CAWS administration increases the expression of interferon γ and complement factors that lead to severe vasculitis in DBA/2 mice

To induce arteritis in B6 (control) and DBA/2 mice, we performed intraperitoneal injection (i.p.) of CAWS (0 or 1 mg/mouse) for 5 consecutive days as reported previously[11]; we will call this process as CAWS administration, hereafter, and set this timing (right after the 5^th injection of CAWS) as 0 week (0 w). Although the number of surviving mice at 2 w, 4 w, 8 w, and 9 w following CAWS administration was remarkably reduced in DBA/2 mice (Figure 1A), similar to our previous report[11], few B6 mice died during this period (data not shown), which is also similar to our previous report[11]. At 2 w, 4 w, 8 w, and 9 w after CAWS administration, the surviving mice were sacrificed and mRNA isolated from PBMCs. Three mice were sacrificed at each time point (the number of sacrificed mice is included in Figure 1B). Tissue sections that were obtained from paraffin-embedded hearts fixed with 10% neutral formalin and stained with hematoxylin and eosin (H&E) revealed vasculitis in the aorta and coronary artery of DBA/2 mice, with fibrotic thickening of the arterial media and adventitia and infiltration of inflammatory cells (Figure 1C). In tissue sections obtained from the hearts of mice at 2 w, 4 w, 8 w, and 9 w following CAWS administration, we observed infiltration of inflammatory cells in the aortic media and adventitia of DBA/2 mice, a histological finding typical of aortitis, as early as 2 w following CAWS administration, and this was observed up to the final 9 w time point (right panels in Additional file1: Figure S1). B6 mice also showed infiltration of inflammatory cells in the aortic media and adventitia, although the onset of aortitis in B6 mice occurred later than in DBA/2 mice, not appearing until 4–6 w following CAWS administration (left panels in Additional file1: Figure S1). Since these results were similar to those of our previous report[10, 11], we concluded that the mice used for DNA microarray analysis in this study responded to CAWS in a similar manner to those described in our previous report.

To identify genes whose expression levels were up- or downregulated in the PBMCs of DBA/2 mice, but not in B6 mice, following CAWS treatment, we examined the transcriptome profiles of mRNA prepared from DBA/2 and B6 PBMCs at 2 w, 4 w, 8 w, or 9 w following CAWS administration using Agilent’s Whole Mouse Genome Microarray (G4122F). When the relative intensities of the signals at 2 w, 4 w, 8 w, or 9 w were compared with those at 0 w following CAWS administration, the genome-wide expression profiles were found to be distinct between these two mouse strains (Additional file1: Figure S2A). Heat map representation of the signal intensities revealed that a number of genes were dramatically up- or downregulated in DBA/2 mice compared to B6 mice (Additional file1: Figure S2B).

The most notable genes that showed enhanced expression following CAWS treatment in DBA/2 mice belong to the “complement system”, in particular the genes encoding C3, C4, complement factor b (Cfb or BF), complement factor h (Cfh or HF1), and ficolin-A (Fcna) (indicated by turquoise arrows in Figure 2A and Additional file1: Figure S2B). The complement 3a receptor 1 (C3ar1) gene showed modestly enhanced expression at 2 w, 4 w, and 8 w following CAWS administration and conspicuously enhanced expression at 9 w (turquoise arrow in Figure 2A). C5ar1 gene also showed enhanced expression at 4 w and 8 w, but B6 mice also showed enhanced expression at this timing, albeit less dramatically. Expression of complement C9 (C9), Serping1 (Serpin peptidase inhibitor, clade G1), complement factor i (Cfi), and hemolytic complement (HC) was extremely low in B6 mice at 2 w, 4 w and 9 w, but their expression was detectable in DBA/2 mice, displaying a comparatively enhanced ratio in DBA/2 mice compared with B6 mice. By contrast, complement C7 (C7), decay accelerating factor (DAF or CD59), and complement receptor 2 (Cr2) showed decreased expression following CAWS treatment in DBA/2 mice, although these genes were weakly expressed in B6 (Figure 2A).

