Induction of a Long Noncoding RNA Transcript, NR_045064, Promotes Defense Gene Transcription and Facilitates Intestinal Epithelial Cell Responses against Cryptosporidium Infection

Epithelial cells along the mucosal surface provide the front line of defense against luminal pathogen infection in the gastrointestinal tract and are an important component of digestive mucosal immunity (1). These epithelial cells represent an integral component of a highly regulated communication network that can transmit essential signals to cells in the underlying gastrointestinal mucosa and that themselves, in turn, serve as targets of mucosal immune mediators (1). TLRs and NF-ĸB signaling are key to epithelial immune defense and have been implicated in secretion of antimicrobial peptides, release of cytokines and chemokines to mobilize immune effector cells, and activation of adaptive immunity (2, 3). Upon microbial challenge, gastrointestinal epithelial cells quickly initiate a series of innate immune reactions, including production of antimicrobial molecules and release of chemokines and cytokines (2, 3). These chemokines and cytokines then mobilize and activate immune effector cells to the infection sites (2).

Increasing evidence suggests that long noncoding RNAs (lncRNAs) function to regulate gene transcription through specific interactions with other cellular factors, including proteins, DNA, and other RNA molecules (4). Modes of lncRNA action may include regulation of gene expression in cis or in trans through recruitment of proteins or molecular complexes to specific loci, scaffolding of nuclear complexes, titration of RNA-binding proteins, and pairing with other RNAs to trigger posttranscriptional regulation (4, 5). Many lncRNAs, including long intergenic noncoding RNAs (lincRNAs), are induced in innate immune cells (6) and may play a role in the regulation of innate defense (7). For example, numerous lincRNAs are differentially regulated in virus-infected cells (8) and in dendritic cells or macrophages following stimulation by ligands for TLR4 and TLR3 (4). LincRNA-Cox2, one of the most highly induced lincRNAs in macrophages, has been shown to mediate both the activation and repression of distinct classes of immune genes (9). We have recently demonstrated that lincRNA-Cox2 regulates inflammatory gene transcription in gastrointestinal epithelial cells through modulating ATP-dependent chromatin remodeling (10, 11).

Cryptosporidium is a protozoan parasite that infects the gastrointestinal epithelium and other mucosal surfaces in humans and animals (12). This parasite is an important opportunistic pathogen in AIDS patients and a common cause of diarrhea in young children in developing countries (13). Despite its significant morbidity, mortality, and cost to society, there is currently no fully effective therapy available (14). The majority of human cryptosporidial infections are caused by two species: C. parvum and C. hominis (14). Cryptosporidium attaches to the apical membrane surface of epithelial cells and forms an intracellular but extracytoplasmic vacuole in which the organism remains (12). As such, epithelial cells play a central role in activating and orchestrating host immune responses (15). Indeed, the invasion of intestinal epithelial cells by C. parvum in vitro activates TLR4/NF-ĸB signaling, resulting in the production and secretion of various cytokines and chemokines, antimicrobial peptides (β-defensins and cathelicidins), and NO, which may kill C. parvum or inhibit parasite growth (16, 17).

In this study, we investigated the role of lncRNAs in intestinal epithelial cell antimicrobial defense using murine models of intestinal cryptosporidiosis. We observed significant alterations in lncRNA expression profiles in intestinal epithelial cells following C. parvum infection and identified a panel of lncRNA genes that are controlled by the NF-кB signaling pathway in host cells. One of such lncRNAs, NR_045064 (18), may play a role in regulating epithelial defense against C. parvum infection through facilitating the transactivation of several defense genes.

C. parvum oocysts of the Iowa strain were purchased from a commercial source (Bunch Grass Farm, Deary, ID). Murine intestinal epithelial cell line (IEC4.1) was a kind gift from Dr. P. Yang (McMaster University, Hamilton, Canada) and muINTEPI, a murine intestinal epithelial cell line (19), was purchased from InSCREENeX Cellular Screening Technologies (Germany). SC-514 (100 μM, Calbiochem), a potent IKK-2 inhibitor (20), and JSH-23 (10 μM, Santa Cruz Biotechnology, Dallas, TX), a cell-permeable, selective inhibitor of nuclear translocation of NF-кB p65 and its transcription activity (21, 22), were used to inhibit NF-кB activation. The phosphorothioate oligodeoxynucleotide (ODN) CpG-ODN 1668 (5′-TCCATGACGTTCCTGATGCT-3′) and LPS (from Escherichia coli strain K12) were purchased from InvivoGen (San Diego, CA) and IFN-γ and TNF-α were purchased from R&D Systems (Minneapolis, MN). SC-514, TNF-α, IFN-γ, and LPS at the used concentrations showed no cytotoxic effects on IEC4.1 and muINTEPI cells.

