Skip Navigation


Rheumatology Advance Access originally published online on March 9, 2006
Rheumatology 2006 45(9):1087-1095; doi:10.1093/rheumatology/kel061
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
45/9/1087    most recent
kel061v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (11)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Decker, P.
Right arrow Articles by Rammensee, H.-G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Decker, P.
Right arrow Articles by Rammensee, H.-G.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Monocyte-derived dendritic cells over-express CD86 in patients with systemic lupus erythematosus

Patrice Decker1,, Ina Kötter2, Reinhild Klein2, Beate Berner2 and Hans-Georg Rammensee1

1Institute for Cell Biology, Department of Immunology, University of Tübingen and2Internal Medicine II, University Hospital, Tübingen, Germany.

Correspondence to: Dr Patrice Decker, Institute for Cell Biology, Department of Immunology, Auf der Morgenstelle 15, D-72076, Tübingen, Germany. E-mail: patrice.decker{at}uni-tuebingen.de


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
Objectives. Dendritic cells (DCs) play a key role in regulating immune responses, especially in priming naïve T-cells. Recently, DCs have been suggested to be involved in systemic lupus erythematosus (SLE) development by activating autoreactive T-helper lymphocytes. As a consequence, we compared the activation state of human monocyte-derived DCs (MDDCs) obtained from lupus patients and normal individuals.

Methods. The MDDCs were generated in vitro from blood from healthy donors and lupus patients. Immature and mature MDDCs were analysed by flow cytometry for several cell surface molecules. In parallel, cytokine secretion was determined by ELISA before and after MDDC activation. In each experiment, lupus DCs were compared with normal DCs.

Results. Here, we show for the first time that lupus MDDCs spontaneously over-express CD86 in the absence of any DC activation signal as compared with normal MDDCs (P=0.025). Moreover, activation-induced IL-6 secretion was increased in lupus DCs with high CD86 over-expression as compared with normal DCs (P=0.010). Interestingly, the percentage of MDDCs in lupus preparations is negatively correlated with disease activity scores (SLEDAI; P=0.031).

Conclusions. Lupus MDDCs are pre-activated suggesting that they might be more efficient antigen-presenting cells. This result might partly explain how the peripheral tolerance is broken in SLE.

KEY WORDS: CD86 over-expression, Monocyte-derived dendritic cell, Lupus, Dendritic cell maturation, Peripheral tolerance.


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Systemic lupus erythematosus (SLE) is a systemic autoimmune disease of unknown aetiology with skin, joint, renal, cardiovascular and nervous manifestations. It is characterized by a large amount of circulating autoantibodies (autoAb) with diverse specificities as a result of abnormal activation of autoreactive B and T-helper lymphocytes. Dendritic cells (DCs) play a key role in regulating immune responses, either by priming naïve T-cells [1] or tolerizing autoreactive T-cells [2]. T-cell priming requires mature DCs whereas immature DCs can induce T-cell tolerance. Thus, dysregulation of DC maturation towards chronic DC activation may skew self-antigen presentation from tolerance to autoimmunity. Recently, DCs have been suggested to be involved in SLE development [3] and myeloid/plasmacytoid DCs (mDCs/pDCs) have been found at lower concentrations in patient blood as compared with normal individuals [4]. Since DCs are known to migrate to lymph nodes upon activation (as reviewed in [5]), it has been hypothesized that blood DCs are activated in lupus patients, resulting in low blood cell numbers. As a consequence, it is rather difficult to analyse activation states of blood DCs since activated DCs are supposed to leave the bloodstream. As an alternative, it is possible to use DCs derived in vitro from monocytes. Importantly, monocytes have been shown to differentiate to DCs in vivo [6] in mice. Thus, it might be possible to overcome the problem of migration of activated blood DCs by analysing monocyte-derived DCs (MDDCs) activation in vitro and avoid any resulting bias in data interpretation.

Interestingly, autoAb production in mouse lupus has been shown to be preferentially dependent on CD86 [7]. Although peripheral blood B lymphocytes show an increased CD86 expression in lupus patients [8], no or little difference was observed for CD86 or CD80/CD86 expression, respectively on non-stimulated monocytes between lupus patients and normal donors [9, 10]. Moreover, CD86 expression on lupus MDDCs has not been investigated, except recently [11] where no difference between lupus and normal MDDCs was observed using a different approach and where cytokine secretion was not investigated. Before the latter study was published, we have thus decided to compare the activation state of MDDCs derived from lupus patients and normal donors and we show for the first time that lupus MDDCs mature spontaneously as observed by CD86 over-expression. This result might partly explain how the peripheral tolerance is broken in SLE patients and why autoreactive T-cells are activated.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Donors
Twenty-two SLE patients fulfilling the American College of Rheumatology criteria and 16 normal individuals (3 men/13 women, mean age = 31 yrs), all randomly selected, were involved in this study. Patient characteristics are described in Tables 1 and 2. All samples were collected between 8:00 AM and 1:00 PM and were treated in the same conditions. Experiments with human cells have been approved by the local ethics committee (reference 146/2001V) and samples were obtained after informed consent of the donors. In all patients, the disease activity score (SLEDAI) was determined.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Patient characteristics