To determine whether changes in the expression levels of these genes is physiologically significant, a scatter plot was used to analyze the distribution of the expression levels of many genes at a glance. Indeed, C3, C4, Cfb, Cfh, Fcna, and Cr2 showed expression levels high enough to conduct physiologically significant comparison (>10 raw signal intensity) in DBA/2 mice at 2 w or 4 w following CAWS administration (Figure 2B). By contrast, complement genes that showed very low expression levels compared to above described activated complement genes in both B6 and DBA/2 mice were not deemed significant (Figure 2B). C3ar1 mRNA levels were low at 2 w or 4 w and its expression level was enhanced only at 9 w following CAWS administration. Thus, we conclude that the enhanced mRNA levels of C3, C4, Cfb, Cfh, and Fcna observed after CAWS administration in DBA/2 mice are significant.

We next performed an ontology search using Ingenuity Pathways Analysis (Ingenuity Systems, http://​www.​ingenuity.​com). This powerful query system facilitates the rapid visualization of relationships between sets of relevant genes to predict the putative physiological significance of these genes. We found that alternative complement pathway genes, including C3, Cfb (=BF), and Cfh (=HF1), appear to play major roles in activation of the complement system (Figure 3). By contrast, classical pathway genes, including complement C1 (C1s and C1r) and Serping G1 (but not C4), and lectin pathway genes including mannose-binding lectin (Mbl) appear to play less significant roles because they showed low levels of expression even in DBA/2 mice (Figure 3). The low mRNA level of C7 measured in DBA/2 mice suggests that downstream regulation of the complement system is not important in the induction of vasculitis in the aortas or coronary arteries of DBA/2 mice.

To confirm that the microarray signals accurately reflected mRNA levels, expression profiling using quantitative reverse transcription-polymerase chain reaction (qRT-PCR) analysis was performed for C3, C4, Cfb, Cfh, and FcnA genes using the same RNA samples used for the microarray analysis. Indeed, all of the genes analyzed by qRT-PCR showed a similar pattern of mRNA distribution following CAWS administration; namely, they showed increased mRNA levels of the denoted genes in DBA/2 mice compared with B6 mice (Figure 4), confirming the accuracy of the microarray analysis data.

We next examined the expression of interferon-related genes because expression of the Ifng gene was enhanced following CAWS administration in DBA/2 but not in B6 mice, as revealed by both DNA microarray and qRT-PCR analyses (Figure 4F and Figure 5A). Notably, Ifng was the only interferon-related gene that exhibited upregulated DBA/2-specific expression following CAWS administration (Figure 5A): moreover, mRNA expression of Ifng was more conspicuously upregulated than that of other interferon gene subtypes in DBA/2 mice (Additional file1: Figure S3). In the “interferon signaling pathway”, interferon regulatory factor-1 (Irf1) also showed increased expression, although less significantly than interferon itself, in DBA/2 mice at 4 w and 8 w (black arrow in Figure 5A). Notably, the mRNA level of Irf9 was decreased at 2 w, and this depressed level persisted throughout the experiment (green arrow in Figure 5A). This may explain why many of the interferon-stimulated genes were not activated (Figure 5A); the Irf9/Isgf3g complex binds to the interferon-stimulated response element to activate the transcription of interferon-stimulated genes[23].

In the pathway identified as the “Role of Prk in interferon induction and antiviral response”, mRNA levels of eukaryotic translation factor 2 alpha kinase 4 (Eif2aka) and Irf1 were enhanced (Figure 5B). Eif2aka is a heterotrimeric GTPase that controls the selection of correct start codons on mRNA, and thus regulates translation initiation by facilitating the preferential translation of selected transcripts in response to cellular stresses, in this case CAWS administration[24]. Our results suggest that DBA/2 mice are more sensitive to this type of stress than B6 mice.