Intestinal epithelium and enteroids were isolated and cultured as previously described (23). Briefly, small intestines were opened longitudinally and washed with ice-cold Ca²⁺- and Mg²⁺-free PBS, then were cut into 1–2 mm fragments and washed with ice-cold Ca²⁺- and Mg²⁺-free PBS three times. The cut fragments were incubated in ice-cold 2 mM PBS/EDTA at 4°C for 30 min with gentle rotation followed by vigorous shake until the PBS solution was mostly opaque with dislodged crypt and villus particles. Large tissue fragments were removed through a 100-μm cell strainer (BD Biosciences, Franklin Lakes, NJ). The pass-through was centrifuged 150 × g for 5 min at 4°C, and the pellet was collected as the intestinal epithelium.

Models of intestinal cryptosporidiosis using intestinal epithelial cell lines and enteroids were employed as previously described (17, 23). Infection was done in culture medium (DMEM-F12 with 100 U/ml penicillin and 100 μg/ml streptomycin) containing viable C. parvum oocysts (oocysts with host cells in a 1:4 ratio). For in vivo infection, we adapted a murine model of intestinal cryptosporidiosis in neonatal mice (24). Briefly, mice at the age of 6 d after birth received C. parvum oocysts by oral gavage (10⁵ oocysts per mice). Mice receiving vehicle (PBS) by oral gavage were used as control. At 24 and 48 h after Cryptosporidium or vehicle administration, animals were sacrificed and ileum intestinal tissues were collected. At least five animals from each group were sacrificed, and ileum tissues were obtained for biochemical analysis. The C57BL/6J mice (The Jackson Laboratory) were used and approved by the Creighton University Biosafety and Institutional Animal Care and Use Committees. Infection was quantified by using real-time PCR, immunofluorescence microscopy, and immunohistochemistry, as previously reported (17, 23).

For quantitative analysis of mRNA and lncRNA expression, comparative real-time PCR was performed as previous reported (8–11, 17, 23) using the SYBR Green PCR Master Mix (Applied Biosystems, Carlsbad, CA). The sequences for all the primers are listed in Supplemental Table I.

Custom-designed RNA oligos against NR_045064 and a scrambled RNA were synthesized by QIAGEN (Germantown, MD). Sequences of small interfering RNAs (siRNAs) are: 5′-GAAAUGCUAACCGAGCUCAUU-3′ (sense) and 5′-UGAGCUCGGUUAGCAUUUCUU-3′ (antisense) for NR_045064 and nonspecific scrambled sequence 5′-UUCUCCGAACGUGUCACGUUU-3′ (sense) and 5′-ACGUGACACGUUCGGAGAAUU-3′ (antisense) for the control. The Wdr5 siRNA was purchased from Santa Cruz Biotechnology. siRNAs were transfected into IEC4.1 cells with Lipofectamine RNAiMAX (Invitrogen). For NR_045064 siRNA transfection into enteroids, the electroporation approach was used with the Neon transfection system (Thermo Fisher Scientific, Waltham, MA; electroporation parameter: 1600 v, 10 ms, pulse 3). The full length of NR_045064 was cloned into the pcDNA3.1(+) vector according to the manufacturer’s protocol. NR_045064-pCDNA 3.1(+) was transfected to cells with Lipofectamine 2000 (Invitrogen).

