The mouse study was approved by The National Animal Research Authority (NARA) (approval ID: 07/11167, ID-178).

Female BW and female age-matched BALB/c mice were purchased from Jackson Laboratory, Bar Harbor, Main, USA. Kidney tissue was collected as described (Fenton et al., 2009), and BW mice were grouped according to stages of murine lupus nephritis. Group 1 pre-nephritic mice (no glomerular deposits of chromatin or IgG), Group 2 mice with mesangial nephritis (mesangial deposits of chromatin-IgG complexes), and Group 3 mice with end-stage disease (deposits of chromatin-IgG complexes in mesangium and in GBM, Fenton et al., 2009).

The recombinant proteins used in this study were: human TNFα (210-TA) and IL-1β (201-LB 1) from R&D system (Minneapolis, MN, USA), and Fas ligand (FasL) protein (ab109359) from Abcam (Cambridge, UK). Caspase 3 control cell extracts (untreated and cytochrome c treated Jurkat cells) were obtained from Cell Signaling Technology (Boston, USA). Alexa Fluor® 488 annexin V was purchased from Life Technologies (California, USA). Kineret (anakinra)—IL-1 receptor antagonist (IL-1Ra) was obtained from SOBI, Sweden.

The following antibodies were used in this study: rabbit anti-DNase I (ab113241, Abcam-Cambridge, UK), goat anti-Trap1(sc-69289), and rabbit anti-DNase I (sc30058) from (Santa Cruz, Texas 75220 U.S.A), rabbit anti-DNase I (LS-B4846), rabbit anti-FasR (LS-C152529, SC-74540), rabbit anti-FasL antibody (LS-C176019), mouse anti-FasL antibody (LS-C35867), and mouse anti-caspase 3 antibody (LS-B2125) were from Lifespan Biosciences (Seattle, WA). Cleaved caspase 3-specific antibody (9661) were from Cell Signaling Technology (Europe, B.V.), rabbit anti-actin antibody (A2066) from Sigma-Aldrich (St. Louis, MO) and mouse IL-1 beta /IL-1F2 Antibody (AF-401-NA) from R&D System. Alexa 595-conjugated anti-rabbit IgG, Alexa 488-conjugated anti-goat IgG, DAPI and Alexa 488-conjugated anti-mouse IgG secondary antibodies were all from Invitrogen (California, USA). Alexa Fluor® 488 annexin V was purchased from Life Technologies (California, USA), HRP-conjugated anti-rabbit IgG were from Invitrogen (California, USA), and HRP-conjugated anti-mouse IgG were purchased from Dako (Oslo, Norway).

Primary human renal proximal tubule epithelial cells (RPTEC) (Clontec, Lonza, Basel, Switzerland) have recently been validated for the present experiment (Thiyagarajan et al., 2015). The cells were grown in Clontec REGM™ BulletKit (CC-3190) containing Renal Epithelial Cell Basal Medium with the following growth supplements: hEGF, Hydrocortisone, Epinephrine, Insulin, Triiodothyronine, Transferrin, GA-1000, and fetal bovine serum, at 37°C in 95% humidified air and 5% CO₂. The cells were grown to 80% confluence (without growth arrest) and stimulated with TNFα (0, 20 ng/ml) or IL-1β (0, 2.5 ng/ml), if not otherwise stated in the text. These concentrations were based on results from serial-dilution experiments as demonstrated below. The cells were harvested at 24 and 48 h for mRNA analysis by qPCR and for protein expression analysis by Western blot and confocal microscopy. All cell stimulation experiments were performed in triplicates and repeated at least three times.

In this study, 2 × 10⁵ cells/well in six-well dishes and 2 × 10⁴ cells/well in confocal plates were seeded 24 h before transfection. After 24 h the media was removed and transfection was proceeded with fresh media containing 20 ng/ml TNFα without and with 20 nM scrambled smart pool DNase I siRNA or IL-1β siRNA (Dharmacon) and negative siRNA control (Sigma) using Lipofectamine 2000 (Invitrogen). The cells were harvested after 48 h after transfection. siRNA against P62 was used as a positive control for transfection in separate wells.