In the “acute phase response signaling” pathway (Figure 5C), three genes other than C3 and Cfb were upregulated; CCAAT/enhancer-binding protein beta (Cebpb or C/EBPβ), IL-5, and lipopolysaccharide-binding protein (Lbp). Cebpb, a member of the C/EBP family of basic region-leucine zipper (bZIP) proteins, is largely expressed in macrophages and is important for the antibacterial activity of macrophages through binding to the regulatory regions of several acute phase and cytokine genes[25]. Enhanced expression of Cebpb following CAWS administration suggests that downstream target genes of Cebpb, including IL-4 (see below), may be widely activated. IL-5 exerts pleiotropic effects on various target cells, including B cells, eosinophils, and basophils, and overexpression of IL-5 in vivo significantly increases the number of eosinophils and B cells[26]. Lbp is involved in the acute phase immunologic response to Gram-negative bacterial infections through binding to bacterial lipopolysaccharide (LPS), thereby eliciting immune responses by presenting the LPS to the cell surface pattern recognition receptors CD14 and Tlr4[27]. In this case, CAWS, rather than LPS, may have been recognized as an infectant.

In the “acute phase myeloid leukemia signaling” pathway (Figure 5D), two genes were upregulated: spleen focus-forming virus proviral integration 1 (Sfpi1 or PU.1) and colony stimulating factor 1 receptor (Csf1r). Sfpi1 is a transcription factor that plays a key role in hematopoietic cell fate decisions through mutual negative regulation that occurs between Sfpi1and GATA-1, a zinc finger transcription factor that is the central mediator of erythroid gene expression[28]. Csf1r is a single pass type I membrane protein that acts as a receptor tyrosine kinase for a cytokine called colony stimulating factor 1, which is a primary growth factor and monocyte/macrophage chemokine that regulates survival, proliferation and differentiation of macrophages and other mononuclear phagocytic lineage cells[29]. Enhanced expression of Csfr1 suggests the activation of macrophages following CAWS administration.

A scatter plot of Irf1, Irf9, C3, Cebpb, Csf1r, Trf, Ifng, Eif2aka, and Cfb expression showed expression levels high enough to conduct physiologically significant comparisons (>10 raw signal intensity) in DBA/2 mice at 2 w or 4 w following CAWS administration (Figure 2B), suggesting that their enhanced mRNA levels indicated by the heatmap (Figure 5E) are significant. By contrast, mRNA levels of Lbp and Sfpi1 may be too low for comparison between B6 and DBA/2 mice to be significantly different.

Some transcription factor genes also showed upregulated expression only in DBA/2 mice with mRNA levels peaking at 4 w (highlighted by black arrows in Figure 6), 8 w (turquoise arrows), 9 w (red arrows), and 2 w/9 w (green arrows). Irf1, Irf7, and Cebpb were discussed above. Stat1, a signal transducer and activator of transcription, plays a crucial role in the pathology of atherosclerosis resulting from the effects of interferon-γ produced by T cells. STAT1 also induces increased expression of several proinflammatory and pro-atherogenic mediators of signaling mediated by TLR4[30]. Expression of Nupr1, a highly basic and loosely folded multifunctional protein, is induced in response to several cellular stresses[31] and provides protection against metabolic stress-mediated autophagic-associated cell death resulting from activation of the Nupr1-aurora kinase A pathway[32]. T-box 20 (Tbx20), a transcriptional regulator required for cardiac development, plays a dual role as both a transcriptional activator and a repressor, promoting or repressing distinct genetic programs within adult heart in response to cellular stress[33]. Increased expression with twin peaks observed at 2 w and 9 w suggests that CAWS injection altered regulation of Tbx20, thereby generating cardiac phenotypes (Figure 1C).

Zinc finger protein 384 (Zfp384), also called Nmp4 (nuclear matrix protein 4), is a nucleo-cytoplasmic shuttling transcription factor that regulates the expression of collagen and matrix metalloproteinases and represses bone formation induced by parathyroid hormone[34]. Ankyrin repeat and SOCS box-containing protein 2 (Asb2) interacts with mixed lineage leukemia (MLL) protein, a key epigenetic regulator of normal hematopoietic development, and overexpression of Asb2 degrades MLL to reduce its transactivation activity[35]. Zinc finger protein 3612 (Zfp3612), a member of the tristetraprolin family of tandem CCCH finger proteins that bind to AU-rich elements in the 3’-untranslated region of mRNAs, is a critical modulator of definitive hematopoiesis[36]. Zinc finger protein 819 (Zfp819), a member of the C2H2-zinc finger (C2H2-Znf) family of proteins, appears to function as a transcriptional and cell cycle/apoptosis regulator and play a role in the establishment and maintenance of pluripotency. Although these proteins play roles in several important cellular events such as hematopoiesis, these functions do not appear to be relevant to the pathogenesis of CAWS-induced vasculitis.