The −1974 to +140 region of the NR_045064 gene, which contains four putative NF-кB binding sites, was cloned into the multiple cloning site of the pGL3–Basic Luciferase vector (Thermo Fisher Scientific). The following primers were used to amplify the sequence: 5′-GGGTACCAGTCAGTTTTACTATG-3′ (forward, with the restriction site for KpnI) and 5′-CCAAGCTTCCACGCAGAAGG-3′ (reverse, with the restriction site for HindIII). The vectors with the mutants of the putative NF-кB binding sites using the GeneArt Site-Directed Mutagenesis PLUS kit (Thermo Fisher Scientific) and the empty vector were used as the control. The sequences for primers used for mutants are listed in Supplemental Table I. Cultured cells were transfected with the reporter construct overnight and then exposed to C. parvum infection for 8 h in the presence or absence of SC-514 or JSH-23 followed by assessment of luciferase activity. The luciferase activity was normalized to the control β-galactosidase level and compared with that of the pGL3 basic vector.

Whole cell extracts were isolated using the standard approach as previously reported (25). Briefly, cells were grown to 80% confluence and then exposed to C. parvum for various times. Cells were washed with PBS, and the whole cell protein was extracted in M-PER Mammalian Protein Extraction Reagent (Thermo Fisher Scientific) with proteinase inhibitor according to the product instruction. Protein concentration of each whole cell lysate was determined and subsequently analyzed by Western blot. The following Abs were used for blotting: anti-H3K4me1 (Abcam), anti-actin (Sigma-Aldrich), and anti-Wdr5 (Cell Signaling Technology).

Cells were lysed with the lysis buffer (20 mM Tris-HCl, pH 8, 150 mM NaCl, 1% NP-40, 20 μM MG-132, 1 mM PMSF, 10 μg/ml leupeptin, and 2 μg/ml pepstatin). A total of 1000 μg of lysate protein was incubated with the primary Abs at 4°C overnight to immunoprecipitate the protein complexes. Immune complexes were then collected by direct binding to Protein A–Sepharose. The immunoprecipitates were then blotted with the corresponding Ab against the protein as indicated. Anti-Mll3 (MilliporeSigma) and anti-Mll4 (MilliporeSigma) were used for immunoprecipitation (IP)/Co-IP analysis.

The formaldehyde crosslinking RNA IP (RIP) was performed as described (26). Briefly, lysates were precleaned with 20 μl of PBS-washed Magna ChIP Protein A+G Magnetic Beads (MilliporeSigma, MA). The precleaned lysate (250 μl) was then diluted with the whole cell extract buffer (250 μl), mixed with the specific Ab-coated beads, and incubated with rotation at 4°C for 4 h, followed by four times washing with the whole cell extract buffer containing protease and RNase inhibitors. The collected immunoprecipitated RNP complexes and input were digested in RNA PK Buffer pH 7 (100 mM NaCl, 10 mM TrisCl pH 7,1 mM EDTA, 0.5% SDS) with addition of 10 μg of proteinase K and incubated at 50°C for 45 min with end-to-end shaking at 400 rpm. Formaldehyde cross-links were reversed by incubation at 65°C with rotation for 4 h. RNA was extracted from these samples using Trizol according to the manufacturer’s protocol (Invitrogen) and treated with DNA-free DNase Treatment and Removal I Kit according to the manufacturer’s protocol (Ambion, Austin, TX). The presence of RNA was measured by quantitative, strand-specific RT-PCR using the iCycler iQ Real-Time PCR Detection System (BioRad). Gene-specific PCR primer pairs are listed in Supplemental Table I. Chromatin IP (ChIP) analysis was carried out as previously reported (10, 11). The following Abs were used for ChIP analysis: anti-p300 (Abcam), anti-H3k4me1 (Abcam), anti-H3k36me3 (Cell Signaling Technology), and anti-Wdr5 (Cell Signaling Technology). Chromatin isolation by RNA purification (ChIRP) analysis was performed as previously reported (27). Briefly, a pool of tiling oligonucleotide probes with affinity specific to NR_045064 was used and glutaraldehyde cross-linked for chromatin isolation. The sequences for each probe are listed in Supplemental Table I. The DNA sequences of the chromatin immunoprecipitates were confirmed by real-time PCR using the same primer sets covering the gene promoter regions of interest as for ChIP analysis. A pool of scrambled oligo probes and primers for LacZ were used as controls.

C57BL/6J mice (6 d old) (The Jackson Laboratory) were administrated with CpG-ODN 1668 (30 mg/kg body weight, oral) and intestine epithelium were collected at 4 and 24 h, as previously reported (28, 29). For LPS stimulation, C57BL/6J mice (4–6 wk old) (The Jackson Laboratory) were injected with LPS (15 mg/kg bodyweight, i.p.) as previously reported (30–32), and intestine epithelium were collected at 4 and 24 h postinjection. Tissue isolated from mice after PBS administration were used as the control.

