Control of the innate epithelial antimicrobial response is cell-type specific and dependent on relevant microenvironmental stimuli

Abstract

Immune defence against microbes depends in part on the production of antimicrobial peptides, a process that occurs in a variety of cell types but is incompletely understood. In this study, the mechanisms responsible for the induction of cathelicidin and beta-defensin antimicrobial peptides were found to be independent and specific to the cell type and stimulus. Vitamin D3 induced cathelicidin expression in keratinocytes and monocytes but not in colonic epithelial cells. Conversely, butyrate induced cathelicidin in colonic epithelia but not in keratinocytes or monocytes. Distinct factors induced beta-defensin expression. In all cell types, vitamin D3 activated the cathelicidin promoter and was dependent on a functional vitamin D responsive element. However, in colonic epithelia butyrate induced cathelicidin expression without increasing promoter activity and vitamin D3 activated the cathelicidin promoter without a subsequent increase in transcript accumulation. Induction of cathelicidin transcript correlated with increased processed mature peptide and enhanced antimicrobial activity against Staphylococcus aureus. However, induction of beta-defensin-2 expression did not alter the innate antimicrobial capacity of cells in culture. These data suggest that antimicrobial peptide expression is regulated in a tissue-specific manner at transcriptional, post-transcriptional and post-translational levels. Furthermore, these data show for the first time that innate antimicrobial activity can be triggered independently of the release of other pro-inflammatory molecules, and suggest strategies for augmenting innate immune defence without increasing inflammation.

Figures

Figure 1

Cathelicidin mRNA expression in keratinocytes and colon epithelial cells. Normal human keratinocytes (NHEK) and colonocytes (HT-29) were stimulated with hormones, biological stimuli, cytokines or microbial products for 24 hr. Concentrations selected for this study exceeded those previously described to be effective in these cells (see Materials and methods). Abundance of mRNA was determined by real-time RT-PCR for cathelicidin and the housekeeping gene GAPDH and normalized to vehicle treated controls. Data shown are means (± SD) of the results from a single stimulation experiment performed in triplicates and are representative of at least three independent experiments. (**: P < 0·01; Student's t-test).

Figure 2

Cathelicidin peptide expression in keratinocytes and colon epithelial cells. (a–c) NHEK grown on chamber-slides were stimulated with the vehicle, or 1,25(OH)2 VD3; (d–f) colonocytes (HT-29) were stimulated with the vehicle or butyrate. Cells were stained with a polyclonal anti-LL-37 antibody and nuclei were detected with DAPI in (a,b,d,e). Immunofluorescence is displayed at 400× magnification. Processing to active cathelicidin peptide was evaluated by SELDI-TOF analyses of NHEK and HT-29 cells in (c) and (f) after stimulation with VD3 or butyrate, respectively. A peak at 4 494 Da corresponding to the mature LL-37 peptide was detected in NHEK and HT-29 cells. Data from one representative of three independent experiments are displayed.

Figure 3

Differential expression of HBD-1, HBD-2 and cathelicidin. The induction of human β-defensin 1 (HBD-1), human β-defensin 2 (HBD-2) and cathelicidin were directly compared by real-time RT-PCR. (a) Normal human keratinocytes, (b) HT-29 colon epithelial cells and (c) U937 monocytes were stimulated with a panel of potential stimuli at concentrations described in Materials and methods for 24 hr. Data shown are means (± SD) of a single experiment performed in triplicates and is representative of three independent experiments. (*P < 0·05; **P < 0·01; ***P < 0·001; Student's t-test).

Figure 4

AMP expression dissociates from cytokine expression and induction of differentiation or vitamin D response. IL-8 expression was evaluated by real-time RT-PCR in (a) Normal keratinocytes and (b) HT-29 colon cells after stimulation with different factors as described in Materials and methods. (c) Involucrin expression, as a surrogate marker for keratinocyte differentiation, was analysed by real-time RT-PCR. (d) The expression of the vitamin D receptor and the effect of 1,25(OH)2 vitamin D3 on vitamin D responsive genes was investigated in colon epithelial cells. HT-29 cells were stimulated with the control (lane 1), 1,25(OH)2 VD3 (lane 2) or butyrate (lane 3), harvested and analysed by Western blot employing a specific anti-VDR antibody. A band at approx. 50 kDa was detected corresponding to VDR protein. In (e) and (f), the expression of the vitamin D responsive gene Cyp24A1 (24-hydroxylase) was analysed by real-time RT-PCR in colon epithelial cells (e) or NHEK (f) stimulated with butyrate, 1,25(OH)2 VD3, 25-OH VD3 or the vehicle control. All data shown are means (± SD) of a single experiment performed in triplicate and representative of one of at least three independent experiments. (*P < 0·05; **P < 0·01; ***P < 0·001; Student's t-test).

