Results

Degraded CGN Induce Colonic Inflammation
All rats developed diarrhea during degraded carrageenan administration and gross evidence of blood was frequently detected in the stools. Colon length dramatically decreased in all treated rats with a more pronounced effect being observed in the 40 kDa dCGN treated group (Fig. 1A). Furthermore, prolonged exposure to 40 kDa dCGN resulted in high macroscopic and histological scores of inflammation (Fig. 1B, C). Only weak myeloperoxidase activity was detected in both control and dCGN-treated groups (Fig. 1D), indicating that granulocytes did not play a major role in the inflammation at that stage. Histological examination revealed various degrees of mucosal inflammation. Rats treated with 10 kDa dCGN showed edema, epithelium atrophy and slight lymphocyte infiltration (data not shown). These symptoms were totally absent in the colon of control rats (Fig. 1E). More severe mucosal injuries including ulceration, hyperplastic epithelium, crypt distortion and a strong macrophage infiltration, were observed in the 40 kDa dCGN-treated rats (Fig. 1F). No sulphated polysaccharides were detected by toluidine blue staining of colon mucosa from rats treated with either the 10 or 40 kDa dCGN (not shown). Although we cannot exclude that dCGN mat not have retained in the section during the histology procedure, this indicates that these polymers may not have been phagocytosed.
10.1371/journal.pone.0008666.g001 Figure 1  Degraded CGN induced colon inflammation in rats.
Histograms showing the effect of degraded CGN on: colon length (A); macroscopic (B) and histological (C) inflammation score of colon; Myeloperoxidase (MPO) activity (D). Control rats (white bars); 10 kDa degraded CGN-treated rats (grey bars); 40 kDa degraded CGN-treated rats (black bars). * p<0.05 from control. ** p<0.01 from control. Histological analysis of colon from control rats (E), and from 40 kDa dCGN-treated rats (F).

Degraded CGN Induced-TNF-α Production by Monocytes In Vitro 
In order to study the capacity of dCGN to stimulate TNF-α production, peripheral blood monocytes were cultivated in the presence of dCGN (0.1 to 1 mg/ml). Very low levels of TNF-α were induced in PBM after stimulation with native CGN (Fig. 2A). Addition of 0.1 mg/ml 10 kDa dCGN resulted in approximately a 60-fold increase in TNF-α production by PBM. This was a dose-dependent effect that reached a 180-fold increase when cells were exposed to 1 mg/ml of 10 kDa dCGN (Fig. 2A). A 250-fold increase in TNF-α production was detected at 1 mg/ml 40 kDa dCGN (Fig. 2A). TNF-α production increased in time reaching a maximum level at 8 hours of culture (Figure 2B). After 24 h, the amount of secreted TNF-α was still one third of the total TNF-α. Lipopolysaccharide (LPS), a known activator of immune cells also induced TNF-α production with similar kinetics as dCGN (Fig. 2B). However, the amount of TNF-α produced by LPS was 4-fold less than the one produced by dCGN and it was not detected after 8 hours of culture (Fig. 2B). Similarly, monocytic THP-1 cells cultivated in the presence of variable concentration of dCGN showed an increase in TNF-α production (Fig. 2C). This increase in TNF-α production was significantly smaller (about 10-fold) than the one presented by PBM (Fig. 2A). No TNF-α was released from THP-1 cells exposed to native CGN (not shown). TNF-α production by THP-1 cells was not dose dependent to the amount of dCGN used. Also there was no difference between the two forms (10 and 40 kDa) of dCGN (Fig. 2C). Interestingly, TNF-α release from THP-1 cells stimulated with dCGN reached a maximum level at 32 h, while stimulation with LPS reached a maximum level at 56 h (Fig. 2D).
10.1371/journal.pone.0008666.g002 Figure 2  Degraded CGN stimulated TNF secretion from monocytes.
Levels of TNF released from peripheral blood monocytes (A–B) and THP-1 cells (C–D) after stimulation with dCGN. A: TNF release induced by native CGN (open bars), 10 kDa dCGN (grey bars), or 40 kDa dCGN (black bars). B: Kinetics of TNF release induced by nothing (control; black diamonds), 0.1 mg/ml 10 kDa dCGN (black triangles), 0.1 mg/ml 40 kDa dCGN (black squares), or 10 µg/ml LPS (open squares). C: TNF release induced by nothing (control; open bars), 10 µg/ml LPS (hatched bars), or increasing concentrations of 10 kDa dCGN (grey bars), or 40 kDa dCGN (black bars). D: Kinetics of TNF release induced by nothing (control; black diamonds), 0.1 mg/ml 10 kDa dCGN (black triangles), 0.1 mg/ml 40 kDa dCGN (black squares), or 10 µg/ml LPS (open squares).