 

View this table:
[in this window]
[in a new window]

  		TABLE 2. Patient characteristics

 
Sample preparation
A total of 30–50 ml of heparinized blood was collected for each donor and was immediately used for peripheral blood mononuclear cells (PBMCs) isolation by standard gradient centrifugation with Ficoll-Hypaque (PAA). The PBMCs were washed several times before use. In most experiments, one normal donor was compared with one SLE patient. In a few experiments, two or three patients were compared with one normal donor. In each individual experiment, normal and lupus bloods were treated exactly the same.

MDDC preparation and maturation
The MDDCs were prepared from plastic-adherent PBMCs (1 h adherence step) cultured in the presence of 100 ng/ml granulocyte-macrophage colony-stimulating factor (GM-CSF, Leucomax 400, Sandoz) and 40 ng/ml interleukin (IL)-4 (R&D) for 6–8 days as described previously [12] in order to induce monocyte differentiation, except that the medium used was X-VIVO 15 (BioWhittaker) in the absence of serum. We checked from time to time that adherent cells were CD14-positive cells in both normal and lupus cultures. Immature MDDCs (non-adherent cells) harvested at day 6, 7 or 8 behaved similarly. Importantly, in each individual experiment MDDC cultures were started in parallel with the same PBMC number for both SLE and normal blood and were treated in the same conditions and for the same duration. Immature MDDCs were then analysed by flow cytometry and were then cultured in medium alone or supplemented with classical DC activators: human tumour necrosis factor (TNF)-{alpha} (Pharmingen, 10 ng/ml) and prostaglandin E2 (PGE2, Sigma, 1 µg/ml) or lipopolysaccharide (LPS, Sigma, from Salmonella typhimurium, 2 µg/ml) and human interferon (IFN)-{gamma} (Pharmingen, 5000 U/ml). After 24 h, supernatants were collected and frozen at –80°C until use and replaced by fresh medium. At day 3, cells were harvested and analysed by flow cytometry.

Flow cytometry
Immature MDDCs (non-adherent cells obtained at day 6–8) were analysed by flow cytometry, with a gate according to FSC/SSC parameters, for CD1a, CD11c, CD14, CD86 and human leucocyte antigen (HLA)-DR cell surface expression using specific monoclonal antibodies (mAb) and corresponding isotype controls (Pharmingen) as well as with propidium iodide using a FACScalibur apparatus with CellQuest software (Becton Dickinson). At day 3 of maturation, cells were harvested and DC maturation was checked by flow cytometry by measuring CD86 cell surface expression in order to estimate CD86 up-regulation.

Cytokine production
Cytokine secretion in culture supernatants was measured by sandwich ELISA using mAb pairs or ELISA kit (Pharmingen) specific for TNF-{alpha}, IL-6, IL-8 and IL-10, according to manufacturer's recommendations. Cytokine concentrations were calculated according to standard curves and normalized according to the percentage of MDDCs in the preparation. Cytokine secretion was determined for MDDCs with or without activation signal. The resulting activation-induced secretion was determined using the ratio: concentration upon activation/concentration without activation. In each experiment, the secretion ratio between lupus and normal DCs was calculated for each cytokine.

Statistical analysis
In order to minimize the variation between experiments (differentiation duration, number of PBMCs available), the study was designed as follows. For flow cytometry analysis, cells were always stained with a specific Ab and with the corresponding isotype control. In all cases, when analysing mean fluorescence intensity (MFI) for a given cell surface molecule, the specific signal was calculated by dividing the signal measured with the Ab by the signal measured with the isotype control (MFI Ab/MFI isotype) in order to take into account eventual background differences. We thus focused on expression levels. Then, MFI of lupus and normal MDDCs were compared by pair in each experiment by calculating the ratio: MFI lupus/MFI normal, meaning that we analysed over-expression of the given cell surface molecule in patients as compared with normal individuals. The population of MFI ratios was transformed using decadic logarithm function (to get a normally distributed population) and then analysed by a two-tailed one-sample Student's t-test against the expected value of 0, i.e. log(1), which should be observed if both groups behave similarly. We thus tested the probability to get two groups with similar characteristics. When analysing percentages of cells positive for a given cell surface molecule, the marker was set according to the corresponding isotype control and fixed at 1% of positive cells as a background in all experiments. The difference in the percentages of positive cells between lupus and normal cultures was analysed as described above, using the ratio of percentages (lupus/normal) calculated for each SLE patient–normal donor pair. Difference in the percentages or concentrations of PBMCs and MDDCs between samples from normal donors and lupus patients or difference in the mean CD86 ratio between patient subgroups were analysed as described above. Values were transformed using decadic logarithm function but then analysed by two-tailed two-sample Student's t-tests.