Figure 5

MEK-ERK activation is necessary but not sufficient for induction of cathelicidin expression in both skin and gut epithelial cells. (a) Keratinocytes (HaCat) were stimulated with 1,25(OH)2 VD3 with or without prior incubation with the MEK-ERK inhibitor U0126 (20 μm, 30 min) and cathelicidin expression assessed by real-time RT-PCR. (b) HT-29 colon cells were stimulated with butyrate with or without prior U0126 treatment and cathelicidin expression was analysed. Data shown are means (± SD) of a single experiment performed in triplicates and are representative of three independent experiments. (**P < 0·01; Student's t-test). (c) To determine if MEK-ERK activation was critical for cathelicidin induction keratinocytes were stimulated with the control (lane 1), 1,25(OH)2 VD3 (lane 3) or EGF (lane 5) for 15 min with or without prior incubation with U0126 (lanes 2,4,6) and subjected to SDS–PAGE and Western blot. (d) HT-29 colon cells were stimulated with the control (lane 1), butyrate (lane 3) or EGF (lane 5) for 15 min with or without prior incubation with U0126 (lanes 2,4,6) and analysed by Western blot. As a positive control a phosphorylated ERK protein was used on both blots (lane 8). Blots were stained with a specific phospho-ERK antibody and visualized by chemiluminescence.

Figure 6

CAMP promoter activity in keratinocytes, colon cells and monocytes. Fragments of the 5′UTR of the human cathelicidin gene CAMP were cloned into a luciferase reporter plasmid. (a) HaCat keratinocytes were transfected with different fragments of the CAMP promoter and stimulated with 1,25(OH)2 VD3 (100 nm). After 24 hr cells were harvested and luciferase activity assayed. Only fragments containing the VDRE at −619 bp to −633 bp relative to the translation start site showed increased transcriptional activity after stimulation. HaCat keratinocytes (b) and U937 monocytes (c) were transfected with promoter constructs containing an intact (pGL3 1500) or deleted VDRE (pGL3 1500-VDRE), stimulated with 1,25(OH)2 VD3 and subsequently luciferase activity measured. Deletion of the VDRE completely blocked VD3-induced transcriptional activity in both cell types. (d) HT-29 colon cells were transfected with pGL3 1500 and pGL3 1500-VDRE construct and stimulated with butyrate (2 mm) or 1,25(OH)2 VD3. Vitamin D3 increased transcriptional activity in HT-29 cells but did not increase mRNA abundance (Fig. 1), while butyrate had no effect on transcriptional activity of the pGL3 1500 construct. (e) To investigate if butyrate increases cathelicidin expression through a transcriptional mechanism, HT-29 cells were stimulated in the presence of Actinomycin D and cathelicidin evaluated by real-time PCR after 8 hr. Despite an inability to CAMP promoter activity, butyrate induced cathelicidin mRNA expression in colon cells was blocked by inhibition of mRNA transcription by Actinomycin D. All data shown are means (± SD) of single experiments performed in triplicates and are representative of at least three independent experiments. (*P < 0·05; **P < 0·01; Student's t-test).

Figure 7

Antimicrobial activity of stimulated keratinocytes and colon epithelial cells. (a) NHEK were stimulated with 1,25(OH)2 VD3 for 24 hr, cells were harvested and cell lysates were coincubated with S. aureusΔmprF and bacterial growth was monitored over time to determine antimicrobial activity. Conditions not containing cell lysates or containing cell lysates from unstimulated cells were used as controls. (b) HT-29 colon cells were stimulated with butyrate for 24 hr and cell lysates were incubated with S. aureusΔmprF. (c) NHEK were stimulated with 1,25(OH)2 VD3, or PMA to induce HBD2 expression and cell lysates were incubated with S. aureusΔmprF. Bacterial growth was measured by OD600. All data shown are means (± SD) of triplicates. (*P < 0·05; **P < 0·01; Student's t-test).