Scatter plot analysis indicated that mRNA levels of Irf1, Cebpb, Stat1, Nurp1, Irf7, and Brd8 were high enough to conduct physiologically significant comparison between B6 and DBA/2 mice according to their heatmaps (Figure 6B) at both 2 w and 4 w following CAWS administration (Figure 6C), suggesting that their superficially augmented expression is physiologically significant. By contrast, mRNA levels of Asb2, Zfp819, Zfp384, Zfp37, Zfp3612, and Tbx20 were too low to conclude that heatmap comparison of these genes between B6 and DBA/2 mice is physiologically significant.

Finally, CAWS administration resulted in a slight, but comparable increase in Dectin-2 mRNA levels in both B6 and DBA/2 mice (Additional file1: Figure S7A). Moreover, scatter plot analysis indicated that the mRNA levels of Dectin-2 mRNA were lower than those of Dectin-1, a similar C-type lectin molecule (Additional file1: Figure S7B).

Total RNA of leukocytes from whole blood of B6 and DBA/2 mice was stabilized by RNAlater™ (Ambion) and then isolated using LeukoLOCK™, a total RNA Isolation System (Ambion, Austin, TX, USA), according to the manufacturer’s instructions. The integrity of the total RNAs used for the microarray analysis was confirmed using the RNA 6000 Nano LabChip Kit (p/n 5067–1511) on the Agilent 2100 Bioanalyzer (G2938C; Agilent Technologies Inc., Palo Alto, CA), and only samples with an RNA integrity number (RIN) above 8.8 were used for gene expression profiling.

Microarray analyses were performed as single-color hybridizations. Total RNAs obtained from mouse PBMCs were independently reverse- transcribed using oligo-dT primers containing the T7 RNA polymerase promoter sequence to generate cDNAs and MMLV-RTase. These were then subjected to in vitro transcription using T7 RNA polymerase to label the cRNAs with Cy3-CTP (Amersham Pharmacia Biotech, Piscataway, NJ) using a Fluorescent Linear Amplification Kit (Agilent Technologies). Purified Cy3-labeled cRNAs (825 ng) from individual B6 or DBA/2 mice were then hybridized on the microarrays. Washing, scanning, and gene analysis with Agilent’s Whole Mouse Genome Microarray (4×44K; G4122F), on which 31,226 known genes were spotted, were conducted according to the manufacturer’s protocol (Agilent Technologies). Agilent Feature Extraction software (v. 9.5.1) was used to assess spot quality and extract feature intensity statistics. The Subio Platform and Subio Basic Plug-in (v1.11; Subio Inc., Aichi, Japan) were then used to calculate the between-sample fold change. When a large number of genes were automatically classified by the software into a signaling pathway heatmap, only the genes whose mRNA levels were distinct between B6 and DBA/2 mice following CAWS treatment were selected. The details of the microarray data have been deposited in the Gene Expression Omnibus (GEO; http://​informahealthcar​e.​com/​doi/​abs/​10.​3109/​08923973.​2013.​830124) database (accession number GSE44803).

Mouse tissues were fixed in 4% paraformaldehyde immediately after removal, then embedded in paraffin and cut into sections (4 μm thick). Sections were stained with H&E according to standard procedures.

Assay-on-Demand TaqMan probes with relevant primers were used for qPCR analysis using the ABI PRISM 7900 according to the manufacturer’s instructions (PE Applied Biosystems, Foster City, CA). Total RNA (500 ng) obtained using the PAXgene Blood RNA Kit was reverse-transcribed with the High Capacity cDNA Archive Kit (ABI). PCR consisted of an initial denaturation (95°C, 10 min) followed by 40 cycles of denaturation (95°C, 15 sec) and annealing/extension (60°C, 1 min). A standard curve was generated from the amplification data for each primer using a dilution series of PBMC RNA as the template. Fold change values were normalized to GAPDH expression levels using the standard curve method according to the manufacturer’s protocol.

All experiments in which mice were used were performed with the approval of the ethics committee of Tokyo University of Pharmacy and Life Sciences.