RPTEC were grown to 80% confluence and cells were stimulated with 2.5 ng/ml of IL-1β alone or together with 200 ng/ml of IL-1 receptor antagonist (IL-1Ra). Cells were harvested after 0-6-12-24-42-48 h and analyzed by qPCR.

RPTEC were grown to 80% confluence and cells were stimulated with 20 ng/ml of TNFα for 48 h. Cell supernatant and cell lysate from both unstimulated and TNFα stimulated cells were analyzed for IL-1β using IL-1β Quantikine ELISA kit from R&D systems. Quantitative determination of human IL-1β in both cell supernatant and cell lysate were performed according to the instruction by the manufactures. Experiments were run in triplicates and results were represented from two individual experiment.

For the FasL experiment, RPTEC were grown to 80% confluence and cells were stimulated with 20 ng/ml of TNFα for 48 h. After 48 h of stimulation, RPTEC were incubated with various concentrations of FasL protein (0, 25, 125 ng/ml) and grown for 48 h. Thereafter the cells were harvested and stored at −80°C. The apoptotic cells were analyzed and counted using TUNEL assay and Annexin V staining.

Total RNA was isolated from RPTEC using TRIzol ® (Invitrogen, CA, US), as described by the manufacturer. RNA quality and concentration were determined spectrophotometrically by Nano Drop (Nano Drop technologies, Wilmington, USA). RNA samples were reverse-transcribed with random primers using High Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, USA).

Quantitative real time PCR (qPCR) was performed using ABI Prism 7900HT Sequence Detection System (Applied Biosystems). Pre-designed FAM-labeled gene expression assays (Applied Biosystems) were purchased and accession numbers for human analyses are given in Table 1. TATA binding protein (TBP) was used as endogenous control. The relative expression levels were calculated using the ddCT method.

For Western blot analysis, cells were harvested using LDS buffer (Invitrogen) and the protein concentration was determined using protein quantification assay from Macherey-Nagel (Duren, Germany) according to the manufacturer instruction. Approximately 10 μg protein per well were loaded onto a 4–12% Nu PAGE Bis-Tris gel (Invitrogen). SDS-PAGE and Western blotting were performed according to standard procedures as described in detail (Kalaaji et al., 2006) and actin was used as a loading control.

RPTEC were seeded in a special 8 μ well-slide (IB80826) from Ibidi (Munich, Germany). The cells were fixed in 4% paraformaldehyde (KEBO, Oslo, Norway) in 1xPBS and incubated for 10 min on ice. The cells were permeabilized using 0.1% Triton X-100 (VWR International Oslo, Norway) in PBS for 5 min, followed by blocking in 3% BSA for 30 min and incubation with the appropriate primary and secondary antibodies for 60 and 30 min, respectively. Protein localization was detected using Zeiss-LSM510 Meta confocal microscopy (Oberkochen, Germany).

Immunohistochemical (IHC) staining of DNase I was performed as described and Polink-2 Plus HRP with DAB kit (Newmarket Scientific, UK) was used as detection system. IHC staining of IL-1β was performed on frozen kidney sections as previously described (Hedberg et al., 2013).

The chromatin immuno-precipitation (ChIP) experiment was performed using anti-DNase I antibody as the assay-specific antibody, or non-specific rabbit IgG immunoglobulins as a negative control. Sheared chromatin and ChIP using anti-histone H3 antibodies were included as positive controls. The gene to which DNase I was bound was identified by qPCR. The ChIP assay was performed using unstimulated RPTEC or RPTEC stimulated with 20 ng/ml of TNFα. Cells were harvested 48 h after stimulation, and a Magna ChIP A assay (Merck Millipore, MA, USA) was performed according to the manufacturer's protocol. Briefly, proteins were cross-linked to the DNA with formaldehyde, cells were lysed, and samples were sonicated twice for 30 s on ice at a 22-mm amplitude using a Soniprep 150 (MSE, London, UK). Samples were incubated with 5.0 μg of rabbit polyclonal anti-DNase I antibody (ab113241 Abcam Inc., MA, USA) and with 5.0 μg of rabbit polyclonal anti-DNase I antibody (sc30058) with appropriate controls. Immuno-precipitates were enriched with protein A magnetic beads. Subsequently, chromatin complexes were eluted, the crosslinks were reversed, and DNA was isolated. The presence of FasR DNA was evaluated using qPCR with primers corresponding to the FasR promoter region (5′ GGGTCTTCCTCATGGCACTAAC 3′ and 3′CTCCTGAGGGCTTTCCATCAC 5′). Thereafter, samples were loaded on a 2% agarose gel and visualized using ProXima C16 software version 3.0 (Isogen Life Science, De Meern, The Netherlands).

