• Crohn's disease;
  • dendritic cells;
  • HSP70;
  • TNF-α


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The objectives were to study the effect of microbial 70 kDa heat shock protein (HSP70) on the production of tumour necrosis factor (TNF)-α and interleukin (IL)-12 by dendritic cells (DC) from patients with Crohn's disease. TNF-α concentration was increased significantly when DC from Crohn's disease were stimulated with HSP70 or CD40L and this was associated with signalling by the extracellular signal regulated kinase (ERK) 1/2 and p38 mitogen activated protein (MAP) kinase pathway. IL-12 production was also increased when DC were stimulated with HSP70. Cells eluted from inflamed intestinal mucosa from Crohn's disease, stimulated with HSP70, CD40L or lipopolysaccharide produced significantly greater TNF-α and IL-12 concentrations than cells from uninflamed mucosa. Significant inhibition of TNF-α production was demonstrated when DC from peripheral blood mononuclear cells or cells eluted from intestinal mucosa of Crohn's disease were treated with either the HSP70 inhibitory peptide (aa 457–496) or peptides derived from CD40 and CD40L. These inhibitory peptides target the CD40–CD40L and the emerging CD40–HSP70 co-stimulatory pathway. Our findings offer a novel strategy to prevent excessive production of TNF-α in Crohn's disease.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The immunopathogenesis of Crohn's disease (CD) has not been elucidated so far, but a dysregulated immune response directed against gut bacteria has been postulated (reviewed in [1]). Genetic predisposition to CD is clearly important [2–4]. Increased frequency of Mycobacterium paratuberculosis and its DNA have been demonstrated in CD [5,6]. However, many Gram-negative and -positive microorganisms and viruses have been associated with CD [7,8]. A breakdown in tolerance to the bacterial flora of the gut may be present because of an increased T cell proliferative response to the intestinal flora in the inflamed bowel [9].

Heat shock protein (HSP) is found in most Gram-positive and Gram-negative intestinal microorganisms. Increased expression of the host 70 kDa HSP (HSP70) has been demonstrated in the intestinal mucosal and submucosal mononuclear cells, as well as epithelial cells in CD [10]. Circulating antibodies and T cell responses to HSP have been reported in inflammatory bowel disease (IBD) [11,12]. A protective role of HSP70 in intestinal epithelial cells against oxidant and other stressful agents has been suggested [13,14]. In non-human primates rectal mucosal administration of HSP70 or HSP65 elicits T and B cell responses and the production of CC chemokines by circulating mononuclear cells [15].

There is compelling evidence that in CD TH1 responses are generated with interleukin (IL)-12 [16–18], interferon (IFN)-γ[19,20] and tumour necrosis factor (TNF)-α[21–24] by intestinal mononuclear cells. One striking clinical development has been the beneficial effect of administration of anti-TNF-α antibodies to patients with CD but not ulcerative colitis (UC) [25,26]. In the cytokine network TNF-α is critical in the control of CD, rheumatoid arthritis and other autoimmune diseases [27]. However, it might be more desirable to prevent rather than neutralize the production of TNF-α, in order to control the immunopathogenesis of CD. This concept was explored, based on the finding that the C-terminal portion of HSP70 (aa359–609) stimulates human monocytes or DC to produce TNF-α and IL-12 [28,29], and that a peptide epitope (aa 457–496) inhibits TNF-α and IL-12 production by DC stimulated with HSP70 or CD40L [30].

The aims of this investigation were to study the effect of HSP70 on cytokine production by DC from patients with CD, compared with those from UC and controls. A further aim was to see whether the interaction between CD40 and CD40L can be inhibited by peptides derived from the sequences of HSP70, CD40 and CD40L. HSP70 or CD40L significantly enhances TNF-α production by DC from patients with CD, and this was inhibited significantly by HSP70-derived peptide 457–496 or those from CD40–CD40L.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References


Human recombinant granulocyte–macrophage colony-stimulating factor (GM-CSF) was obtained from Leucomax (Sandoz Pharmaceuticals, Surrey, UK) and human recombinant IL-4 from R&D Systems (Abingdon, Oxon, UK). Soluble CD40 ligand trimer (CD40LT) was kindly donated by Dr F. Villinger (Atlanta, GA, USA). Lipopolysaccharide (LPS) from Escherichia coli strain 0111B4 was obtained from Sigma (Poole, Dorset, UK).

Preparation of microbial HSP70

The recombinant Mycobacterium tuberculosis HSP70 was prepared from the E. coli pop strain using the pJLA603 vector [31]. HSP70 was purified by ion exchange chromatography using Q-Sepharose resin, followed by ATP-affinity chromatography. The preparation was treated further with polymixin B-coated beads (Sigma-Aldrich, Dorset, UK) to remove LPS. The LPS content of the HSP preparation was determined by the Limulus amoebocyte lysate assay (Sigma-Aldrich) and showed < 0·0006 units/µg of HSP70 or 5 pg/µg of the HSP preparation.

Investigation of contamination of the HSP70 preparations with LPS

Any contamination of HSP70 with LPS was examined as described previously [28]. Briefly, using the intracellular calcium chelator BAPTA/AM, TNF-α production was inhibited in a dose-dependent manner when monocytes were stimulated with HSP70 or the two peptide-binding fragments but not with LPS. Proteinase K differentiated the inhibitory effect of TNF-α production by HSP70 and its two C-terminal fragments but not that of LPS. We then determined the minimal concentration of LPS, which affects the stimulating activity of DC by HSP70. We have shown previously that 10 ng/ml of LPS was required to double the concentration of IL-12 by the two HSP70 preparations or CD40LT [30]. The LPS contamination of the HSP70 preparations was 0·0006 units (or 5 pg) per µg of HSP70, which is 2000 times lower than would be required for LPS to stimulate production of IL-12.

