Published online before print March 23, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Departments of Immunology and
Pulmonary and Critical Care Medicine, Erasmus MC, Rotterdam, The Netherlands
1 Correspondence: Department of Pulmonary and Critical Care Medicine, Erasmus MC, Dr. Molewaterplein 50, 3015GE Rotterdam, The Netherlands. E-mail: t.nikolic{at}erasmusmc.nl
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) is important for the initiation and progression of autoimmunity and the final destruction of ß-cells. Previous studies suggested that an aberrant development of DC and M is related to their pathogenic function. To study this in vivo, we investigated NOD mouse monocytes, the direct precursors of DC, and M. The recently described discrimination between immature (Ly-6Chigh) and mature (Ly-6Clow) monocytes enabled us to investigate the apportioning between blood monocyte populations in the NOD mouse, which had an abnormally high number of mature monocytes in circulation, and this phenomenon appeared to be intrinsic to the NOD background, as nonobese resistant (NOR) and NOD-H2b mice also showed this altered balance. After depletion by apoptosis-inducing liposomes, the reappearance and transition of immature-to-mature monocytes had similar kinetics as control mice but led again to the presence of a larger, mature monocyte compartment in the blood. In addition, although monocytes from C57BL mice down-regulated their capability to adhere to fibronectin and intercellular adhesion molecule-1 upon maturation, the mature NOD monocytes retained their high adhesion capacity, characteristic of immature cells. Furthermore, both monocyte subpopulations of NOD mice showed enhanced differentiation into M-like F4/80high cells in vitro. In conclusion, mice with the NOD background have raised numbers of mature monocytes in the circulation and a proinflammatory, M-directed monocyte development.
Key Words: autoimmunity • development • antigen-presenting cells • phagocytes
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) are closely related to the initiation and progression of autoimmune diabetes. Significant accumulation of DC and M around islets at 4–5 weeks of age precedes the peri-insular concentration of lymphocytes and their subsequent infiltration in the islets [1
, 2
].
Monocytes are direct precursors of DC and M [3
4
5
]. It has been generally accepted that the majority of DC and M in nonlymphoid tissues originate from blood monocytes, especially in inflammation [6
, 7
]. Mouse blood monocytes can be separated phenotypically and functionally into at least two subsets [8
, 9
]. Monocytes that have recently emigrated from the bone marrow (BM) and express high levels of the Ly-6C molecule represent the first subset. These Ly-6Chigh, immature monocytes are readily attracted to sites of inflammation, have the potential to develop into DC or F4/80+ inflammatory M [5
, 9
], and coexpress CD62L/L-selectin and CC chemokine receptor 2 (CCR2), which enables the cells to be attracted toward inflammation-induced CC chemokine ligand 2 (CCL2)/monocyte chemoattractant protein-1 (MCP-1) signals [8
, 10
]. The second subset of blood monocytes is generated through a maturation step from the immature monocytes, which is marked by a reduction of the surface expression of Ly-6C and an increase in expression of CD43 [9
]. These Ly-6Clow monocytes, which express a low level of CD11c and a high level of CXC 3 chemokine receptor 1 (CX3CR1; fractalkine receptor), migrate to nonlymphoid organs in the absence of inflammation and have been proposed to be the precursors for the steady-state pool of peripheral DC and M [8
].
In the study that described two circulating monocyte subsets in the mouse [8 ], authors have suggested that mouse CX3CR1low (Ly-6Chigh) monocytes correspond to human CD14++CD16–monocytes, and the CX3CR1high (Ly-6Clow) monocytes correspond to the CD14+CD16+ human monocyte population, based on the cell phenotype as a standard. Others have used different criteria to subdivide circulating human monocytes as well, such as the ability to adhere to fibronectin [11 ]. The fibronectin-adherent, so-called proinflammatory "P-monocytes," constitute 20–30% of the circulating monocyte population, express high levels of adhesion molecules, and have an enhanced chemotactic responsiveness, phagocytosis capacity, and proinflammatory cytokine production capability [12 , 13 ]. We found that fibronectin-adherent monocytes are equally present in the CD14+CD16+ and CD14++CD16– populations [14 ], indicating that the divisions based on the fibronectin adherence and on the CD16 expression are probably not identical.