Annexin V staining was performed according to the manufacturer's instructions. In short, unstimulated, TNFα-stimulated or TNFα-stimulated in presence of FasL, adherent and non-adherent RPTEC were washed in cold phosphate-buffered saline (PBS). Washed non-adherent cells were centrifuged, the supernatant discarded, and the cell-pellet was resuspended in annexin-binding buffer (10 mM HEPES, 140 mM NaCl, 2.5 mM CaCl₂, pH 7.4). Adherent and non-adherent RPTEC were then diluted to a cell density of about ~1 × 10⁶ cells/ml. In situations where cell death was marginal, all non-adherent RPTEC were pelleted and subjected to analyses. Five microliters of Alexa Fluor® 488 annexin V were added to each 100 μl of cell suspension before the cells were incubated at room temperature for 15 min. Next, the cells were subjected to DAPI for 1 min and washed with annexin-binding buffer. The non-adherent cells were placed on a slide. Both the non-adherent and adherent annexin V and DAPI stained RPTEC were subjected to a Carl Zeiss LSM 700 confocal microscopy for fluorescence detection.

The terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) kit was purchased from Roche (Indiana, USA) and the assay was performed according to the manufacture's instruction for both non-adherent and adherent cells. Briefly, unstimulated, TNFα stimulated and TNFα stimulated and FasL treated RPTEC cells were washed before the adherent cells were fixed in 4% paraformaldehyde (PFA) in PBS for 10 min on ice. The non-adherent cells were fixed with 2% PFA for 60 min in room temperature prior to centrifugation and removal of supernatant. The cells were permeabilized using 0.1% Triton X-100 (VWR International) in PBS for 5 min. The cells were subsequently treated with TUNEL Mix containing TUNEL label (11767291910; Roche) and the terminal deoxynucleotidyl transferase enzyme (11767305001; Roche), and incubated at 37°C for 1 h. The cells were washed with PBS and nuclei were stained by DAPI and subsequently subjected to confocal microscopy using the Carl Zeiss LSM 700 confocal microscopy. As a positive control in the TUNEL assay, RPTEC were exposed to 10U recombinant DNase I to induce single-stranded (TUNEL-positive) nicks, and the negative control was represented by TUNEL in absence of terminal deoxynucleotidyl transferase.

RPTEC were treated with 20 ng/ml of TNFα with or without 125g of FasL for 48 h. Following treatments, caspase 3/7 activity were measured using the Caspase-Glo 3/7 assay kit (Promega, Madison USA). Briefly, plates with cells were removed from the incubator and allowed to equilibrate to room temperature for 30 min. Caspase-Glo reagent (100 μl) was added to each well before it was gently mixed with a plate shaker at 300–500 rpm for 30 s. The cells were then incubated at room temperature for 30 min−1 h. The luminescence of each sample wasmeasured in a Clariostar microplate reader (BMG-Labtech, Ortenberg, Germany). The experiments were performed in triplicate and repeated in three independent experiments.

Graph pad Prism was used to calculate the significance and the data are presented as mean of three parallels (± SD). An unpaired t-test was performed to test differences between the cytokine stimulated/siRNA transfected and non-transfected cells and for ELISA. One-way ANOVA with Dunett post-hoc test was performed at different concentrations of cytokines to non-stimulated cells and for cas3/7 GLO Assay. For each parameter; p < 0.05 was considered significant and represented as ^*. If P-value is ≤ 0.05 represented as ^* ≤ 0.005 as ^** and P-value ≤ 0.0005 as ^***. All observations were included and Spearman was used for significance testing.

Fuji Image J program was used to measure the nuclear staining intensity for all the confocal images.