All values are given as mean ± SEM. Means of groups were from at least three independent experiments and compared with Student t test (unpaired) or the ANOVA test when appropriate. The array data were analyzed and compared statistically by LC Sciences using the Agilent Feature Extraction Software, in accordance with the Minimum Information about a Microarray Experiment guidelines. The p values <0.05 were considered statistically significant.

To characterize the lncRNA expression profile in intestinal epithelial cells following Cryptosporidium infection, a genome-wide transcriptome analysis was performed using an in vitro model of intestinal cryptosporidiosis employing the IEC4.1 cell line, a transformed but nontumorigenic intestinal epithelial cell line from neonatal mice (5–7 d old). IEC4.1 cells were exposed to C. parvum infection for 24 h, and genome-wide RNA transcriptome analysis was carried out using the Agilent SurePrint G3 Mouse Gene Expression Microarray (G4852A), which provides full coverage of genes and transcripts with the most up-to-date content, including mRNAs and lncRNAs (http://www.chem.agilent.com/store/en_US/Prod-G4852A/G4852A). All array data were described in accordance with the Minimum Information about a Microarray Experiment guidelines and deposited at the Gene Expression Omnibus database (with the accession number GSE112247; https://www.ncbi.nlm.nih.gov/geo). A total of 1406 protein-coding genes were upregulated and 2536 protein-coding genes downregulated in cells following infection (the fold changes >1.25 combined with a p < 0.05 as the threshold) (Supplemental Table II and database GSE112247). A total of 1212 lncRNAs were upregulated and 627 lncRNAs downregulated in the infected cells (Supplemental Table III and database GSE112247). Top 20 upregulated and downregulated lncRNA candidates are shown in heatmap in Fig. 1A. C. parvum–induced upregulation of selected lncRNA genes was further confirmed by quantitative real-time PCR in IEC4.1 cells and muINTEPI cells following infection for 24 h (Fig. 1B). Consistent with previous studies (17, 33, 34), protein-coding genes, such as Cxcl2, Nos2, and Csf2, were upregulated in host cells following C. parvum infection (Supplemental Fig. 1A, Supplemental Table II and database GSE112247). Using an ex vivo infection model employing enteroids from neonatal mice (23), we detected increased expression levels of selected lncRNAs in enteroids following C. parvum infection (Fig. 1B). An increase in their expression levels was also observed in the intestinal epithelium in infected neonatal mice at 72 h after oral administration of the parasite (Fig. 1B). The top five most upregulated lncRNAs in the intestinal epithelium following infection in vivo are NR_045064, lncRNA-Chr5:5405, lncRNA-Chr14:1060, NR_045311, and lincRNA-Chr9:4130 (Fig. 1B).

Differential alterations in the lncRNA expression profile of infected host cells suggest that lncRNA gene expression is finely controlled in intestinal epithelial cells in response to C. parvum infection. In our previous studies, we demonstrated that C. parvum infection activates the NF-кB pathway in gastrointestinal epithelial cells through microbial recognition of TLR4 and TLR2 (35, 36). Therefore, we asked whether activation of the NF-кB pathway is involved in the transcription of select lncRNAs upregulated by C. parvum. We exposed IEC4.1 cells to C. parvum infection in the presence of SC-514, an IKK2 inhibitor that prevents p65-associated transcriptional activation of the NF-кB pathway (20). SC-514 blocked C. parvum–induced upregulation of many lncRNAs (Fig. 2A). One of such lncRNAs is NR_045064, a noncoding transcript of 754 nt from chromosome 15 (Fig. 2B) (18). Expression of this NR_045064 was detected in many of the tissues and organs obtained from mice, such as the brain, heart, and lungs (Supplemental Fig. 1B). NF-кB–dependent upregulation of NR_045064, as well as several other selected lncRNAs induced by C. parvum infection in IEC4.1 cells, was further confirmed in muINTEPI cells following infection in the presence or absence of SC-514 or JSH-23 (Fig. 2C). Induction of NR_045064 by C. parvum infection was detected in IEC4.1 cells as early as 8 h following infection, persisting for up to 48 h in the infected cells (Fig. 2D). Transactivation of the NR_045064 gene was also detected in IEC4.1 and muINTEPI cells following stimulation by LPS or TNF-α but not by IFN-γ (Fig. 2E). Based on TFSEARCH (http://www.cbrc.jp/research/db/TFSEARCH.html) and MOTIF (http://motif.genome.jp/) database searches (37, 38), putative NF-кB binding sites were identified within the potential promoter region of the NR_045064 gene (Fig. 2F). We then cloned the potential promoter region of the NR_045064 gene and inserted the sequence into the luciferase reporter vector (Fig. 2F). C. parvum infection increased the luciferase activity in cells transfected with the luciferase construct that encompassed the promoter region of NR_045064 gene, but not in cells transfected with the empty vector control (Fig. 2F). Moreover, increased luciferase activity associated with the promoter region of the NR_045064 gene induced by C. parvum infection was blocked with the NF-кB inhibitors or by mutation of the putative NF-кB binding sites (Fig. 2F).

