PMC

Quantitative real-time PCR analysis of selected mRNAs in pools of villus and crypt epithelial cells. Real-time fluorescence monitoring of PCR reactions using cDNA from villus and crypt samples obtained by LCMD. PCR primer pairs for HD5, Reg3A, aldolase-B or β-Actin were used in PCR reactions with cDNA templates from villus epithelial cells (VEC) or crypt epithelial cells (CEC). NTC, no template control. The cDNA was used as a template in real-time PCR analysis with specific primers (Table 1).

Localization of epithelial biomarkers by immunohistochemistry. Expression of HD5 in crypt epithelium and aldolase-B in villus epithelium is shown by immunohistochemistry analysis. HD5 (A) and aldolase-B (B) protein expression were evaluated by immunohistochemistry using peroxidase detection of mouse monoclonal anti-HD5 and anti-aldolase-B antibodies in consecutive serial tissue sections. Expression of HD5 was detected exclusively in Paneth cells, while aldolase-B was observed primarily in the epithelial cells of the villus region, providing supportive protein level evidence for the transcriptional differences elucidated by microarray and RT-PCR based assays.

Detection of known villus epithelial and Paneth cell biomarkers. Relative transcriptional levels of previously characterized Paneth cell and villus epithelial biomarkers. Fold differences in expression between the crypt and villus epithelial regions are shown for HD5, HD6, trypsin, lysozyme, PLA2IIA (crypt epithelium) and carboxyesterase-2, PEP carboxykinase, aminopeptidase N, apolipoprotein A-IV, aldolase-B (villus epithelium) to demonstrate the magnitude of differential expression between the 2 cellular compartments and provide supportive evidence that the appropriate cells were captured by LCMD. Statistical validation of the differences in expression levels were determined by student t-test (P value ≤ 0.05).

LCMD of villus epithelial and Paneth cells. A. Villus and crypt epithelial cells were excised from sections of formalin fixed human ileal tissue samples by LCMD. Areas where cells from each lineage were captured are labeled. The mRNA from the isolated cells was amplified, labeled and analyzed using both microarray and RT-PCR based methodologies. B. Schematic diagram of the cell types that comprise the human ileal epithelium. Epithelial cell lineages, including adsorptive, enteroendocrine, goblet, and Paneth cells are color-coded and their distribution within the epithelium is shown. The location of stem cells, shown in green, and the bidirectional cell migration is depicted in the magnified crypt region on the right.

Reg3A and RETLNB are transcribed exclusively in crypt epithelial cells. Transcript levels of the anti-microbial, Reg3A (A) and a potential regulator of Reg3A, RETLNB (B) in crypt epithelial cells and villus epithelium of three ileal tissue samples. The mRNA levels of Reg3A were similar to the high levels measured for alpha-defensins, supporting a potentially important and orchestrated role for these proteins in innate defense responses in the human ileum. Fold differences in expression between the crypt and villus epithelial regions are show for 3 independent patient samples (P1, P2, and P3). Statistical validation of the fold-differences in mRNA levels was determined by student t-test (P value ≤ 0.05).

Transcriptional profile of genes associated with villus epithelial cell maturation, adhesion, and defense response. Multiple genes involved in epithelial cell adhesion, growth and maturation, and defense response were found to be transcribed exclusively in the villus epithelium. Interestingly, the genes involved in epithelial adhesion and differentiation were distinct from those identified in Paneth cells, suggesting that each lineage employs a separate and specific set of molecular mechanisms to promote appropriate growth, development, and positioning. Increased transcription of multiple mucins transcripts in the villus epithelium was also detected, an expected result given the observed level of goblet cells in tissue sections of the ileal samples that were used for LCMD. Statistical validation of the fold-differences in mRNA levels was determined by student t-test (P value ≤ 0.05).

Identification of crypt epithelial cell genes associated with differentiation, positioning, and secretory functions. Four genes expressed in crypt epithelial cells, but not in villus epithelium, and potentially involved in epithelial cell maturation (homeobox B5), positioning (ephrin receptor B3, semaphorin F), and granule secretion (Rab26) were identified through microarray based transcriptional profiling. The lack of expression of these genes in villus epithelium may indicate that they function in concert exclusively in the Paneth cell lineage to promote cell-specific differentiation and migration patterns that lead positioning at the bottom of crypts and antimicrobial peptide secretion. Statistical validation of the fold-differences in mRNA levels was determined by student t-test (P value ≤ 0.05).

Quantitative real-time PCR analysis of selected mRNAs in human gastrointestinal tissues. Genes were selected for analysis from those that showed differential crypt-villus expression by microarray analysis, including mRNA encoding HD5, Reg3A, aldolase-B, Eps8L2, Muc3β, semaphorin3F, RETLNB, ApoA4, Eph-Rb3 and Pdxp. The mRNA copy counts were determined from standard curves using gene-specific plasmids and are expressed as a fraction relative to measure values for β-actin. Ileal samples (n = 5) and colon (n = 5) were from non-diseased surgical samples of adjoining tissue resected for bowel obstruction or colon cancer. Single samples of commercially obtained (pooled) RNA were analyzed for the other tissues. Error bars represent standard deviations from the geometric mean. For single specimen samples (E, D, J) inter-assay variation was <15% for mRNA copy counts ≥10 exp4/10 ng RNA, and 40% for mRNA copy counts ≤10 exp2/10 ng RNA. E, esophagus; D, duodenum. J, jejunum; I, ileum; C, colon.

High throughput gene expression profiling by microarray analysis. A. M vs A plot showing genes identified as differentially expressed (2-fold or more, P value ≤ 0.05) between villus and crypt epithelial cells in red. Genes were identified through comparison of median expression levels in 3 independent paraformaldehyde-fixed ileal tissue samples from 3 different individuals. B. Hierarchical clustering of the genes differentially expressed (2-fold or more, P value ≤ 0.05). Common patterns of gene transcription were determined by non-biased clustering analysis, revealing a set of genes whose expression was enriched in each of the 2 lineages, villus and crypt epithelial cells. C. Pathway analysis of the differentially expressed genes reveals physiological processes that predominate the biology of villus and crypt epithelial cell lineages. Genes enriched in the 2 lineages evaluated were analyzed statistically to determine which pathways and processes were most represented each cell type. The top 5 pathways/processes implicated for each lineage are shown.