As we have previously demonstrated (Thiyagarajan et al., 2015), stimulation of RPTEC with 20 ng/ml of TNFα in vitro for 48 h upregulates DNase I protein (Figure 1A). Nuclear intensity of DNase I after TNFα stimulation shows a significantly higher staining in the nuclei of TNFα stimulated cells (Figure 1B). Stimulation of RPTEC with 20 ng/ml of TNFα significantly upregulates DNase I mRNA (Figure 1C). Also, TNFα upregulates IL-1β mRNA levels (Figure 1D) and IL-1β protein levels (Figure 1E) and translocates DNase I into the nucleus (Figure 1A, Trap1 (green) is used as a cytoplasmic marker). Because TNFα is a potent inducer of apoptosis (Krown et al., 1996; He and Ting, 2002), the nuclear translocation of DNase I in response to TNFα is probably linked to an apoptotic process. Particularly, the FasR mRNA (Figure 1F) and the FasR protein expression levels (Figure 1G) were significantly upregulated in TNFα-stimulated cells compared with unstimulated cells. As shown in Supplementary Figure 1A, TNFα-stimulation of RPTEC induced FAS gene expression in a dose-depended manner. While FasR mRNA and FasR protein were constantly expressed in unstimulated RPTEC and increased upon TNFα-stimulation, FasL mRNA (qPCR, Supplementary Figure 1B) were undetectable after stimulation of the cells by TNFα or by IL-1β (Supplementary Figures 1B,C).

Since TNFα-stimulation of RPTEC upregulates IL-1β expression levels (Figures 1D,E) (Turner et al., 2007; Thiyagarajan et al., 2015), we analyzed whether IL-1β was involved in regulation of DNase I expression levels and/or its nuclear translocation. As shown in Figure 2A, IL-1β-stimulation did not significantly affect expression levels of DNase I in situ as demonstrated by confocal microscopy (Figure 2A). However, nuclear intensity measurements of DNase I after IL-1β stimulation show that IL-1β clearly promotes a nuclear translocation of DNase I (Figures 2A,B and Supplementary Figures 2A,B). Percentage of nuclear stained cells in stimulated vs. unstimulated cells were calculated and showed nuclear staining of almost 85% of stimulated cells compared to unstimulated cells (Supplementary Figure 2A). IL-1β did not affect the DNase I mRNA (Figure 2C), nor the DNase I protein expression levels significantly (Figure 2D). However, Importantly, IL-1β-stimulation induced a significant upregulation of FasR mRNA (Figure 2E) and FasR protein levels (Figure 2F), while it did not affect FasL expression levels in RPTEC (Supplementary Figures 1C). Interestingly, low doses of IL-1β were sufficient to induce upregulation of FasR mRNA levels (Figure 2E, Supplementary Figure 3A). Corresponding mRNA levels of IL-1β in response to different doses of IL-1β stimulation are demonstrated in Supplementary Figure 3B. This experiment indicates that as little as 0.037 ng/ml is sufficient to significantly upregulate both FasR and IL-1β mRNA levels above limits necessary for DNase I translocation. These data are important for understanding the results of experiments where TNFα-stimulation of RPTEC was intended to be inhibited with siRNA specific for IL-1β, or by addition of IL-1Ra (see below). Further studies are therefore needed to understand if endogenous, newly synthesized IL-1β is important for DNase I translocation (studies in progress).

To determine if newly synthesized DNase I was required for DNase I nuclear translocation, RPTEC were stimulated with TNFα and simultaneously transfected with DNase I-mRNA-specific or IL-1β-mRNA-specific siRNAs (Figure 3A). Furthermore, analyses were undertaken to determine if nuclear translocated DNase I is responsible for upregulation of the FAS gene. Thus, we investigated the links between TNFα, IL-1β, DNase I expression levels and its nuclear translocation and possible ability to enhance FasR expression levels.