Preparation of synthetic peptides

Peptide 457–496 was synthesized by Bachem Ltd (Bubendorf, Switzerland) to a purity of 93·7%, determined by high pressure liquid chromatography, the sequence of which is: IVHTAKDKGTGKENTIRIQEGSGLSKEDIDRMIKDAEA H. We have also studied peptides derived from the CD40 N-terminal tandem repeat (aa 27–36 and 39–58) with the CD40L extracellular domain 18mer (aa 102–119) to find out if these peptides inhibit the CD40–CD40L co-stimulatory pathway. These were synthesized as above and the sequences of these peptides are as follows: CD40 aa 27–36 REKQYLINSQ; aa 39–58 SLSQPGQKLVSDSTEFTETE; residues underlined indicate substitution of Cys with Ser, to prevent aggregation through disulphide bonds. CD40L aa 102–119 KEETKKENSFEMQKGDQN.

Selection of patients and biopsy procedure

Patients were selected for this study from the gastrointestinal out-patient clinics at Guy's and St Thomas’ Hospitals. The study was approved by the Local Research Ethical Committee and written informed consent was obtained from the patients. A total of 29 patients with Crohn's disease (CD), 22 with ulcerative colitis (UC) and 13 healthy control subjects were included in this investigation. The diagnosis of CD or UC was determined by standard clinical, radiological, endoscopic and histopathological criteria. About 50 ml of blood was taken from each subject for the immunological studies. An additional six colonic, rectal or ileal biopsy specimens were taken at colonoscopy during routine investigation of patients for diagnosis, assessment of disease activity and cancer surveillance. Biopsy specimens were obtained in 13 patients with CD and 12 patients with UC; when biopsies were available from more than one site, these were pooled separately for inflamed and non-inflamed mucosal tissues. Five patients with iron deficiency or altered bowel habits showed normal mucosa on endoscopy and biopsies were taken which also failed to show any histopathological changes. Cells eluted from these five patients were used as controls.

Preparation of human monocytes and monocyte-derived dendritic cells (DC)

Human primary monocytes were isolated from peripheral blood mononuclear cells (PBMC) by centrifugation on a Ficoll-Hypaque density gradient (Amersham Biosciences, Little Chalfont, Bucks, UK). The CD14+ monocytes were enriched by depletion of CD14 cells using a monocyte isolation kit (MACS, Miltenyi Biotec, Surrey, UK). The purity of isolated monocytes was consistently greater than 90% when analysed by flow cytometry with antibody to CD14. Human DC were generated by culturing monocytes with GM-CSF (400 U/ml) and IL-4 (100 U/ml) for 5 days [32]. These monocyte-derived DC were generally considered to be immature DC, defined by surface expression of DC markers CD83, CD80, CD86, CD40, and were CD14 negative.

Assays of TNF-α and IL-12 production

Immature DC were stimulated with HSP70 (10 µg/ml), CD40LT (10 µg/ml) or LPS (1 µg/ml) and cultured for 2 days. IL-12 and TNF-α in the culture supernatants were assayed by enzyme-linked immunosorbent assay (ELISA) using specific antibodies (BD Pharmingen, Cowley, UK), with sensitivity limits of 8 pg/ml for TNF-α and 15 pg/ml for IL-12. The supernatants were diluted ×2 for IL-12 or TNF-α assay. The results were expressed in pg/ml.

The effect of co-stimulation of monocytes or DC with the synthetic peptides and HSP, CD40L or LPS, inducing TNF-α and IL-12 production

To study the effect of the synthetic peptides, human monocytes (1 × 106/ml) or monocyte-derived DC (2 × 105/ml) were incubated with 12·5 and 25·0 µg/ml of the peptides and the cells were stimulated with 10 µg/ml HSP70, 10 µg/ml CD40LT or 500 ng/ml LPS. After culture for 3 days TNF-α and IL-12 were assayed in the supernatants.

Assay of p38 and extracellular signal regulated kinase (ERK) 1/2 phosphorylation MAP kinases

For the phospho-p38 and ERK 1/2 assays, DC were treated with HSP70, CD40L or LPS in the presence or absence of p38 inhibitors (SB 203580; Sigma) or ERK 1/2 inhibitor (PD 098059) for 30 min, then lysed in isotonic buffer with 6 M urea and 0·1% Triton X-100. The phospho-p38 and ERK 1/2 were assayed by ELISA using IC duo-set reagents (R&D Systems, Oxford, UK), according to the manufacturer's instructions. Samples were also subjected to sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) and Western blotting to demonstrate p38 specific phosphorylation.