Cytokine concentrations were determined by ELISA and the ratios between the concentration produced by lupus MDDCs and the concentration produced by normal MDDCs was calculated for all cytokines in each individual experiment. The population of cytokine ratios was then transformed using decadic logarithm function and analysed by a two-tailed one-sample Student's t-test against the expected value of 0, i.e. log(1).

Correlations were determined using Spearman correlation test. The Spearman correlation coefficient (R) and the probability (P) are reported.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Comparison of cell content between normal and lupus samples
First of all, the PBMC concentrations (million cells per millilitre of blood) calculated for fresh blood after Ficoll centrifugation were compared between normal and lupus blood. As shown in Table 3, lupus blood contained significantly fewer PBMCs as compared with normal blood (P<0.0001). According to this result, we decided to start with exactly the same PBMC number (instead of the same blood volume) for normal and lupus MDDC cultures in each individual experiment. Importantly, Scheinecker et al. [4] found no difference in the percentage of monocytes among PBMCs between lupus patients and normal individuals, suggesting that starting with the same PBMC number for lupus and normal cultures will not introduce a bias in our experiments. After in vitro differentiation for 6–8 days, immature MDDCs were harvested and analysed by flow cytometry. In all experiments, the DC gate used only contained CD14-negative, but CD11c-positive cells. Interestingly, the numbers of immature MDDCs obtained from both normal and lupus cultures did not differ (Table 3), indicating that the differentiation rate from monocytes is similar in both groups (calculated from the starting PBMC number). It also suggests that if an intrinsic defect is responsible for low myeloid DC numbers in patients, this defect does not depend on differentiation from monocytes. Interestingly, the percentage of immature MDDCs in lupus cultures was negatively correlated with the SLEDAI (P = 0.031). When we compared normal and lupus immature MDDC numbers obtained from the same volume of blood, we observed decreased DC numbers for SLE patients (Table 3) although the difference was not statistically significant (P = 0.065). On the contrary, a clear positive correlation was observed between the PBMC concentration and the immature MDDC concentration (calculated from the starting blood volume), either for the lupus group (Spearman correlation coefficient R = 0.53; P = 0.018; Fig. 1) or for all donors together (R = 0.52; P = 0.002; Fig. 1). Thus, the higher the PBMC concentration, the higher the number of harvested immature MDDCs.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Cell countsa

 

Figure 1
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 1. Correlation between PBMC and MDDC concentrations in 20 SLE patients and 14 normal individuals. PBMC and MDDC numbers were estimated for each donor after isolation by gradient centrifugation or after in vitro differentiation, respectively, and concentrations were expressed in million per millilitre of starting blood. Each dot represents a single lupus patient (squares) or a single normal donor (triangles) analysed for PBMC and MDDC concentrations. The correlations were determined using Spearman correlation test (R = 0.52; P = 0.002 for all donors (solid line) and R = 0.53; P = 0.018 for SLE patients (dashed line)).

