Gene expression profiling of mouse bladder inflammatory responses to LPS, substance P, and antigen-stimulation

Abstract

Inflammatory bladder disorders such as interstitial cystitis (IC) deserve attention since a major problem of the disease is diagnosis. IC affects millions of women and is characterized by severe pain, increased frequency of micturition, and chronic inflammation. Characterizing the molecular fingerprint (gene profile) of IC will help elucidate the mechanisms involved and suggest further approaches for therapeutic intervention. Therefore, in the present study we used established animal models of cystitis to determine the time course of bladder inflammatory responses to antigen, Escherichia coli lipopolysaccharide (LPS), and substance P (SP) by morphological analysis and cDNA microarrays. The specific aim of the present study was to compare bladder inflammatory responses to antigen, LPS, and SP by morphological analysis and cDNA microarray profiling to determine whether bladder responses to inflammation elicit a specific universal gene expression response regardless of the stimulating agent. During acute bladder inflammation, there was a predominant infiltrate of polymorphonuclear neutrophils into the bladder. Time-course studies identified early, intermediate, and late genes that were commonly up-regulated by all three stimuli. These genes included: phosphodiesterase 1C, cAMP-dependent protein kinase, iNOS, beta-NGF, proenkephalin B and orphanin, corticotrophin-releasing factor (CRF) R, estrogen R, PAI2, and protease inhibitor 17, NFkB p105, c-fos, fos-B, basic transcription factors, and cytoskeleton and motility proteins. Another cluster indicated genes that were commonly down-regulated by all three stimuli and included HSF2, NF-kappa B p65, ICE, IGF-II and FGF-7, MMP2, MMP14, and presenilin 2. Furthermore, we determined gene profiles that identify the transition between acute and chronic inflammation. During chronic inflammation, the urinary bladder presented a predominance of monocyte/macrophage infiltrate and a concomitant increase in the expression of the following genes: 5-HT 1c, 5-HTR7, beta 2 adrenergic receptor, c-Fgr, collagen 10 alpha 1, mast cell factor, melanocyte-specific gene 2, neural cell adhesion molecule 2, potassium inwardly-rectifying channel, prostaglandin F receptor, and RXR-beta cis-11-retinoic acid receptor. We conclude that microarray analysis of genes expressed in the bladder during experimental inflammation may be predictive of outcome. Further characterization of the inflammation-induced gene expression profiles obtained here may identify novel biomarkers and shed light into the etiology of cystitis.

Figure 1.

A-F: Acute and chronic inflammation. A: H&E stains of cross-sections of mouse urinary bladder responding to acute (A and D) and chronic (B, C, E, F) intravesical stimulation with LPS. Notice intense leukocyte infiltrate in the acute response (A, D) and intense perivascular infiltrate of plasma cells, lymphocytes, and monocytes in the chronic model (B, C, E, F). D–F: Immunohistochemistry of cross-sections obtained from chronic inflamed bladders stained with rat anti-mouse macrophage monocyte antibody and secondary antibody anti-mouse Fab-HRP. Black arrow and circle indicate a plasma cell, red arrows indicate PMNs, white arrows indicate lymphocytes; and green arrows indicate macrophages/monocytes.

Figure 2.

A–B: Venn diagram of ratio of gene expression obtained in bladder isolated from sensitized C57BL/6J mice that were instilled with 150 μl of one of the following substances: pyrogen-free saline, SP (10 μmol/L), LPS (100 μg/ml), or antigen DNP₄-OVA (1 μg/ml). Twenty-four hours after, bladders were removed, RNA was extracted, reverse-transcribed to cDNA, and hybridized to 1.2K mouse membranes (Clontech). Venn diagrams were obtained using raw data and filtering genes that were at least threefold up-regulated when comparisons were made within each animal group between the two conditions, antigen and saline, SP and saline, and LPS and saline. Results in A represent number of genes that were up-regulated and in B genes that were down-regulated. For each group 1200 genes were analyzed by two different hybridizations using two different pools of RNA isolated from sensitized mice that were challenged with saline, antigen, LPS, and SP. Mice were sacrificed following three different time points (1, 4, and 24 hours after acute stimulation). Therefore, a total of 28,800 data points were analyzed.