Elution of mononuclear cells from mucosal tissues

Biopsy material was collected in sterile saline. The cells were passed through a 70 µm sterile nylon cell strainer (Becton Dickinson Biosciences, Oxford, UK Ltd) and treated by collagenase digestion. Briefly, the tissue was suspended in warm RPMI-1640 supplemented with 10% fetal calf serum (FCS), 2 mM l-glutamine and penicillin/streptomycin (100 U/ml), and digested with collagenase at 2 mg/ml for 2 h at 37°C (Sigma-Aldrich UK Ltd). The cells were then filtered through the cell strainer, washed in complete medium, counted using trypan blue exclusion and resuspended in the appropriate medium. The cells were cultured with HSP70, CD40L or LPS as above. In parallel, PBMC were stimulated with the same reagents, cultured for 2 days and the culture supernatant assayed for TNF-α and IL-12. The possibility that the enzyme treatment might have affected the function of eluted mucosal cells was examined by treatment of PBMC by the same procedure as the mucosal tissue, and their capacity to generate TNF-α and IL-12 was unaffected when compared with untreated PBMC.

Statistical analysis

The data between the groups were analysed by the non-parametric Wilcoxon's signed rank test or Mann–Whitney confidence interval test. The results of inhibition studies were analysed by the paired Student's t-test.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Comparative investigation of TNF-α and IL-12 production by stimulating immature DC and monocytes with HSP70, CD40L or LPS

Production of TNF-α was increased significantly (Fig. 1a) when immature DC from patients with CD were stimulated with HSP70 (431 ± 112 pg/ml), compared with controls (133 ± 35 pg/ml; t = 2·55, P = 0·02) or CD40L (1303 ± 422) compared with controls (363 ± 94; t = 2·18, P = 0·046). In contrast, no significant increase in TNF-α production was found in cells from UC whether stimulated with HSP70 or CD40L (Fig. 1a). Stimulation of DC with LPS failed to show significant differences in TNF-α concentrations, although these were lower in CD than in controls.


Figure 1. Stimulation of monocyte-derived dendritic cells (DC) with heat shock protein (HSP)70, CD40L or lipopolysaccharide (LPS) producing (a) tumour necrosis factor (TNF)-α and (b) interleukin (IL)-12 in Crohn's disease (CD) □ (n = 16 and 12), ulcerative colitis (UC) bsl00008 (n = 9 and 8) and healthy controls ▪ (n = 13 and 14), respectively; statistical analyses were performed between CD and controls and UC and controls for each stimulant (*P < 0·05, **P < 0·02). Stimulation of monocytes in these groups producing (c) TNF-α (n = 10–16) and (d) IL-12 (n = 7–9), respectively.

Download figure to PowerPoint

Stimulation of monocytes with HSP70 or CD40L showed increased production of TNF-α with cells from CD (Fig. 1c), but this did not reach statistical significance when compared with controls. Cells from UC were not significantly different from controls in the concentration of TNF-α when stimulated with HSP70 or CD40L. In contrast, LPS elicited less TNF-α, both in CD and UC. A comparison between DC and monocytes stimulated with HSP70 or CD40L showed no significant difference in TNF-α production in any of the three groups of subjects.

Production of IL-12 by monocyte-derived DC was also increased in CD and UC when stimulated with HSP70 (Fig. 1b). However, stimulation of DC with CD40L or LPS elicited a much greater increase in IL-12 concentration than those stimulated with HSP70. Surprisingly, IL-12 production by CD40L stimulated by DC in CD (4594 ± 1068 pg/ml) was significantly less than that from controls (8812 ± 907 pg/ml; P = 0·025, Fig. 1b). Although LPS stimulated a similar decrease in concentration of IL-12 by DC, this was not significant (P = 0·065). DC from patients with UC also showed decreased IL-12 production, but again this was not significant (P = 0·294) (Fig. 1b).

A parallel study of monocytes stimulated with HSP70 showed an increased production of IL-12 in cells from CD, but not in those from UC (Fig. 1d). However, as with DC, CD40L and LPS induced less IL-12 in monocytes from CD. None of the differences in IL-12 concentrations between healthy controls and the two groups of patients was statistically significant. Overall, the results suggest that significant differences in TNF-α and IL-12 production by monocyte-derived DC stimulated with HSP70 or CD40L were found only in DC from CD, compared with those from healthy controls.

Inhibition of HSP70-stimulated TNF-α and IL-12 production by DC with HSP70-derived peptide 457–496

Peptide 457–496 derived from the sequence of HSP70 has been shown to reduce TNF-α and IL-12 production by normal DC or monocytes stimulated with HSP70 or CD40L [35]. With this set of experiments we wanted to assess HSP70-peptide-mediated reduction of TNF-α and IL-12 production by DC from CD and UC patients, compared with controls. The results are presented for optimum inhibition of TNF-α or IL-12 production in healthy controls and the two groups of patients (Table 1a,b). The concentration of TNF-α or IL-12 from DC of healthy controls stimulated with HSP70 was reduced significantly by p457–496 (P = 0·001 and 0·004, respectively; Table 1a,b). In CD both TNF-α and IL-12 were significantly inhibited by p457–496 (P = 0·020 and 0·050, respectively). DC from patients with ulcerative colitis, however, failed to show significant decrease in TNF-α production with p457–496. It is important to note that the concentrations of TNF-α and IL-12 produced by cells from CD were decreased by p457–496 to 18% and 31%, respectively, below the levels of these cytokines produced by cells from healthy controls (Table 1a,b). Thus, TNF-α and IL-12 production can be reduced significantly by treatment with p457–496 in DC from CD but not UC.