Confocal microscopy analyses of DNase I (red) was performed on RPTEC after stimulation with TNFα and the effect of simultaneous transfection of the cells with DNase I mRNA-specific or IL-1β mRNA-specific siRNAs. Trap1 staining (green) was used as a cytoplasmic marker. Unstimulated RPTEC presented weak cytoplasmic DNase I expression but no nuclear DNase I staining (Figure 3A, first row). After stimulation of RPTEC with TNFα alone, the DNase I protein expression levels increased, DNase I translocated into the nuclei (Figure 3A, second row), and DNase I mRNA levels increased slightly (Figure 3C). After TNFα-stimulation and simultaneous transfection of the cells with DNase I mRNA-specific siRNA, DNase I mRNA expression levels were reduced to a level below that in unstimulated cells (Figure 3C), however, nuclear DNase I translocation was still weakly detectable (Figure 3A, third row). Importantly, transfection of these cells with DNase I siRNA did not affect the IL-1β mRNA levels (Figure 3C). Nuclear intensity measurement of DNase I in RPTEC cells transfected with DNase I and IL1β siRNA were represented (Figure 3B). This explains why DNase I siRNA does not interfere with endogenous IL-1β-mediated upregulation of FasR mRNA levels (Figure 3C). Thus, DNase I that translocates into the nucleus most likely derives from previously transcribed and stored DNase I protein, and its translocation is promoted by IL-1β.

Whether nuclear DNase I translocation is an IL-1 receptor-depending process was analyzed by addition of IL-1Ra to RPTEC stimulated with IL-1β. This resulted in strong reduction in IL-1β mRNA levels (Supplementary Figure 3C). However, IL-1Ra was not sufficient to eliminate the upregulation of IL-1β gene expression levels completely in response to IL-1β-stimulation in RPTEC (Supplementary Figures 3C,D) or to TNFα-stimulation (as demonstrated in Supplementary Figure 3D). This may probably explain why addition of IL-1Ra did not markedly affect mRNA levels of FasR in IL-1β- or in TNFα-stimulated cells (Supplementary Figure 3E).

RPTEC stimulated with TNFα and transfected with DNase I specific siRNA upregulated FasR mRNA significantly (Figure 3C) in contrast to the FasR protein expression level (Figure 4A) which remained unaffected in harmony with the reduced DNase I translocation in this experiment. Importantly, prevention of nuclear DNase I translocation by IL-1β siRNA in RPTEC also reduced TNFα mediated upregulation of FasR mRNA level (Figure 3C). In addition, FasR protein expression was reduced (Figures 4B,C).

Different concentration of TNFα stimulation upregulates IL-1β and FAS mRNA significantly in other cell lines (Supplementary Figure 4A). In addition, nuclear DNase I translocation and its association with IL-1β mRNA levels and protein expression were also observed in vivo as demonstrated by data in Supplementary Figure 4. IL-1β gene and protein expression are shown in young (<6 weeks) and older mice (20 weeks) (Supplementary Figures 4A,B, respectively) and demonstrate higher IL-1β levels in 20 weeks old mice compared to young mice (Supplementary Figures 4A,B, for IL-1β mRNA and IL-1β protein in situ). A low amount of DNase I nuclear staining was seen in animals with low levels of renal IL-1β expression (Supplementary Figure 4C), compared to animals with higher IL-1β expression levels in which tubular cell nuclei were extensively stained (Supplementary Figure 4C, 20 weeks).

As demonstrated above, IL-1β induced nuclear DNase I translocation in TNFα-stimulated RPTEC, accompanied by upregulation of FasR mRNA and protein expression levels. Since translocated nuclear DNase I in this experiment was enzymatically inactive and its appearance in the nuclei did not result in chromatin fragmentation, (Thiyagarajan et al., 2015 see below), we hypothesized that nuclear DNase I was involved in regulation of the FAS gene. To test this hypothesis ChIP analyses was performed (Figure 4D) by using non-specific IgG (neg.ctr), one anti-55 kDa DNase I (Santa Cruz (sc)), and one anti-DNase I antibody specific for translocated DNase I [52 kDa DNase I, Abcam (ac)]. To determine if DNase I was associated with FasR promoter sequences in chromatin fragments precipitated by the antibodies, FasR specific qPCR was performed (Figure 4D, upper part represent data from unstimulated RPTEC and lower part demonstrate data from TNFα-stimulated RPTEC). FasR-specific qPCR on chromatin from unstimulated RPTEC immunoprecipitated with control-IgG and ac-DNase I was negative (neg.ctr). Also the anti-DNase I specific antibody from Santa Cruz (sc) did not precipitate FasR sequences. In contrast, the anti-DNase I antibody specific for translocated DNase I precipitated FasR sequences in chromatin from TNFα-stimulated RPTEC. The qPCR on purely sheared chromatin demonstrated presence of FasR DNA sequences (Input sample) and as a positive control, an anti-histone H3 antibody precipitated chromatin fragments that yielded a positive FasR-specific qPCR (pos ctr).