 
Lupus immature MDDCs over-express CD86
In each individual experiment, lupus and normal MDDCs were compared. Importantly, it has already been shown that there is no difference for CD86 expression between lupus and normal monocytes [10]. After in vitro differentiation of normal and lupus monocytes for 6–8 days, immature normal and lupus MDDCs were harvested, analysed by flow cytometry and compared for the expression of cell surface molecules. The ratio of the MFI of a given cell surface molecule between lupus and normal MDDCs was calculated in order to get the level of over-expression of this cell surface molecule in lupus patients. Both normal and lupus immature MDDCs were CD14–CD11c+HLA-DRlow with no obvious difference between the two groups. On the contrary, although both normal and lupus immature MDDCs were CD86-positive, a clear difference was observed in the level of CD86 expression. Thus, we observed that CD86 was over-expressed on lupus immature MDDCs in the absence of DC activation signal (as summarized in Fig. 2). An example of the CD86 over-expression by immature lupus MDDCs as compared with immature normal MDDCs is shown in Fig. 3. Whereas both lupus and normal immature MDDCs show a similar staining with the isotype control (as expected), a clear CD86 over-expression is observed with lupus MDDCs as compared with normal MDDCs. Interestingly, the whole MDDC population over-expresses CD86 and not only a sub-population. Thus, most patients (15 out of 20) over-express CD86 (ratio > 1, Fig. 2) and seven patients (35%) have a CD86 expression 50% higher than normal individuals (ratio ≥ 1.5, mean ratio = 2.54, Fig. 2). As a mean, the level of CD86 expression was 55% higher on lupus MDDCs as compared with normal MDDCs and this difference was statistically significant (P = 0.025). In order to prove that analysis was not biased by using ratios, we tested the correlation between CD86 expression and CD86 over-expression on lupus MDDCs. Thus, we could show that the level of CD86 expression on immature lupus MDDCs positively correlates with CD86 over-expression by these cells (R = 0.48; P = 0.033). When we analysed the percentage of CD86-positive MDDCs (percentage ratio between lupus and normal cells), we observed that lupus DC cultures contained more CD86-positive DCs. As a mean, lupus cultures contained 26% more CD86-positive DCs than normal cultures (ratio = 1.26). However, this difference did not reach statistical significance (P = 0.064). Nevertheless, a strong and statistically significant positive correlation was observed between the ratio of CD86 MFI and the ratio of the percentages of CD86-positive DCs (R = 0.73; P = 0.0003). Thus, the higher the CD86 over-expression by lupus MDDCs, the higher the concentration of CD86-positive DCs in lupus cultures as compared with normal cultures. No difference was observed for HLA-DR expression between lupus and normal immature MDDCs, either for the level of expression or for the percentage of positive cells. Moreover, the percentage of propidium iodide-positive MDDCs was similar in lupus and normal cultures. This result, together with the observation that the percentages of immature MDDCs among starting PBMCs are similar in lupus and normal cultures (Table 3), suggests that the survival of lupus DCs is not affected, especially not by the therapy.


Figure 2
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 2. CD86 over-expression in SLE patients as compared with normal individuals in the absence of activation signal. CD86 expression was measured on immature normal and lupus MDDCs by flow cytometry. CD86 over-expression by SLE MDDCs was then determined by calculating the ratio: lupus CD86 MFI/normal CD86 MFI. On the left, the population of over-expression values corresponding to the 20 patients is shown (each dot represents a single SLE patient–normal donor pair; 20 different patients were compared with 14 different normal donors in 14 independent experiments); on the right, the mean over-expression value and the confidence interval (95%) are represented. The horizontal line at y = 1 represents the expected value for CD86 over-expression when the groups compared do not differ (ratio = 1, meaning no over-expression). As observed in most cases, lupus MDDCs over-express CD86 as compared with normal MDDCs (ratio > 1). The patient CD86 over-expression characteristics are as follows: mean over-expression = 1.55; confidence interval (95%) = 1.12–1.98 for n = 20 patients. This mean over-expression by lupus immature MDDCs was significantly different from the expected value as determined by a two-tailed one-sample Student's t-test (P = 0.025).

 

Figure 3
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 3. Example of the CD86 over-expression observed with immature lupus MDDCs as compared with immature normal MDDCs for a single SLE patient–normal donor pair. Immature MDDCs from arthritic patient No. 4 (Table 1) and from the corresponding normal donor used in the same experiment were harvested at day 6 and analysed by flow cytometry using a phycoerythrin-conjugated isotype control or a phycoerythrin-conjugated anti-human CD86 antibody. DCs were gated according to FSC/SSC parameters (DC morphology). All the gated cells were CD14-negative but CD11c-positive cells. (A) Normal MDDCs stained with the isotype control; (B) lupus MDDCs stained with the isotype control; (C) normal MDDCs stained for CD86; (D) lupus MDDCs stained for CD86. The vertical line in dot plots represents the marker for cells positive for phycoerythrin fluorescence and was set according to the isotype control. The percentages of positive cells are indicated. A clear shift is observed for CD86 expression with lupus MDDCs as compared with normal MDDCs.

 
Upon incubation with either TNF-{alpha} and PGE2 or LPS and IFN-{gamma}, both lupus and normal MDDCs matured and up-regulated CD86, as observed by increases in MFI and percentages of positive cells. At day 3 of maturation, CD86 expression reached a similar level in both normal and lupus MDDCs (data not shown). Thus, both normal and lupus MDDCs reached full maturation, indicating that normal immature DCs compensated the lower CD86 expression upon activation. In other words, normal immature MDDCs need an activation signal to express high levels of CD86, whereas lupus immature MDDCs reach high CD86 expression without activation.