Table 1.  The effect of heat shock protein (HSP)70-stimulated dendritic cells (DC) on (a) tumour necrosis factor (TNF)-α and (b) interleukin (IL)-12 production when treated with p457–496 derived from HSP70, and (c) TNF-α, and (d) IL-12 production when treated with pCD40 (p27–36/39–58) and pCD40L (p102–119).
 nMean ± s.e.m. pg/ml
(a) Effect of p457–496 on TNF-α
 1 Controls13187 (64) 78 (41)4·150·001
 2 Crohn's disease13308 (85)153 (54)2·670·020
 3 Ulcerative colitis11330 (162)274 (170)1·840·096
(b) Effect of p457–496 on IL-12
 1 Controls 9436 (159)246 (119)4·060·004
 2 Crohn's disease10523 (178)301 (119)2·270·050
 3 Ulcerative colitis10380 (279)333 (282)6·820·001
(c) Effect of pCD40/CD40L on TNF-α
 1 Controls13192 (63)88 (43)4·060·001
 2 Crohn's disease13309 (90)131 (45)3·10·009
 3 Ulcerative colitis11350 (180)291 (220)0·80·4
(d) Effect of pCD40/CD40L on IL-12
 1 Controls 5212 (43)113 (24)4·620·009
 2 Crohn's disease10405 (134)249 (105)3·810·004
 3 Ulcerative colitis 6450 (266)364 (229)2·080·092

Inhibition of HSP70-stimulated TNF-α and IL-12 production by DC with pCD40/CD40L

We have also used peptides derived from the CD40–N terminal tandem repeat 30mer (p27–36/39–58) and the CD40L extracellular domain 18mer (p102–119) to investigate whether these peptides can inhibit TNF-α and IL-12 production in the CD40–CD40L co-stimulatory pathway. As with p457–496, significant inhibition of TNF-α production by DC was elicited with the CD40/CD40L peptides (Table 1c) in the controls (P = 0·001) and CD (P = 0·009) but not UC. Inhibition of IL-12 by the CD40/CD40L peptides was again similar to that for TNF-α, with significant inhibition of DC from control subjects (P = 0·009) or CD (P = 0·004), but not UC (Table 1d).

P38 and ERK 1/2 phosphorylation MAP kinases

Monocyte-derived DC from CD patients stimulated with CD40L showed a significant increase in TNF-α production (Fig. 1a) but a significant decrease in IL-12 concentration (Fig. 1b). A similar response, although not statistically significant, was also seen with UC (Fig. 1a,b). This dissociated response between TNF-α and IL-12 production was investigated further by the ERK 1/2 and p38 phosphorylation MAP kinase pathways of DC from patients with CD. DC stimulated by HSP70, CD40L or LPS elicited a dose-dependent inhibition of TNF-α production with the ERK 1/2 inhibitor (PD098059) (Fig. 2d,e,f), but not with the p38 inhibitor (SB023580) (Fig. 2a,b,c). However, production of IL-12 was dissociated when the DC were stimulated with HSP70 or CD40L as a dose-dependent decrease of IL-12 (Fig. 2a,b) resulted with the p38 inhibitor (Fig. 2a,b), whereas ERK 1/2 inhibitor enhanced IL-12 production (Fig. 2d,e). The effect of the ERK inhibitor on LPS-induced IL-12 production was inconclusive. The data presented in Fig. 2 are the mean ± s.e.m. of three experiments. These responses did not appear to be specific to DC from CD, as those from normal control subjects showed similar results (data not presented).


Figure 2. Monocyte-derived dendritic cells (DC) from Crohn's disease stimulated with heat shock protein (HSP)70, CD40L or lipopolysaccharide (LPS) to produce tumour necrosis factor (TNF)-α and interleukin (IL)-12p40, and treated with increasing concentrations of p38 inhibitor (SB023580) (a, b, c) or extracellular signal regulated kinase (ERK) inhibitor (PD098059) (d, e, f); presented as mean ± s. e.m. pg/ml.

Download figure to PowerPoint

Stimulation of TNF-α and IL-12 production by HSP70, CD40L or LPS of mononuclear cells eluted from intestinal biopsy specimens of patients with CD, UC and normal subjects

Eluted cells from five normal colonic mucosal specimens failed to be stimulated by HSP70, CD40L or LPS to produce TNF-α (0–3·5 pg/ml). In CD and UC the mucosal tissues were divided into inflamed and uninflamed, based on endoscopic appearance and then histological examination. In CD stimulation of eluted cells from uninflamed mucosal tissue with the three reagents produced very little TNF-α; the means were 7·3 ± 3·4, 8·7 ± 4·9 and 12·7 ± 4·7 pg/ml, respectively (Fig. 3a). However, cells eluted from the inflamed mucosa of patients with CD, compared with those from uninflamed mucosa, produced significantly increased concentrations of TNF-α when stimulated by HSP70 (41·7 ± 13·7 pg/ml, P = 0·001), CD40L (41·0 ± 14·7 pg/ml, P = 0·055), which is just over the significant level, and LPS (51·9 ± 21·0 pg/ml, P = 0·007).


Figure 3. Stimulation by heat shock protein (HSP)70, CD40L or lipopolysaccharide (LPS) of tumour necrosis factor (TNF)-α production by cells eluted from mucosal biopsies from patients with (a) Crohn's disease (inflamed n = 13 and uninflamed n = 13) or (b) ulcerative colitis (inflamed n = 8, uninflamed n = 12) and inhibition of HSP70-stimulated cells with p457–496 or pCD40/CD40L (c) in Crohn's disease (inflamed n = 11, uninflamed n = 12) and (d) in ulcerative colitis (inflamed n = 7, uninflamed n = 11).