Effect of Native and Degraded CGN on THP-1 Proliferation and Cell Cycle
Preliminary observations by enumeration of THP-1 cells exposed to different concentrations of native and dCGN (10 and 40 kDa) during 2, 5 and 7 days, showed a decline in cell number (data not shown). This suggested that dCGN might cause an alteration in the cell cycle. Cell cycle analysis using flow cytometry showed an accumulation of THP-1 cells in G0/G1 phase, which was associated with a decrease number of cells in the S phase (Fig. 3A). The percentage of cells in the G0/G1 phase was 45.2% for control cells, 62.6% for C10 dCGN (at 2 mg/ml), and 64.2% for C40 dCGN-treated cells (at 2 mg/ml) (Fig. 3B). The effect of dCGN on cell cycle was dose-dependent (Fig. 3B). Neither native nor dCGN had an effect on the number of cells in the G2/M phase (Fig. 3). This effect is not due to cytotoxicity of dCGN even at the highest concentration (i.e. 2 mg/ml) since cell viability was not affected (data not shown).
10.1371/journal.pone.0008666.g003 Figure 3  Degraded CGN induced THP1 cell cycle arrest in G1 phase.
THP-1 cells in exponential growth phase were incubated in the presence or absence of carrageenan for 24 h before being stained with propidium iodide. Cell DNA content was then analyzed by flow cytometry. A: Histograms of cells treated with medium only (control), 10 kDa dCGN (C10), or 40 kDa dCGN (C40). B: Percentage of cells in each phase of the cell cycle when treated with medium only (control), different concentrations of 10 kDa dCGN (C10), of 40 kDa dCGN (C40), or of native CGN (Native).

ICAM-1 Expression Is Induced by Degraded CGN and Is Responsible for Monocytes Aggregation In Vitro 
In order to study the effect of dCGN on the expression of cell surface antigens, PBM and THP-1 cells were incubated for 36 h in the presence and absence of dCGN. The expression of various cell surface molecules was analyzed by flow cytometry as described in materials and methods. Both forms of dCGN clearly stimulated expression of ICAM-1 (CD54) on PBM and THP-1 cells (Fig. 4A). The increase in ICAM-1 expression was higher on THP-1 cells treated with 40 kDa dCGN (Fig. 4B). Another surface antigen, the lymphocyte function-associated antigen 3 (CD58) was slightly reduced on PBM after treatment with 40 kDa dCGN (Fig. 4B). Interestingly, expression of major histocompatibility complex molecules of class I (HLA-ABC) and of class II (HLA-DR), as well as the monocyte marker CD14, seemed to be reduced by treatment with dCGN (Fig. 4B). However, these differences were not statistically significant.
10.1371/journal.pone.0008666.g004 Figure 4  Degraded CGN stimulated ICAM-1 expression in monocytes.
Peripheral blood monocytes (PBM) or THP-1 cells were incubated in the presence or absence of carrageenan for 24 h before being stained for various cell surface antigens. Antigen expression was then analyzed by flow cytometry. A: Histograms of ICAM-1 expression in cells treated with medium only (control), 10 kDa dCGN (C10), or 40 kDa dCGN (C40). B: Fluorescence intensity for expression of the antigens HLA-ABC, HLA-DR, CD14, ICAM-1, and CD58 in cells treated with medium only (control), 10 kDa dCGN (C10), or 40 kDa dCGN (C40). Data are mean +/− SEM.   Treatment with dCGN also induced a strong aggregation of monocytes, detected by phase contrast inverse microscopy (Fig. 5). Although this effect was easily observed in monocytes incubated with the 10 kDa dCGN (Fig. 5B), a more robust cell aggregation was observed in monocytes incubated with the 40 kDa dCGN (Fig. 5C). ICAM-1 has been proposed to be the main adhesion molecule responsible for monocyte aggregation. To confirm this, monocytes were incubated with both types of dCGN in the presence of an anti-ICAM-1 antibody, an anti-CD58 antibody and an isotype control IgG1 antibody. The anti-ICAM-1 antibody effectively blocked the cell aggregates induced by dCGN (Fig. 5D, 5E), strongly suggesting that indeed ICAM-1 is responsible for monocyte aggregation. Both the control IgG1 and the anti-CD58 antibody did not modify monocyte aggregation (data not shown).
10.1371/journal.pone.0008666.g005 Figure 5  Degraded CGN induced monocytes aggregation in vitro.
Monocytes were incubated in the absence (A) or the presence of 1 mg/ml 10 kDa dCGN (B), 1 mg/ml 40 kDa dCGN (C), or 1 mg/ml 40 kDa dCGN plus 2.5 µg/ml anti-ICAM-1 antibody (D). Cells were observed by phase contrast inverse microscopy at 150X magnification. Inserts in A and B show a close up of cells at 300X magnification. E: Number of monocyte aggregates in 24 wells plate of PBM cell cultured with nothing (control), with 10 kDa degraded CGN (C10), or with 40 kDa degraded CGN (C40). Some cultures had also 2.5 µg/ml anti-ICAM-1 antibody. Data are mean +/− SEM.