Cytokine secretion
Supernatants of DC cultures where harvested 24 h after culture in medium alone or after activation with TNF-{alpha} and PGE2 or LPS and IFN-{gamma} and were tested for the presence of TNF-{alpha}, IL-6, -10 and -8. In each experiment, the secretion ratio between lupus and normal DCs was calculated for each cytokine. This ratio was thus calculated for the secretion with or without activation and for the resulting activation-induced secretion (as defined in the Materials and methods section).

TNF-{alpha} and IL-10 were not detected in the absence of DC activation signal, either with normal or lupus MDDCs (data not shown). Regarding IL-6, a slight secretion was detected in the absence of activation signal with both normal and lupus MDDCs, but at a similar extent (mean concentration = 122.9 pg/ml). On the contrary, a strong IL-8 secretion was observed with both normal and lupus MDDCs without any activation signal (mean concentration = 5.8 ng/ml). Nevertheless, no statistically significant difference was observed between normal and lupus MDDCs. Upon activation, all cytokines were secreted at higher levels (for example, mean IL-6 concentration = 390.0 pg/ml after TNF-{alpha}/PGE2 activation; mean IL-8 concentration = 19.6 ng/ml after TNF-{alpha}/PGE2 activation), but no significant difference was observed between normal and lupus MDDCs regarding the final cytokine concentrations or the activation-induced cytokine concentrations (data not shown). Finally, as shown in Fig. 4 a clear positive correlation was observed between CD86 over-expression and activation-induced IL-6 secretion ratios (lupus vs normal MDDCs; R = 0.70; P = 0.010), indicating that the more lupus DCs over-express CD86, the more they over-induce IL-6 secretion in the presence of TNF-{alpha}/PGE2 as compared with normal DCs.


Figure 4
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 4. The level of over-induction of IL-6 secretion upon activation positively correlates with CD86 over-expression by lupus MDDCs. IL-6 secretion was measured by ELISA for normal and lupus MDDCs without or with activation by TNF-{alpha} and PGE2. Then, the activation-induced IL-6 secretion was calculated as described in ‘Materials and methods’. In each experiment, the activation-induced IL-6 secretion ratio between lupus MDDC secretion and normal MDDC secretion was calculated (as we did for CD86 to calculate over-expression) in order to determine the ‘over-induction’ by lupus cells. The IL-6 ratios were then plotted against the CD86 ratios determined in Fig. 2. The Spearman correlation coefficient (R) and the P-value were calculated. A clear positive correlation was observed between both variables (R = 0.70; P = 0.010). Each dot represents a single SLE patient–normal donor pair; 12 different patients were compared with nine different normal donors in nine independent experiments.

 
Correlations with clinical parameters
Interestingly, a negative correlation was observed between MDDC concentrations (expressed in million cells per millilitre of starting blood) and C3 levels in SLE patients (P = 0.022). Moreover, we found a negative correlation between C3 levels and anti-nuclear antibody (ANA) levels (P = 0.021). Regarding CD86 over-expression, no correlation was observed with SLEDAI, patient age, disease duration or differential blood cell counts. Importantly, there was no correlation between CD86 over-expression and CRP or ESR levels. Likewise, no correlation was observed between CD86 over-expression and neopterin, C3/C4, ANA or anti-DNA autoAb levels.

Interestingly, we have observed a negative correlation between the mycophenolate mofetil (MMF) doses used in patients and ANA levels (P = 0.039) as well as between MMF doses and activation-induced IL-6 secretion ratios (P = 0.020), supporting the use of MMF in lupus therapy.

In order to exclude a role of medication in CD86 levels, we have analysed correlations between CD86 expression or over-expression and the doses of each drug used in SLE patients. Importantly, no correlation was found between CD86 expression and the dose of any drugs. Regarding CD86 over-expression, a negative correlation was observed with MMF: the higher the MMF dose the lower, the CD86 over-expression (P = 0.035). A similar tendency was observed with azathioprine (AZA), although it did not reach significance (P = 0.053). Moreover, when we compared patients treated with MMF and/or AZA and patients who received neither MMF nor AZA, CD86 over-expression was higher in non-treated patients (P<0.001), suggesting that this therapy would rather inhibit CD86 over-expression. On the contrary, a positive correlation was observed between CD86 over-expression and hydroxychloroquine (HCQ) dose (P = 0.045). Similarly, patients receiving HCQ and/or chloroquine (CQ) have a higher CD86 over-expression than patients receiving neither HCQ nor CQ (P = 0.008). No correlation was observed with prednisolone, CQ, cyclosporine A or methotrexate used individually. Thus, two different tendencies were observed: whereas MMF/AZA therapy inhibits CD86 over-expression, CQ/HCQ therapy favours CD86 over-expression. As a consequence, we compared patients treated with HCQ, CQ, MMF and/or AZA to patients who did not receive any of these drugs: no difference in CD86 over-expression was observed between both groups (P = 0.996) and importantly in both groups patients over-express CD86. This result suggests that CD86 over-expression by lupus DCs is not due to the therapy when considering the whole lupus population.