Download figure to PowerPoint

Cells eluted from UC mucosa produced similar concentrations of TNF-α to CD cells when they were stimulated with the three reagents (Fig. 3b). However, a significantly higher concentration of TNF-α (P < 0·05) was found with mucosal cells eluted from inflamed compared with those from uninflamed mucosa when stimulated with CD40L but not with HSP70 or LPS (Fig. 3b). Intraepithelial and subepithelial (lamina propria) cells were not studied separately, as the specimens of mucosal tissues were inadequate to yield a sufficient number of cells.

IL-12 production was then evaluated by stimulating the cells eluted from inflamed colonic mucosa of CD or UC. The concentration of IL-12 was greater in cells from CD than UC when stimulated with HSP70, CD40L or LPS, but none of these differences reached significant levels (Table 2a).

Table 2.  Stimulation of interleukin (IL)-12 production by heat shock protein (HSP)70, CD40L or lipopolysaccharide (LPS) of mononuclear cells eluted from colonic biopsy specimens from patients with Crohn's disease or ulcerative colitis, and inhibition of HSP70-stimulated cells with p457–496 or pCD40/CD40L.
Mucosal cellsnConcentration of IL-12, mean ± s.e.m. pg/mlnHSP70-stimulated cells inhibited with
(a) Crohn's disease
 Uninflamed 824·9 (10·2)39·9 (11·6)32·6 (11·9) 820·3 (8·0)2·8 (1·8)20·3 (8·0) 6·6 (3·8)
 Inflamed1028·1 (7·3)30·9 (12·4)58·0 (30·5)1024·3 (8·0)6·6 (2·9)24·3 (7·2)11·5 (2·9)
 P-value  0·753 0·396 1·0 Uninflamed Inflamed0·061 0·022 0·043  0·023 
(b) Ulcerative colitis
 Uninflamed 6 6·1 (2·8) 7·2 (5·1)16·2 (8·9) 619·2 (12·1)4·5 (2·9)19·2 (12·1) 5·0 (2·9)
 Inflamed 519·9 (12·7) 3·4 (3·2)23·2 (14·4) 517·3 (9·2) Uninflamed Inflamed5·8 (4·2) 0·18 0·3017·3 (9·2)  0·19  0·1712·5 (7·2)

Inhibition of HSP70-stimulated TNF-α and IL-12 production by cells eluted from intestinal biopsy specimens by peptides derived from HSP70 (p457–496) or CD40/CD40L (p27–36/p39–58 and 102–119)

Both peptides inhibited TNF-α production by more than 50% from a mean of 41·9 ± 14·1 to 10·8 ± 4·8 pg/ml (P = 0·016) with p457–496, and from 41·9 ± 14·1 to 15·2 ± 5·1 pg/ml with p27–36/39–58/p102–119 (P = 0·018), when the cells eluted from inflamed mucosa of CD were stimulated with HSP70 (Fig. 3b). Cells from uninflamed mucosal tissue produced little TNF-α, but even this was inhibited significantly (Fig. 3b). Cells eluted from both inflamed and uninflamed mucosal tissues of UC also showed inhibition of TNF-α production with the two sets of peptides, but surprisingly, significant inhibition was found only in those from uninflamed mucosa (Fig. 3d).

Similar investigation of HSP70 induced IL-12 production (Table 2a) showed significant inhibition by both p457–496 (P = 0·022) and pCD40/CD40L (P = 0·023) for cells eluted from inflamed mucosa of CD; cells from uninflamed mucosa also showed significant inhibition with pCD40/CD40L. However, significant changes in IL-12 production were not found with cells eluted from UC (Table 2b). It should be noted that the enzyme treatment of mucosal specimens used to elute cells did not affect the functional capacity of the treated cells to generate TNF-α or IL-12, as equally treated PBMC, stimulated with HSP70, CD40L or LPS showed similar production of these cytokines as untreated PBMC (data not presented).


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

DC have been accorded a central role in the interphase between immunity and tolerance [33,34]. We have pursued the present study of DC in Crohn's disease with special reference to responses to HSP70, which is found in Gram-positive and Gram-negative gut organisms. As CD40 is a receptor for microbial HSP70 [28], the CD40L (CD154) trimer was used as a positive control. LPS is present in Gram-negative gut organisms and binds to CD14 [35], and was therefore used as a relevant CD40 negative control.

Stimulation by HSP70 of monocyte-derived DC showed significant increases in TNF-α and IL-12 production in CD but not UC. However, stimulation with CD40L resulted in a significant increase in TNF-α but a decrease in IL-12 in DC from CD. The dissociation between TNF-α and IL-12 production by stimulating DC from CD with CD40L might be accounted for by the ERK 1/2 phosphorylation MAP kinase pathways, as the inhibitor enhanced IL-12, while inhibiting TNF-α production. The increase in TNF-α concentration produced by DC from CD stimulated with HSP70 or CD40L is consistent with that found in the literature [21–23]. Although a decrease in IL-12 production by stimulation with CD40L, but not with HSP70, in the circulating monocyte-derived DC was demonstrated in the present study, a slight increase in IL-12 was found with cells eluted from inflamed tissues and this is consistent with the data found in the literature [17,18]. It should be pointed out that HSP70 stimulated relatively modest amounts of TNF-α and IL-12, compared with those stimulated by CD40L or LPS.