Different pathological conditions appear to correspond with changes in the composition of the monocyte pool. In humans, the CD14+CD16+ monocyte population expands greatly in acute and chronic infections, systemic inflammatory syndromes, AIDS, and renal failure [15 16 17 18 19 20 ], and other conditions stimulate the prevalence of the CD14++CD16neg population [21 ]. Also in mice, acute and chronic infection models revealed a shift in the monocyte balance in the circulation [9 , 22 ]. However, normal numbers of CD14/CD16-defined monocyte populations were found in T1D patients and in multiple sclerosis patients [14 , 23 ], and we found a raised number of fibronectin-adherent monocytes in T1D patients as compared with healthy controls [14 ]. Furthermore, we previously reported a hampered capability of monocytes of T1D patients to develop during an overnight culture into DC-like cells [24 ].
To investigate a putative reflection of the autoimmune-prone status of the NOD mice in their blood monocyte profile and compare this with published data for human T1D monocytes, we applied a recently developed methodology [9
] to study NOD mouse monocytes. Here, we describe an excess of mature Ly-6Clow monocytes in the blood of mice with the NOD background, which have an abnormal, high fibronectin-adhesive capacity. In addition, we found that NOD monocytes have an enhanced capability to mature and preferentially acquire a M-like phenotype.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antibodies
Specifications of monoclonal antibodies (mAb) and fluorescent conjugates against surface markers used in this study are listed in Table 1
. Directly conjugated isotype control mAb were purchased from BD Pharmingen. Anti-rat immunoglobulin G (IgG) conjugated with FITC or R-PE (mouse-absorbed; GaRa-FITC or GaRa-PE) were purchased from Caltag Laboratories. Biotinylated antibodies were detected by APC-conjugated streptavidin, purchased from BD Pharmingen.
|
View this table: [in a new window] |
Table 1. mAb Used in This Study
|
Preparation of leukocytes, flow cytometry, or cell sorting
Mice were killed by CO2 exposure. Blood was obtained by heart puncture and collected in heparin-coated tubes. For flow cytometry, erythrocytes were eliminated using BD Pharmingen lysing buffer. For sorting, blood samples were treated with a sterile ammonium chloride lysing solution. Leukocytes were subsequently washed by centrifugation at 1500 rpm for 5 min in PBS, pH 7.8, containing 0.5% bovine serum albumin (BioWhittaker, Verviers, Belgium), resuspended, and counted in a Bürker hemocytometer.
For phenotypic analysis, aliquots of 0.5–1 x 106 cells were pipetted into 96-microwell plates (round-bottom, Nunc, Denmark) and incubated with the prepared mix of mAb. Each incubation step was performed at room temperature for 10 min. Cells were analyzed using a FACSCalibur equipped for four-color flow cytometry, and up to 5 x 105 events were obtained. Data were analyzed using CellQuest software (Becton Dickinson, Sunnyvale, CA).
Cell sorting was performed on a FACSDiva by applying a protocol described previously [3 ]. Briefly, lysed blood samples (pooled from five to 10 mice) were washed in sterile PBS, supplemented with 5% heat-inactivated fetal calf serum (FCS; BioWhittaker) and antibiotics (60 mg/ml penicillin and 100 mg/ml streptomycin; further referred to as sorting buffer), and incubated for 30 min on ice with the mix of CD11b, Ly-6C, and Ly-6G antibodies (Table 1) . Subsequently, cells were washed with sorting buffer and filtered over a 30-µm sieve (Polymon PES, Kabel, Amsterdam, The Netherlands) to avoid clogging of the nozzle. After sorting, the purity of the cell suspensions was analyzed by rerunning sorted samples. Purity exceeded 95%, unless stated otherwise. Cells were kept at 4°C throughout the staining and sorting procedure.
Adhesion test
The adhesion capacity test was performed as described previously [28
]. Briefly, sorted monocytes were suspended in RPMI-1640 culture medium (BioWhittaker) containing antibiotics supplemented with 1% heat-inactivated FCS and then plated at a density of 0.2 x 105 cells per chamber on coated Chambertek glass slides (Nalge Nunc International, Naperville, IL), previously coated with 10 µg/ml fibronectin or 10 µg/ml intercellular adhesion molecule-1 (ICAM-1; Sigma, Steinheim, Germany). After 60 min incubation at 37°C, cells were washed with PBS and fixed with 4% paraformaldehyde (Sigma), supplemented with 3% glucose. Cells were permeabilized using 0.5% Triton X-100 (Sigma Chemical Co., St. Louis, MO) and stained with 0.1 µg/ml FITC-labeled phalloidin (Sigma Chemical Co.) for 30–45 min. After washing and mounting the slides, the cells were counted using a fluorescence microscope. Adhesion was expressed as the number of cells in 10 high-power fields (hpf) at 200x magnification.