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
We show for the first time that lupus immature MDDCs over-express CD86 as compared with normal MDDCs in the absence of any DC stimulus. Interestingly, lupus MDDCs with the highest spontaneous CD86 over-expressions also over-induce IL-6 secretion upon activation. Importantly, CD86 over-expression cannot be attributed to the therapy. These results support the hypothesis that DCs are over-activated in vivo in SLE patients, resulting in DC migration to lymph nodes and low blood DC concentrations.

We have decided to analyse MDDCs and not blood DCs. Indeed, DCs have been suggested to be involved in lupus development, but have also been shown to circulate at low concentrations in patient blood as compared with normal individuals [3, 4]. It was thus suggested that lupus DCs might be activated in patients, leading to DC migration from blood to lymph nodes and resulting in low blood concentrations. As a consequence, studies on the activation state of blood DCs in vivo might be biased, since over-activated DCs should migrate and leave blood circulation.

Only recently, MDDCs from lupus and normal donors have been compared [11] where CD86 over-expression by lupus MDDCs was not observed. Nevertheless, several reasons can explain the differences observed. First of all, Köller et al. [11] have used monocytes obtained by positive selection using CD14 magnetic beads whereas we used untouched monocytes. It has been observed that monocytes obtained by CD14-positive selection differ from those obtained from untouched monocytes [13, 14]. Thus, soluble factors secreted during DC cultures by CD14-activated monocytes might influence in vitro MDDC differentiation as well as the behaviour of the immature MDDCs obtained after 6–8 days of differentiation. In the latter study, the authors showed that immature MDDCs derived from adherent monocytes or CD14-positive monocytes obtained by positive selection expressed comparable levels of CD86 molecules. This was clear when comparing the percentages of CD86-positive DCs. Nevertheless, when comparing the level of expression, the authors also observed that the CD86 MFI was clearly increased in immature MDDCs obtained from untouched monocytes as compared with MDDCs obtained from positively selected monocytes (MFI = 799 vs 207, respectively). Thus, monocytes obtained by positive selection might be activated in culture and already exhausted after a few days, before differentiating in MDDCs.

The presence of serum in medium during MDDC preparation, as well as the nature of the serum used, have also been shown to influence DC differentiation [15]. Thus, immature MDDCs prepared in the presence of fetal calf serum (FCS) expressed higher levels of CD86 than MDDCs cultured in serum-free medium. This could again explain the differences observed with the study by Köller et al. [11] where medium containing 10% FCS was used. In the latter study, it is possible that the authors could not detect any difference in CD86 expression between MDDCs from lupus patients and MDDCs from normal donors because FCS was used, meaning that CD86 was higher in both DC groups and thus, decreasing the CD86 expression ratio between both groups and then lowering the sensitivity.

Interestingly, we have found that the differentiation rate from monocytes to DCs is similar for both lupus patients and normal individuals (when starting with the same PBMC numbers) indicating that a deficient differentiation process of monocytes is not involved in low myeloid DC numbers in lupus blood. On the contrary, if instead of starting with the same PBMC numbers for normal and lupus DC cultures we start with the same blood volume, the number of MDDCs obtained from lupus blood is lower than the number obtained from normal blood and this observation fits with a previous study reporting low blood DC numbers in SLE patients [4]. In agreement, we have shown that the percentage of MDDCs in lupus cultures is negatively correlated with the SLEDAI. Moreover, although immature MDDCs from lupus patients spontaneously over-express CD86, they need activation to secrete high amounts of cytokines, as do normal MDDCs. These results mean that normal and lupus MDDCs behave similarly with regard to cytokine secretion and thus, the high concentrations of DC-derived cytokines observed in patient blood cannot be explained by intrinsic high cytokine secretion levels by lupus MDDCs. Nevertheless, lupus MDDCs with the highest CD86 over-expressions also show the highest activation-induced IL-6 secretion ratios. Interestingly, IL-6 has been reported to be present at higher concentrations in the sera of patients with active disease [16]. Contrarily to the present study, monocytes from lupus patients have been shown to have a limited differentiation flexibility towards MDDCs [17]. Nevertheless, in the later study authors analysed MDDCs early at day 3 of differentiation in the presence of very low concentrations of GM-CSF and IL-4 and did not analyse CD86 expression.