There is a great deal of evidence that the CD40–CD40L co-stimulatory pathway is involved in the immunopathogenesis of inflammatory bowel disease [36]. An increase in CD40+ cells was reported in mononuclear, endothelial and mesenchymal cells from patients with intestinal lesions in CD and UC [37–39]. CD40L+ CD4+ cells and some CD8+ T cells were also increased in CD and UC [38,40]. The proportions of circulating CD40+ lymphocytes and monocytes are also increased in patients with CD [41]. Furthermore, colitis was reported in a transgenic CD40L mouse model [42]. Because HSP70 may function as an alternative to CD40L in the interaction with CD40 [28,29,43–45], this may enhance the already heightened CD40–CD40L interaction in CD. CD40–CD40L and the alternative CD40–HSP70 pathway are especially relevant in the intestinal mucosa, which is exposed to HSP70 from gut bacteria. These potent co-stimulatory pathways provide a mechanism for recruitment of T and B cells, monocytes and DC into the mucosa and production of proinflammatory cytokines, which may maintain chronic inflammation, especially in genetically susceptible subjects.

As TNF-α is critical in the control of CD and autoimmune diseases [27], and HSP70-derived peptide 457–496 inhibits TNF-α production by DC from healthy control subjects [30], we explored the possibility that p457–496 might also inhibit TNF-α in CD. Indeed, addition of p457–496 to HSP70 significantly inhibited TNF-α production by DC from CD but not those from UC. It is noteworthy that p457–496 inhibits TNF-α production to below the level found in healthy controls, but did not prevent TNF-α production completely, thereby allowing normal function of TNF-α. It is of interest that mycobacterial and other microbial HSP70 show extensive homology with human HSP70 [46] and mycobacteria have been implicated directly in the pathogenesis of CD [5,6]. Furthermore, microbial HSP70 elicits chemokines and cytokines via the CD40 co-stimulatory molecules [28,29], and these may affect host immunity. Indeed, the alternative HSP70–CD40 co-stimulatory pathway has been invoked in priming T cells, in the control of tuberculosis [44] and in converting T cell tolerance to autoimmunity [45].

It was of interest to find out if there is any homology between the mycobacterial p457–496 and that found in human HSP70. Aligning the mycobacterial (457–496) with human HSP70 (p485–524) sequences revealed that 29/40 (73%) residues are conserved. However, the critical residues within the 40mer peptide have not, as yet, been defined. This might be significant, as the mycobacterial N-terminal portion of the inhibitory peptide (aa 457–471) shows that 12/15 (80%) residues are identical or conserved when compared with those of the human HSP70 (aa 485–499). The C-terminal portion (aa 480–494) shows 15/15 (100%) homology.

An alternative strategy to inhibiting TNF-α and IL-12 stimulated by the CD40–CD40L or CD40–HSP70 pathway was to use peptides derived from the CD40, N-terminal tandem repeat (aa 27–36 and 39–58) and from the CD40L cytoplasmic tail (aa 102–119). Significant inhibition of production of both cytokines was found with these two peptides in DC from CD and healthy controls, but not UC. Thus TNF-α and IL-12 production by HSP70-stimulated DC from CD can be inhibited by a peptide derived from HSP70 that targets CD40 [30] and by peptides from the CD40 and CD40L molecules.

It was important to establish whether cells eluted from intestinal mucosa affected by CD respond to stimulation with HSP70, CD40L and LPS in a similar manner to that found with PBMC. Indeed, cells eluted from the inflamed tissue when stimulated with HSP70, CD40L and LPS showed a significant increase in TNF-α production in CD but not UC. As both CD and UC are mucosal ulcerative lesions, the raised TNF-α concentration is unlikely to be due to a greater influx of microbial HSP70 through the damaged mucosa, increasing stimulation of TNF-α production. We favour the concept that the increased number of CD40+ cells in CD enables microbial HSP70 to potentiate TNF-α production. We also studied the effect of the two sets of inhibitory peptides on the HSP70-stimulated cells eluted from CD and UC. Significant inhibition of TNF-α and IL-12 production was found, with both p457–496 and p27–36/39–58/102–119 using eluted cells from inflamed CD but not UC mucosa. The small number of cells eluted from the intestinal biopsy specimens obliged us to use the entire cell population, unlike the circulating cells from which monocyte-derived DC were isolated. Cells eluted from the mucosal tissues contain epithelial cells, in addition to lymphocytes, macrophages, DC and neutrophils, and these may have a differential effect on TNF-α and IL-12 production, especially when comparing epithelial cells from inflamed and uninflamed tissues. However, irrespective of any contributory role of epithelial cells, TNF-α and IL-12 production were inhibited significantly by both the HSP70-derived peptide (aa 457–496) and CD40/CD40L peptides (aa27–36/39–58/102–119), which may have a therapeutic effect in CD. Further work is required to establish whether mucosal DC are involved in these changes, but the results are consistent with DC playing a role in the immunopathogenesis of CD.

An increase in TNF-α production was found to be stimulated by HSP70 or CD40L, with both mucosal cells and circulating monocyte-derived DC. The inhibition experiments, targeting the CD40-CD40L or CD40-HSP70 co-stimulatory pathway, raise the possibility that the two sets of inhibitory peptides may offer an alternative strategy for inhibition of TNF-α production in CD by direct rectal or oral administration of either peptide, or possibly both sets of peptides (Fig. 4). This peptide inhibitory strategy might prevent excessive TNF-α production, compared with anti-TNF-α antibodies that neutralize any TNF-α already produced. Such an approach may complement existing treatment with monoclonal antibodies to TNF-α, which have to be given by injection, and usually with cytotoxic drugs.