Degraded CGN Induce an Increase in ICAM-1 and TNF-α mRNA Expression
The increase in surface ICAM-1 expression and TNF-α production by monocytes correlated with an upregulation of mRNA for these molecules. Both 10 kDa and 40 kDa dCGN induced a robust increase in mRNA for both ICAM-1 and TNF-α (Fig. 6). β-actin mRNA levels were not affected by dCGN treatment.
10.1371/journal.pone.0008666.g006 Figure 6  Degraded CGN stimulated ICAM-1 and TNF-α gene expression in monocytes.
Representative samples of RT-PCR analysis showing over expression of ICAM-1 and TNF-α after stimulation of monocytes with 1 g/l of degraded CGN. β-actin expression was used as normalization gene.

Degraded CGN Induce IκB Degradation and NF-κB Activation
The expression of genes encoding for ICAM-1 and TNF-α is controlled by the nuclear factor NF-κB. Site-specific phosphorylation of the inhibitor IκB leads to its degradation by proteasome and to a consequential activation of the NF-κB pathway. Using a reporter plasmid for NF-κB activation, it was confirmed that dCGN induced a strong activation of NF-κB, as reflected by an increase in luciferase activity (Fig. 7A). Both forms of dCGN used induced NF-κB activation in a dose dependent manner. However, the effect was more strongly induced by the 40 kDa dCGN (Fig. 7A). These results were further confirmed by directly detecting NF-κB in the cell nucleus by Western blotting (Fig. 7C) and by FACS (Fig. 7D). These assays also allowed us to determine what NF-κB subunits were activated by dCGN. Both forms (10 or 40 kDa) of dCGN induced activation of the p50 and p65 subunits of NF-κB. This nuclear factor was present in low levels in the cell nucleus and increased considerably after treatment with dCGN. Western blots suggested the the 40 kDa form of dCGN induced a stronger activation of NF-κB (Fig. 7C). A more sentive assay for nuclear factor activation is flow cytometry of nuclei stained with specific antibodies for the nuclear factor of interest. In agreement with the previous data, FACS analysis of nuclei from THP-1 cells showed that there was a basal level of nuclear NF-κB (Fig. 7D). Again, both forms (10 or 40 kDa) of dCGN induced an increase of the p50 and p65 subunits of NF-κB in the nucleus of these cells. The 40 kDa degraded CGN gave a stronger increase of NF-κB (Fig. 7D). These data strongly suggest that the heterodimer p50/p65 is the NF-κB isoform activated by degraded CGN in monocytes. In addition, degradation of the inhibitor IκBα was also observed in cells treated with dCGN (Fig. 7B). No significant IκBα degradation was detected within two hours of dCGN treatment, but IκBα was markedly degraded by four hours of dCGN treatment (Fig. 7B). We focused on IκBα subunit, since it masks the nuclear localisation sequence of p65, it is the most rapidly degraded subunit and the most studied one.
10.1371/journal.pone.0008666.g007 Figure 7  Degraded CGN activated the NF-kB pathway in monocytes.
A: THP-1 cells were transfected with a NF-κB reporter plasmid driving expression of luciferase. Cells were then treated with various concentrations of 10 kDa (triangles), or 40 kDa dCGN (squares). B: THP-1 cells treated with 1 mg/ml of 10 kDa dCGN (C10), or with 1 mg/ml of 40 kDa dCGN (C40) were lysed after various periods of time. Proteins in cell extracts were resolved by SDS-PAGE and then Western blotted for IκBα or α−tubulin as loading control. C: Degraded carrageenans (dCGN) induced activation of NF-κB. THP-1 cells were treated with nothing (control), or with 1 mg/ml of 10 kDa dCGN (C10), or with 1 mg/ml of 40 kDa dCGN (C40) for 30 minutes at 37°C. Nuclei were isolated and lysed. Proteins in nuclear extracts were resolved by SDS-PAGE and then Western blotted for NF-κB p50 subunit (p50) or NF-κB p65 subunit (p65). Lower panels show Western blots of nuclear ERK revealing equivalent amount of protein in each sample. Data are representative of three separate experiments. D: Degraded carrageenan (dCGN) induced activation of NF-κB. Nuclei isolated from THP-1 cells were fluorescence-stained for NF-κB p50 subunit or NF-κB p65 subunit before (filled area) or after cells were treated with 1 mg/ml of 10 kDa dCGN (C10), or with 1 mg/ml of 40 kDa dCGN (C40) for 30 minutes at 37°C. Dashed line corresponds to nuclei stained only with secondary fluorescence antibody. Fluorescence intensity was analyzed by flow cytometry as described.