Thus, combining the results demonstrated that DNase I translocates into the nuclei in response to IL-1β, probably binding to the promoter region of the FAS gene where it may exert a function as a transcription factor for FasR.

In a previous study, we demonstrated that TNFα-stimulation did not directly induce apoptosis in RPTEC (Thiyagarajan et al., 2015). In light of the results described above, we analyzed a link between DNase I translocation and upregulation of FasR in absence of detectable apoptosis. In-depth analyses of several apoptotic, parameters in stimulated cells were examined. After TNFα-stimulation, a significant upregulation of FasR (Figure 1F), caspase 3 and caspase 7 mRNA levels (Figure 5A) was shown. We did not detect activated caspase 3 levels upon TNFα-stimulation (Figure 5B). However, after TNFα stimulation in the presence of 125 ng of FasL, RPTEC showed an increased caspase 3/7 activity compared to the unstimulated and TNFα stimulated cells by Caspase GLO 3/7 assay (Figure 5C).

In order to ascertain that TNFα does not induce apoptosis in RPTEC, other relevant apoptotic parameters were analyzed (Figure 5D). Since RPTEC grow as adherent cells and become non-adherent by cell death, quantification of non-adherent cells is indicative of cell death. The numbers of non-adherent RPTEC were negligible and very similar in the unstimulated or TNFα-stimulated RPTEC cultures, thus arguing against an apoptotic cell death in response to TNFα.

In an analogous approach, the TUNEL assay was negative for unstimulated and TNFα-stimulated adherent cells (Figure 5D, upper panel), while inconclusive for non-adherent cells (Figure 5D, lower panel) because of no or marginal amounts of cells in the culture medium.

Data presented so far demonstrate that TNFα-stimulation alone upregulates the FasR protein expression levels, but did not render the cells TUNEL positive. Therefore, RPTEC demonstrate a relative resistance to TNFα-mediated apoptosis, consistent with previous observations for tubular cells (Boonstra et al., 1997; Jo et al., 2002; Kunter et al., 2005). Resistance to TNFα-induced apoptosis could also be explained by the fact that RPTEC do not constitutively express FasL, which was confirmed by gene expression analysis (Supplementary Figure 1B). In order to induce apoptosis in RPTEC, we added recombinant human FasL to the cells stimulated with TNFα. As demonstrated in Figure 5E, FasL did not bind to unstimulated RPTEC, while after TNFα-stimulation; FasL bound significantly to the cell membranes and was in confocus with FasR. To identify whether the interaction of FasL with FasR induced overt apoptosis, several apoptotic markers were analyzed after TNFα- and FasL-stimulation of RPTEC. In these cells, mRNA levels of caspases 3 and 7 were significantly upregulated in contrast to the anti-apoptotic Bcl2 mRNA that was significantly reduced in presence of FasL (Figure 5A). However, as observed for TNFα-stimulated RPTEC, we could not detect activated caspase 3 in these cells after treatment with exogenous FasL by Western blot (Figure 5B). In Figure 5F, it is demonstrated that unstimulated cells were both TUNEL and Annexin V negative. However, both adherent and non-adherent TNFα-stimulated cells to which FasL was added became TUNEL-positive and Annexin-positive (Figure 5F). In this experiment, we did not observe non-adherent cells with intact morphology, rather, Annexin V stained mostly microparticle-like (Beyer and Pisetsky, 2010; Dey-Hazra et al., 2010; Dinkla et al., 2013) or blebs-like (Cocca et al., 2002) structures in the culture supernatant (Figure 5F, lower panel).

To summarize, the data demonstrate that TNFα-stimulation upregulates FasR indirectly through TNFα-mediated upregulation of IL-1β in RPTEC. IL-1β promotes nuclear translocation of DNase I, which leads to the upregulation of FasR. This cascade makes the cells highly susceptible to FasL-mediated apoptosis.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.