The NF-кB signaling pathway has been demonstrated to be essential to epithelial anti–C. parvum defense (12, 15, 36). Given the fact that NR_045064 is the top induced lincRNA in intestinal epithelium following C. parvum infection in vivo, coupled with its NF-кB–dependent transcription feature, we asked whether transcription of this gene is involved in epithelial anti–C. parvum defense. To test this possibility, we took the RNA interference approach to knockdown NR_045064 in host cells and then measured its impact on C. parvum infection burden following exposure of them to the parasites for various periods of time. The designed siRNA to NR_045064 could significantly knockdown the expression level of NR_045064 in the noninfected and C. parvum–infected IEC4.1 cells (Fig. 3A). Intriguingly, knockdown of NR_045064 but not the nonspecific siRNA controls caused a significant increase in the infection burden in IEC 4.1 cells (Fig. 3B). Complementarily, overexpression of the 754 nt of NR_045064 in IEC4.1 cells resulted in a decrease of infection burden (Fig. 3C). Moreover, siRNAs to knockdown NR_045064 also increased the burden of C. parvum infection in enteroids isolated from neonatal mice (Fig. 3D). Taken together, the above data suggest that induction of NR_045064 may promote epithelial antimicrobial defense against C. parvum infection.

Given the significant impact of NR_045064 induction on the parasite infection burden, we tested whether NR_045064 induction can affect the transcription of epithelial cell defense genes in intestinal epithelial cells following C. parvum infection. Previous studies have demonstrated a set of genes critical to epithelial anti–C. parvum defense, such as Nos2, Csf2, Cxcl2, Ido1, lipocalin 2 (Lcn2), and regenerating islet-derived 3 γ (Reg3g) (1–3). We then measured the expression levels of these genes in cells either transfected the siRNAs to NR_045064 or overexpressed with the NR_045064 vector following C. parvum infection. Upregulation of expression levels for Csf2, Nos2, and Cxcl2 genes was detected in IEC4.1 cells following C. parvum infection (Fig. 4A). Knockdown of NR_045064, but not the nonspecific siRNA controls, attenuated the induction of Csf2, Nos2, and Cxcl2 in cells induced by C. parvum infection (Fig. 4A). Complementarily, overexpression of NR_045064 enhanced C. parvum–induced expression of Csf2, Nos2, and Cxcl2 in IEC4.1 cells (Fig. 4B). Neither siRNA to NR_045064 nor forced expression of NR_045064 resulted in significant changes in the expression levels of Lcn2, Ido1, and Reg3g in the noninfected or C. parvum–infected IEC4.1 cells (Fig. 4A, 4B). The above data suggest that NR_045064 induction is involved in the transcriptional regulation of selected defense genes in intestinal epithelial cells following C. parvum infection.