Figure 4. A novel strategy for inhibition of CD40–CD40L or CD40–heat shock protein (HSP)70 interaction by peptides derived from HSP70 or CD40 and CD40L, stimulating production of tumour necrosis factor (TNF)-α, interleukin (I)-12 and CCL-5.

Download figure to PowerPoint


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This work was supported by the Guy's & St Thomas’ Charitable Foundation. We wish to thank Drs Wendy Clarke, Abdul Mohsen and Bijay Baburajan in helping with the intestinal biopsy specimen.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Duchmann R, Zeitz M. Crohn's disease. In: OgraPL, MesteckyJ, LammME, StroberW, BienenstockJ, McGheeJR, eds. Mucosal immunology. San Diego: Academic Press, 2000: 105580.
  • 2
    Bouma G, Strober W. The immunological and genetic basis of inflammatory bowel disease. Nat Rev Immunol 2003; 3: 321533.
  • 3
    Mathew CG, Lewis CM. Genetics of inflammatory bowel disease: progress and prospects. Hum Mol Genet 2004; 13: 1618.
  • 4
    Hampe J, Frenzel H, Mirza MM et al. Evidence for a NOD2-independent susceptibility locus for inflammatory bowel disease on chromosome 16p. Proc Natl Acad Sci USA 2002; 99: 3216.
  • 5
    Sanderson JD, Moss MT, Tizard ML, Hermon-Taylor J. Mycobacterium paratuberculosis DNA in Crohn's disease tissue. Gut 1992; 33: 8906.
  • 6
    Naser SA, Ghobrial G, Romero C, Valentine JF. Culture of Mycobacterium avium subspecies paratuberculosis from the blood of patients with Crohn's disease. Lancet 2004; 18: 103944.
  • 7
    Liu Y, Van Kruiningen HS, West AB, Cartun RW, Cortot A, Colombel JF. Immunocytochemical evidence of listeria, E. coli, and streptococcus antigens in Crohn's disease. Gastroenterology 1995; 108: 1369404.
  • 8
    Robertson D, Sandler R. Measles virus and Crohn's disease: a critical appraisal of the current literature. Inflamm Bowel Dis 2001; 7: 517.
  • 9
    Duchmann R, Kaiser I, Hermann E, Mayet W, Ewe K, Zum M. Tolerance exists towards resident intestinal flora but is broken in active inflammatory bowel disease (IBD). Clin Exp Immunol 1995; 102: 44855.
  • 10
    Ludwig D, Stahl M, Ibrahim ET et al. Enhanced intestinal expression of heat shock protein 70 in patients with inflammatory bowel disease. Dig Dis Sci 1999; 44: 14407.
  • 11
    Elsaghier A, Prantera C, Bothamley E, Wilkins E, Jindal S, Ivanyi J. Disease association of antibodies to human and mycobacterial hsp70 and hsp60 stress proteins. Clin Exp Immunol 1992; 89: 3059.
  • 12
    Pirzer U, Schonhaar A, Fleischer B, Hermann E, Meyer E, Meyersum Buschenfelde K-H. Reactivity infiltrating T lymphocytes with microbial antigens in Crohn's disease. Lancet 1991; 338: 12389.
  • 13
    Musch MW, Ciancio MJ, Sarge K, Chang EB. Induction of heat shock protein 70 protects intestinal epithelial IEC-18 cells from oxidant and thermal injury. Am J Physiol 1996; 270: C42936.
  • 14
    Wischmeyer PE, Musch MW, Madonna MB, Thisted R, Chang EB. Glutamine protects intestinal epithelial cells: role of inducible HSP70. Am J Physiol 1997; 272: G87984.
  • 15
    Lehner T, Bergmeier LA, Wang Y et al. Heat shock proteins generate β-chemokines which function as innate adjuvants enhancing adaptive immunity. Eur J Immunol 2000; 30: 594603.
  • 16
    Trinchieri G. Interleukin-12: a cytokine produced by antigen-presenting cells with immunoregulatory functions in the generation of T-helper cells type 1 and cytotoxic lymphocytes. Blood 1994; 84: 400627.
  • 17
    Parronchi P, Romagnani P, Annunziato F et al. Type 1 T-helper cell predominance and interleukin-12 expression in the gut of patients with Crohn's disease. Am J Pathol 1997; 150: 82332.
  • 18
    Monteleone G, Biancone L, Marasco R et al. Interleukin 12 is expressed and actively released by Crohn's disease intestinal lamina propria mononuclear cells. Gastroenterology 1997; 112: 116978.
  • 19
    Kakazu T, Hara J, Matsumo T et al. Type 1 helper cell predominance in granulomas of Crohn's disease. Am J Gastroenterol 1999; 8: 214955.
    Direct Link:
  • 20
    Fuss IJ, Neurath MF, Boirivant M et al. Disparate CD4+ lamina propria (LP) lymphokine secretion profiles in inflammatory bowel disease. Crohn's disease LP cells manifest increased secretion of IFN-γ, whereas ulcerative colitis LP cells manifest increased secretion of IL-5. J Immunol 1996; 157: 126170.
  • 21
    Breese EJ, Michie CA, Nicholls SW et al. Tumor necrosis factor α-producing cells in the intestinal mucosa of children with inflammatory bowel disease. Gastroenterology 1994; 106: 145566.
  • 22