The CD86 over-expression on lupus MDDCs could be explained by several mechanisms. First, the regulation of CD86 expression at the cell surface might be impaired and might represent an intrinsic defect in SLE. Second, a serum soluble factor might be present in patients and could be involved in DC maturation. Likewise, it could be a soluble factor secreted in lupus DC cultures. In agreement, we have recently shown that nucleosomes, the main auto-antigen in lupus and found as a circulating complex in patients, induce DC maturation via a MyD88-independent pathway upon direct interaction with these cells [18]. This manifests by CD86 up-regulation on DCs and secretion of cytokines known to be associated with lupus. Third, IFN-{alpha} has been detected in the serum of lupus patients [19] and has been shown to induce normal monocytes to differentiate into mature DCs [3] as estimated by CD86 up-regulation. Nevertheless, DC cultures were performed without IFN-{alpha} or lupus serum in the present study but sensitization of monocytes before the cell culture might favour differentiation. We did not measure serum levels of IFN-{alpha} in blood donors and did not try to correlate them with CD86 over-expression.

Due to low DC numbers (as expected), it was not possible to perform functional analysis such as mixed lymphocyte reactions (MLRs). Nevertheless, we have analysed cytokine secretion. Moreover, it is well established that CD86 is an important molecule for T-cell activation and as mentioned above, autoAb production in mouse lupus has been shown to be preferentially dependent on CD86 [7]. MLRs would give information on the capacity of DCs to stimulate T-cells but not on the exact molecular mechanism. It could not be excluded that other proteins influencing T-cell activation are up- or down-regulated at the surface of lupus DCs. Thus, strong T-cell activation in the presence of lupus DCs could not be attributed for sure to CD86. On the contrary, by focusing on CD86 over-expression we highlighted a molecule which might be involved in a mechanism associated with lupus pathogenesis. Importantly, we also found that CD86 over-expression by lupus MDDCs correlates with high activation-induced IL-6 secretion as compared with normal MDDCs. It should be noted that no study has analysed MLRs, DCs phenotype and cytokine secretion together so far with lupus MDDCs. A few have analysed MLRs and MDDC phenotype but not cytokines. We did not analyse MLRs, but we describe DC phenotype and cytokine secretion.

In conclusion, lupus immature MDDCs over-express CD86 furnishing strong signal 2 for efficient T-cell activation. Moreover, high CD86 over-expression by lupus DCs is associated with high activation-induced IL-6 secretion as compared with normal DCs. Thus, lupus MDDCs are probably more efficient antigen-presenting cells that might efficiently co-stimulate T-cells even in the absence of DC activation signal. This mechanism may reflect a state facilitating T-cell activation and thus, might favour the break of peripheral tolerance in patients and might sustain continuous T-cell activation.

Formula


   Acknowledgments
 
We thank Dr Martin Eichner (Department for Medical Biometry, University of Tübingen, Germany) for help in statistical analysis. This work was supported by a fortüne grant (Forschungsprogramm der Tübinger Medizinischen Fakultät, Nr. 1199-0-0) and a Fritz Thyssen Stiftung grant (Az. 10.03.2.123 [EC] ) to P.D.

