUPR1 promoter with various deletions following exposure to 1.0 mM Ni for 24 h was subsequently evaluated. A striking reduction in luciferase activity for both unexposed and Ni-exposed stable transfectants containing the NUPR1 promoter with the upstream most 299 base pairs deleted (pNUPR309; NUPR1 promoter −2135 to +713) was found (Fig. 5). From these results, it was inferred that this region must be important for high levels of NUPR1 transactivation. However, to determine which transcription factor(s) is critical to NUPR1 transactivation following Ni exposure, the NUPR1 promoter requires extensive examination by deleting specific transcription factor binding sites.

AP-1 subunits, JUN and FOS, were determined to be induced following Ni exposure in BEAS-2B cells. Additionally, knockdown of JUN and FOS significantly suppressed NUPR1 induction by Ni. Both JUN and FOS (i.e. AP-1) have broad transcriptional repertoires, and are involved in many aspects of tumor biology including tumor cell proliferation, apoptosis and survival of tumor cells, and invasive growth and angiogenesis (reviewed in (25). Moreover, a putative JUN binding site is located in this upstream most region of the NUPR1 promoter (−2339 to −2333). Consequently, the role that JUN plays in the transcriptional upregulation of NUPR1 was investigated. An NUPR1 promoter construct without the JUN binding site corresponding to position (−2339 to −2333) was generated (pNUPR_JUN; Fig. S1B). BEAS-2B cells were then stably transfected with the construct, and exposed to 1.0 mM Ni for 24 h or were unexposed. Fig. 5 shows that upon deleting the JUN binding site at (−2339 to −2333), a significant reduction in promoter activity in both the Ni exposed and unexposed cells was detected. This decrease in promoter activity mimics that which was generated using the NUPR1 promoter construct deletion corresponding to pNUPR309; NUPR1 promoter (−2135 to +713). Therefore, we conclude that the JUN binding site at position (−2339 to −2333) in the NUPR1 promoter is important for NUPR1 transactivation and that JUN acts by enhancing NUPR1 transactivation.

It has been reported on several occasions that NUPR1 expression is associated with tumorigenesis in vivo, and in non-small cell lung cancer cells, NUPR1 knockdown has been reported to reduce cell proliferation and colony formation ability (4,12). Therefore, the role that NUPR1 plays in cell proliferation in Ni-transformed cells was investigated. Ni-transformed cells were originally generated by treating BEAS-2B cells with Ni for 30 days and selecting single colonies that developed the ability to grow anchorage-independently in soft agar over a 4 week period (16). Ni-transformed cells were stably transfected with either control shRNA (ctrl) or NUPR1 shRNA (NUPR1 KD). Knockdown efficiencies in two Ni-transformed cell lines were determined to be approximately 50% (Fig. 6A). A colony formation assay was performed with NUPR1 knockdown Ni-transformed BEAS-2B cells to determine if NUPR1 knockdown suppresses colony formation and cell proliferation. As shown in Fig. 6B, colony formation was significantly suppressed in both Ni-transformed cell lines upon NUPR1 knockdown. Representative images are shown in Fig. S2A. Therefore, NUPR1 likely played an important role in cell proliferation in both Ni-transformed BEAS-2B cells.

In normal epithelial cells the occurrence of anoikis, or programmed cell death by disruption of the interactions with the extracellular matrix in suspension culture, renders cells anchorage-dependent (27–29). In most squamous epithelial cells and their transformed counterparts, growth in single-cell suspension is prevented, as cells undergo differentiation and ultimately anoikis (29). Anchorage-independent growth is considered a hallmark of cancer cell growth, particularly with respect to metastatic potential, and therefore, was assessed as an indicator of cell transformation and chemical carcinogenesis (30,31). NUPR1 overexpression was previously shown to induce cell transformation and knockdown of NUPR1 prevented Cr(VI)-induced cell transformation (10). Since Ni-transformed cells showed reduced cell proliferation when NUPR1 was stably knocked down, we sought to determine if Ni-transformed-NUPR1 knockdown cells would forfeit their ability to grow anchorage-independently. Results from the anchorage-independent growth assay showed reduced colony formation in soft agar in NUPR1 knockdown Ni-transformed cells compared to Ni-transformed shRNA control cells (Fig. 6C). Representative images of soft agar colonies are also depicted in Fig. S2B. From this it can be concluded that NUPR1 likely plays an important role in anchorage-independent growth in Ni-transformed BEAS-2B cells.