    Reinecker H-C, Steffen M, Witthoeft T et al. Enhanced secretion of tumour necrosis factor-alpha, IL-6 and IL-1β by isolated lamina propria mononuclear cells from patients with ulcerative colitis and Crohn's disease. Clin Exp Immunol 1993; 94: 17481.
  • 23
    Woywodt A, Neustock P, Kruse A et al. Cytokine expression in intestinal mucosal biopsies. In situ hybridisation of the mRNA for interleukin-1β, interleukin-6 and tumour necrosis factor-α in inflammatory bowel disease. Eur Cytokine Netw 1994; 5: 38795.
  • 24
    Kollias G, Douni E, Kassiotis G, Kontoyiannis D. On the role of tumor necrosis factor and receptors in models of multiorgan failure, rheumatoid arthritis, multiple sclerosis and inflammatory bowel disease. Immunol Rev 1999; 169: 17594.
  • 25
    Targan SR, Hanauer SB, Van Deventer SJ et al. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor α for Crohn's disease. Crohn's Disease cA2 Study Group. N Engl J Med 1997; 337: 102935.
  • 26
    Rutgeerts P, Van Assche G, Vermeier S. Optimizing anti-TNF treatment in inflammatory bowel disease. Gastroenterology 2004; 126: 1593610.
  • 27
    Feldman M, Maini RN. Anti-TNF-α therapy of rheumatoid arthritis: what have we learned? Ann Rev Immunol 2001; 19: 16396.
  • 28
    Wang Y, Kelly CG, Karttunen JT et al. CD40 is a cellular receptor mediating mycobacterial heat shock protein 70 stimulation of CC-chemokines. Immunity 2001; 15: 97183.
  • 29
    Wang Y, Kelly CG, Singh M et al. Stimulation of Th1-polarizing cytokines, C-C chemokines, maturation of dendritic cells and adjuvant function by the peptide binding fragment of heat shock protein 70. J Immunol 2002; 169: 24229.
  • 30
    Wang Y, Whittall T, McGowan E et al. Identification of stimulating and inhibitory epitopes within the HSP70 molecule which modulate cytokine production and maturation of dendritic cells. J Immunol 2005; 174: 330616.
  • 31
    Mehlert A, Young DB. Biochemical and antigenic characterization of the Mycobacterial tuberculosis 71 kD antigen, a member of the 70 kD heat-shock protein family. Mol Microbiol 1989; 3: 12530.
  • 32
    Sallusto F, Lanzavecchia A. Mobilizing dendritic cells for tolerance, priming and chronic inflammation. J Exp Med 1999; 189: 611.
  • 33
    Steinman RM, Hawiger D, Nussenzweig MC. Tolerogenic dendritic cells. In: PaulWE, FathmanCG, GlimcherLH, eds. Annual review of immunology, vol. 21. Palo Alto: Non-profit publishers, 2003: 685–711.
  • 34
    Moser M. Dendritic cells in immunity and tolerance − do they display opposite function [Review]. Immunity 2003; 19: 58.
  • 35
    Wright SD, Ramos RA, Tobias TS, Ulevitch RJ, Mathison JC. CD14, a receptor for complexes of lipopolysaccharide (LPS) and LPS binding protein. Science 1990; 249: 14313.
  • 36
    Danese S, Sans M, Fiocchi C. The CD40/CD40L costimulatory pathway in inflammatory bowel disease. Gut 2004; 53: 103543.
  • 37
    Battaglia E, Biancone L, Resegotti A. Expression of CD40 and its ligand, CD40L, in intestinal lesions of Crohn's disease. Am J Gastroenterol 1999; 94: 327984.
    Direct Link:
  • 38
    Gelbmann CM, Leeb SN, Vogel D. Inducible CD40 expression mediates NFB activation and cytokine secretion in human colonic fibroblasts. Gut 2003; 52: 144856.
  • 39
    Vogel JD, West GA, Sturm A. CD40-mediated immune–nonimmune cell interactions induce mucosal fibroblast chemokines causing T-cell transmigration. Gastroenterology 2004; 126: 6380.
  • 40
    Liu Z, Colpaert S, D’Haens GR et al. Hyperexpression of CD40 ligand (CD154) in inflammatory bowel disease and its contribution to pathogenic cytokine production. J Immunol 1999; 163: 404957.
  • 41
    Sawada-Hase N, Kiyohara T, Miyagawa J et al. An increased number of CD40-high monocytes in patients with Crohn's disease. Am J Gastroenterol 2000; 95: 151623.
    Direct Link:
  • 42
    Clegg CH, Rulffes JT, Harald S et al. Thymus dysfunction and chronic inflammatory disease in gp39 transgenic mice. Int Immunol 1997; 9: 111122.
  • 43
    Becker T, Hartl F-U, Weiland F. CD40, an extracellular receptor for binding and uptake of Hsp70–peptide complexes. J Cell Biol 2002; 158: 127785.
  • 44
    Lazarevic V, Myers AJ, Scanga CA, Flynn JL. CD40, but not CD40L, is required for the optimal priming of T cells and control of aerosol M. tuberculosis infection. Immunity 2003; 19: 82335.
  • 45
    Millar DG, Garza KM, Odermatt B et al. HSP70 promotes antigen-presenting cell function and converts T-cell tolerance to autoimmunity in vivo. Nat Med 2003; 9: 1469523.
  • 46
    Hunt C, Morimoto RI. Conserved features of eukaryotic hsp70 genes revealed by comparison with the nucleotide sequence of human hsp70. Proc Natl Acad Sci USA 1985; 82: 64559.