To explore the underlying mechanisms of NR_045064-mediated gene transactivation, we asked whether NR_045064 induction can impact the transcriptional active histone modifications associated with Nos2 and Csf2 genes induced by C. parvum infection. Histone modifications, such as H3K4 and H3K36 methylations, are generally associated with gene transactivation (39). Using ChIP analysis with designed PCR primers (Supplemental Table I) covering the various regions of regulatory promoters of the genes (Fig. 5A), we detected a significant increase in H3K4me1 in the gene loci of Nos2 and Csf2 in infected IEC4.1 cells (Fig. 5A). Indeed, an increased content of H3K4me1 at the protein level was measured using Western blot in cells following infection at 24 h (Supplemental Fig. 1C). Accordingly, enrichment of H3K4me1 at the gene loci of Nos2 and Csf2 induced by C. parvum were attenuated in cells transfected with the siRNA to NR_045064 (Fig. 5A). Nevertheless, no detectable increase in H3K36me3 was detected in the Nos2 and Csf2 gene loci in infected IEC4.1 cells (Fig. 5B).

We further asked the question of how transcription of Nos2 and Csf2 genes is selectively regulated by NR_045064. Based on TFSEARCH (http://www.cbrc.jp/research/db/TFSEARCH.html) and MOTIF (http://motif.genome.jp/) database searches (37, 38), putative p300 binding sites were identified within the potential promoter regions of the Nos2 and Csf2 gene loci. To define whether NR_045064 is physiologically associated with p300, we performed the RIP assay. A significant amount of NR_045064 was detected in the immunoprecipitates from both the noninfected and infected IEC4.1 cells using anti-p300 (Fig. 6A), suggesting a physiological association between p300 and NR_045064. Nevertheless, no obvious increase of NR_045064 to the p300 complex in the infected cells, probably because whole cell extracts were used for the RIP assay and the anti-p300 was not specific for activated p300. We then asked whether they could coordinately be recruited to the Nos2 and Csf2 gene loci. Indeed, an increased occupancy of p300 to the promotor regions of Nos2 and Csf2 gene loci was detected using the ChIP analysis (Fig. 6B). Knockdown NR_045064 with the siRNAs partially blocked the occupancy of p300 to the promotor regions of the gene loci (Fig. 6B). To define whether NR_045064 is also recruited to the Nos2 and Csf2 gene loci to facilitate their transcription in infected cells, we performed ChIRP analysis to measure the occupancy of NR_045064 to the genes in cells following infection. A pool of biotinylated tiling ODN probes with affinity to the NR_045064 sequence was used to precipitate the chromatin fragments through glutaraldehyde crosslinking and chromatin isolation. The DNA sequences of the precipitated chromatin fragments were identified by PCR using primers specific to various regions of the DNA sequences of the target gene loci (27). We detected an increased occupancy of NR_045064 to the Nos2 and Csf2 gene loci in IEC4.1 cells after exposure to C. parvum for 24 h (Fig. 6C).

The p300 transcription factor is not generally accepted as an RNA-binding protein (RBP) (40). To clarify which RNA-binding elements in the p300/NR_045064 complex may mediate the assembly of NR_045064, we tested the potential involvement of Wdr5, an RBP and a key element of the Mll complex that has been implicated in p300-mediated histone methylation and gene transcription (41). We first performed the RIP assay using anti-Wdr5 in cells following C. parvum infection. A significant amount of NR_045064, but not for the control Rnu2-1 RNA, was detected in the immunoprecipitates from infected cells (Fig. 6D). Assembly of NR_045064 to the Wdr5 complex was not detected in the noninfected control (Fig. 6D). We then performed Co-IP analysis for Wdr5 and the Mll complex in cells following C. parvum infection. An increased association between Wdr5 and the Mll3/Mll4 was detected in infected IEC4.1 cells (Fig. 6E). Knockdown of NR_045064 with the siRNAs significantly blocked the physical association between Wdr5 and Mll3/Mll4 complex (Fig. 6E). Accordingly, an increased recruitment of Mll/Wdr5 complex was detected in the Nos2 and Csf2 gene loci in IEC4.1 cells after exposure to C. parvum for 24 h (Fig. 6F). Notably, both Wdr5 and NR_054064 were recruited to the set 2 promoter region of Csf2 and set 1/2/3 promoter regions of Nos2. However, recruitment of NR_054064, but not Wdr5, was detected in the set 3/5 regions of Csf2 and set 7 region of Nos2. Recruitment of Wdr5, but not NR_054064, was detected in the set 1 region of Csf2 and set 4 region of Nos2 (Fig. 6C, 6F). Interestingly, knockdown of NR_045064 with the siRNAs significantly blocked the recruitment of Wdr5 to all the set regions of Nos2 and Csf2 gene loci in infected cells, even for the set 1 region of Csf2 and set 4 region of Nos2 where direct recruitment of NR_045064 was not detected (Fig. 6C, 6F). Complementarily, siRNA knockdown of Wdr5 also significantly inhibited C. parvum–induced transcription of Nos2 and Csf2 genes in IEC4.1 cells (Fig. 6G). Taken together, the above data indicate that WDR5/MLL-p300–associated chromatin remodeling is involved in NR_045064-mediated transcription of associated defense genes in intestinal epithelial cells following C. parvum infection.