The authors have no conflicts of interest.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Banchereau J and Steinman RM. (1998) Dendritic cells and the control of immunity. Nature 392:245–52.[CrossRef][Medline]
  2. Lanzavecchia A and Sallusto F. (2001) Regulation of T cell immunity by dendritic cells. Cell 106:263–6.[CrossRef][Web of Science][Medline]
  3. Blanco P, Palucka AK, Gill M, Pascual V, Banchereau J. (2001) Induction of dendritic cell differentiation by IFN-alpha in systemic lupus erythematosus. Science 294:1540–3.[Abstract/Free Full Text]
  4. Scheinecker C, Zwölfer B, Köller M, Manner G, Smolen JS. (2001) Alterations of dendritic cells in systemic lupus erythematosus: phenotypic and functional deficiencies. Arthritis Rheum 44:856–65.[CrossRef][Web of Science][Medline]
  5. Cavanagh LL and Von Andrian UH. (2002) Travellers in many guises: the origins and destinations of dendritic cells. Immunol Cell Biol 80:448–62.[CrossRef][Medline]
  6. Leon B, Martinez del Hoyo G, Parrillas V, et al. (2004) Dendritic cell differentiation potential of mouse monocytes: monocytes represent immediate precursors of CD8– and CD8+ splenic dendritic cells. Blood 103:2668–76.[Abstract/Free Full Text]
  7. Nakajima A, Azuma M, Kodera S, et al. (1995) Preferential dependence of autoantibody production in murine lupus on CD86 costimulatory molecule. Eur J Immunol 25:3060–9.[Web of Science][Medline]
  8. Folzenlogen D, Hofer MF, Leung DY, Freed JH, Newell MK. (1997) Analysis of CD80 and CD86 expression on peripheral blood B lymphocytes reveals increased expression of CD86 in lupus patients. Clin Immunol Immunopathol 83:199–204.[CrossRef][Web of Science][Medline]
  9. Garcia-Cozar FJ, Molina IJ, Cuadrado MJ, Marubayashi M, Pena J, Santamaria M. (1996) Defective B7 expression on antigen-presenting cells underlying T cell activation abnormalities in systemic lupus erythematosus (SLE) patients. Clin Exp Immunol 104:72–9.[CrossRef][Web of Science][Medline]
  10. Liu MF, Li JS, Weng TH, Lei HY. (1999) Differential expression and modulation of costimulatory molecules CD80 and CD86 on monocytes from patients with systemic lupus erythematosus. Scand J Immunol 49:82–7.[CrossRef][Web of Science][Medline]
  11. Köller M, Zwölfer B, Steiner G, Smolen JS, Scheinecker C. (2004) Phenotypic and functional deficiencies of monocyte-derived dendritic cells in systemic lupus erythematosus (SLE) patients. Int Immunol 16:1595–604.[Abstract/Free Full Text]
  12. Bender A, Sapp M, Schuler G, Steinman RM, Bhardwaj N. (1996) Improved methods for the generation of dendritic cells from nonproliferating progenitors in human blood. J Immunol Methods 196:121–35.[CrossRef][Web of Science][Medline]
  13. Schutt C, Ringel B, Nausch M, et al. (1988) Human monocyte activation induced by an anti-CD14 monoclonal antibody. Immunol Lett 19:321–7.[CrossRef][Web of Science][Medline]
  14. Breckpot K, Corthals J, Heirman C, et al. (2004) Activation of monocytes via the CD14 receptor leads to the enhanced lentiviral transduction of immature dendritic cells. Hum Gene Ther 15:562–73.[CrossRef][Web of Science][Medline]
  15. Duperrier K, Eljaafari A, Dezutter-Dambuyant C, et al. (2000) Distinct subsets of dendritic cells resembling dermal DCs can be generated in vitro from monocytes, in the presence of different serum supplements. J Immunol Methods 238:119–31.[CrossRef][Web of Science][Medline]
  16. al Janadi M, al Balla S, al Dalaan A, Raziuddin S. (1993) Cytokine profile in systemic lupus erythematosus, rheumatoid arthritis, and other rheumatic diseases. J Clin Immunol 13:58–67.[CrossRef][Web of Science][Medline]
  17. Steinbach F, Henke F, Krause B, Thiele B, Burmester GR, Hiepe F. (2000) Monocytes from systemic lupus erythematosus patients are severely altered in phenotype and lineage flexibility. Ann Rheum Dis 59:283–8.[Abstract/Free Full Text]
  18. Decker P, Singh-Jasuja H, Haager S, Kötter I, Rammensee HG. (2005) Nucleosome, the main autoantigen in systemic lupus erythematosus, induces direct dendritic cell activation via a MyD88-independent pathway: consequences on inflammation. J Immunol 174:3326–34.[Abstract/Free Full Text]
  19. Hooks JJ, Moutsopoulos HM, Geis SA, Stahl NI, Decker JL, Notkins AL. (1979) Immune interferon in the circulation of patients with autoimmune disease. N Engl J Med 301:5–8.[Abstract]
Submitted 28 July 2005; revised version accepted 31 January 2006.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Immunol.Home page
J. C. Crispin, M. Oukka, G. Bayliss, R. A. Cohen, C. A. Van Beek, I. E. Stillman, V. C. Kyttaris, Y.-T. Juang, and G. C. Tsokos
Expanded Double Negative T Cells in Patients with Systemic Lupus Erythematosus Produce IL-17 and Infiltrate the Kidneys
J. Immunol., December 15, 2008; 181(12): 8761 - 8766.
[Abstract] [Full Text] [PDF]

Home page
 Rheumatology (Oxford)Home page
N. Mozaffarian, A. E. Wiedeman, and A. M. Stevens
Active systemic lupus erythematosus is associated with failure of antigen-presenting cells to express programmed death ligand-1
Rheumatology, September 1, 2008; 47(9): 1335 - 1341.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
45/9/1087    most recent
kel061v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (11)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Decker, P.
Right arrow Articles by Rammensee, H.-G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Decker, P.
Right arrow Articles by Rammensee, H.-G.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?