To determine the relevance of our findings of C. parvum–responsive NR_045064-mediated intestinal defense gene transcription in the intestine in general, we tested intestinal epithelial response to ODN and LPS stimulation. Given the activation of TLR9/NF-кB signaling in intestinal epithelium by CpG-ODNs in vitro and in vivo, this model has been used to stimulate NF-кB–mediated innate defense in the gastrointestinal tract (28, 29). Transactivation of the NR_045064 gene was detected in IEC4.1 and muINTEPI cells following stimulation by CpG-ODN 1668 (Fig. 7A). Mice were administrated by oral with CpG-ODN 1668 (30 mg/kg body weight), and intestinal epithelium was isolated. Expression levels of NR_045064 and inflammatory genes were measured. A significant increase in NR_045064 levels was found in the small intestinal epithelium (Fig. 7B). Upregulation of Nos2 and Cxcl2 was also detected in the small intestinal epithelium from mice after ODN injection (Fig. 7B). Similarly, using a well-established mouse septic model induced by LPS to the abdominal cavity (30–32), we detected a significant increase in the expression levels of NR_045064, Nos2, and Cxcl2 in the small intestinal epithelium from mice after LPS injection (Fig. 7C).

Our genome-wide analysis of the transcriptomic profile revealed distinct gene expression patterns for both protein-coding genes and lncRNA genes in intestinal epithelial cells following C. parvum infection. Similar to the protein-coding genes, transcription of a panel of lncRNA genes appears to be controlled by NF-кB signaling. One of such lncRNAs, NR_045064, is induced at an early phase in host cells following C. parvum infection, and functionally, induction of NR_045064 is involved in the regulation of epithelial antimicrobial defense responses through facilitating defense gene transcription. Therefore, lncRNAs may represent a new arm of regulation in defense mechanisms in intestinal epithelial cells against C. parvum infection.

Various models of transcriptional control of lncRNA genes have recently been proposed based upon their genomic locations (5). Some may have their own regulatory elements (e.g., promoters and enhancers) and thus are regulated independently (42–44). Others may share the regulatory elements with nearby genes and be coregulated along with other genes (45, 46). Regulation of the NR_045064 gene at the transcriptional level by NF-κB involves binding of the NF-κB subunits to the putative promoter region of the NR_045064 gene. Putative binding sites exist at the promoter region of the gene locus. Moreover, luciferase reporter assay associated with the transcriptional activity induced by NF-κB activation was detected in intestinal epithelial cells following C. parvum infection. Thus, NF-κB–mediated transcription of lncRNA genes may be an important part of the transcriptional responses in intestinal epithelial cells to C. parvum infection.

One of the major findings of this study is that NR_045064 is required for efficient epithelial defense against Cryptosporidium infection. Loss-of- or gain-of-function of NR_045064 can significantly impact the parasite burden in intestinal epithelial cells. Previous studies demonstrated that knockdown a single lncRNA could alter the expression profiles of many genes (9–11, 47–49). For example, knockdown of lincRNA-Cox2, a TLR/NF-κB–responsive lincRNA in murine macrophages (9), resulted in a global inhibition of transcription of the late primary inflammatory genes in cells in response to LPS stimulation (10). Given the essential role of a limited defense gene for intestinal epithelial cells against C. parvum infection (15), we focused our analysis of NR_045064 influence on defense gene transcription. Our data support that induction of NR_045064 is involved in the transcriptional regulation of selected defense genes in cells following C. parvum infection. This impact appears to be gene specific, as the upregulation of the Nos2 and Csf2 genes, not the other tested defense genes such as Lcn2 and Ido1, was attenuated in infected cells treated with the siRNA to NR_045064. The recruitment of NR_045064 to the Cfs2 and Nos2 gene loci in cells following infection further support that this function of NR_045064 on Nos2 and Csf2 expression is at the transcription level.

Our data support that the transcriptional regulation of Nos2 and Csf2 genes by NR_045064 in cells in response to C. parvum infection involves p300/WDR5/MLL complex–mediated epigenetic gene transcription. The SET1/COMPASS family of histone H3K4 methylases are essential regulators of transcription and key mediators of normal development and disease (50, 51). Key subunits of this family proteins include MLL1, MLL3, MLL4, SET1A, SET1B, WDR5, RbBP5, ASH2L, and DPY30 (50). Of these subunits, MLL3 and MLL4 possess histone methylation activity and are involved in transcriptional coactivation of many genes (50–53). Functionally, the WDR5/MLL3/4–containing complex has been associated with p300-mediated gene transcription through modulation of histone modifications (41). First, we identified that several of the key subunits of the complex were recruited to the Nos2 and Csf2 gene loci in cells following infection, including p300 and Wdr5. Importantly, NR_045064 was also recruited to the similar regions of the gene loci in infected cells. Second, the WDR5/MLL–containing complex-associated H3K4 methylation was also enriched at the corresponding regions of the gene loci in cells following infection. Knockdown of Wdr5 attenuated the enrichment of above histone modifications at the Nos2 and Csf2 gene loci. Finally, NR_045064 is involved in the recruitment of the complex to the gene loci, as NR_045064 was assembled into the Wdr5-containing complex, and knockdown of NR_045064 resulted in a decreased recruitment of p300 and Wdr5 and associated H3K4 methylation at the Nos2 and Csf2 gene loci in cells following infection.

Notably, both Wdr5 and NR_054064 were recruited to the set 2 promoter region of Csf2 and set 1/2/3 promoter regions of Nos2. However, recruitment of NR_054064, but not Wdr5, was detected in the set 3/5 regions of Csf2 and set 7 region of Nos2, suggesting other Wdr5-indenepdent mechanisms for NR_054064 recruitment to different regions. Intriguingly, knockdown of NR_045064 could block the recruitment of Wdr5 to the set 1 region of Csf2 and set 4 region of Nos2 where recruitment of NR_045064 was not detected. Underlying mechanisms merit further investigation.

NR_045064 may function as a scaffold molecule through its interactions with the RNA-binding proteins within the Wdr5/Mll complex. Wdr5 is a known RNA-binding protein, and the Wdr5/Mll complex is involved in p300-mediated gene transcription (41). Whether NR_045064 can directly interact with Wdr5 requires further characterization using in vitro RNA pull-down assay. However, knockdown of NR_045064 resulted in dissociation between Wdr5 and Mll3 or Mll4 as revealed by our Co-IP analysis. This implicates to us that NR_045064 may function as a scaffold to assemble various key elements for the Wdr5/Mll complex. Alternatively, some lncRNAs have been demonstrated to interact with DNA molecules to form a triple-helical structure (54, 55). As such, NR_045064 may “guide” the initial recruitment of the Wdr5/Mll complex to the gene loci through direct binding to a specific DNA motif. We favor the speculation that NR_045064 may function as a scaffold molecule for the assembly of various p300/Wdr5/Mll components to the Nos2 and Csf2 gene loci in cells following Cryptosporidium infection.

NF-κB–mediated NR_045064 expression and its subsequent impact on transcription of defense genes may also be involved in intestinal inflammation and infection by other pathogens. The induction of NR_045064 in intestinal epithelium following stimulation with ODN or LPS, along with the expression of several inflammatory and defense genes, suggests that NF-κB–mediated NR_045064 expression and its subsequent impact on transcription of inflammatory and defense genes may be an important component of NF-κB–mediated responses in intestinal epithelium in general. Future studies should investigate the regulation of other genes by NR_045064 and potential significance in inflammatory process. Comprehending the mechanisms of how lncRNAs may modulate dynamic gene transcription through chromatin remodeling could advance our knowledge of the molecular mechanisms of defense responses in many inflammatory and infectious diseases, such as intestinal cryptosporidiosis.

We thank Barbara L. Bittner for assistance in writing the manuscript. The IEC4.1 cells were a gift from Dr. Ping-Chang Yang (McMaster University, Hamilton, Canada).

The authors have no financial conflicts of interest.