Culture of sorted cells
For analysis of in vitro differentiation capacity of monocytes, sorted cells were seeded into 96-well culture plates at a concentration of 1 x 106/ml in RPMI-1640 medium supplemented with antibiotics and 10% FCS. Cells were incubated at 37°C with 5% CO2, stimulated or not with 100 ng/ml lipopolysaccharide (LPS; Sigma). After 24 h incubation, cells were used for phenotypic analysis.
Statistical analysis
To determine differences between the groups, data were compared by a two-tailed Student’s t-test using the SPSS software package. Results are presented as the mean ± SEM, unless indicated differently.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (59K): [in a new window] |
Figure 1. Characterization and phenotypic analysis of blood monocytes in the NOD mouse. (A) NOD peripheral blood monocytes can be distinguished as side-scatter (SSC)lo and CD11bhi cells. Expression of Ly-6C further separates monocytes into two populations, as seen from the biphasic profile of the histogram. (B) Although Ly-6C levels are lower on NOD monocytes, Ly-6C separation identifies the same two monocyte populations as in the C57BL mice, as judged by CD11c, CD43, and CD62L expression. However, NOD mice have a lower percentage of Ly-6Chi monocytes. (C) Additional markers tested, differentially or similarly expressed between two monocyte subsets, show similar expression between NOD and C57BL mice within a particular population. Difference in CD31 is strain-related, and F4/80 and CD45RB are not statistically significant (Student’s t-test). Dot-plots and histograms show a representative staining of a minimum of three independent stainings for each marker. FSC, Forward-scatter; FcR , Fc receptor for IgG.
|
|
View this table: [in a new window] |
Table 2. Phenotype of Monocyte Subpopulations
|
marker F4/80 had a reproducibly higher surface expression on both subsets of NOD monocytes. In addition, CD31 expression was significantly higher in NOD mice when compared with C57BL (Table 2)
but not when compared with BALB/c mice (not shown). Both monocyte subtypes similarly expressed all other markers tested in different mouse strains (Fig. 1C
and Table 2
).
Mice with the NOD background have more mature (Ly-6Clow) monocytes in the blood
Although the phenotype of the NOD monocyte subsets did not differ from that of C57BL mice, we noticed that the frequency of the Ly-6Chigh monocytes was repeatedly lower in NOD mouse blood when compared with C57BL blood (Fig. 1B)
. Correspondingly, Ly-6Clow monocytes in the NOD blood formed the major monocyte population.
Is this over-representation of mature monocytes in the blood specific to the NOD mouse? We determined the frequency of monocyte subsets in several mouse strains (Fig. 2A ). Between 58% and 67% monocytes were of the Ly-6Chigh type in the strains of mice unrelated to the NOD. In contrast, in the strains with the NOD background, Ly-6Chigh monocytes never exceeded 44% of the total monocyte pool. Correspondingly, the higher frequency of Ly-6Clow monocytes was found in healthy and diabetic NOD mice, as well as NOR and NOD-H2b mice (55–65% for the NOD-related strains vs. 30–35% for the NOD-unrelated strains; Fig. 2A ). Hence, all mice with the NOD background had a reversed ratio of monocyte subsets in favor of the Ly-6Clow cells.
 View larger version (31K): [in a new window] |
Figure 2. Different balance between Ly-6Chi and Ly-6Clo monocytes in the blood of mice with a NOD genetic background. (A) Analysis of the frequency of blood monocytes in different mouse strains revealed a specific shift toward Ly-6Clo monocytes in all mouse strains with the NOD genetic background. (B) Calculation of the absolute cell number showed that a similar number of Ly-6Chi monocytes are found in the NOD as in the C57BL, the C3H, and the BALB/c mice. In contrast, the number of Ly-6Clo monocytes in the NOD mouse blood is significantly higher than in any control mouse strain tested. Graphs represent average values ± SEM calculated from 26 C57BL mice, 20 NOD mice, and six to nine mice of other mouse strains.
|
Ly-6Clow monocytes from NOD mice display unusually high adhesion to fibronectin and ICAM-1, typical for Ly-6Chigh monocytes
Adhesion capacity is also used as a discriminating property of different monocyte populations [11
]. To evaluate the ability of the two NOD monocyte subsets to adhere to fibronectin or ICAM-1, we sorted Ly-6Chigh and Ly-6Clow monocytes from the mouse blood and tested their adhesiveness to these compounds in vitro.
As shown in Figure 3 , Ly-6Chigh monocytes isolated from NOD and C57BL adhered to fibronectin and ICAM-1 equally. In contrast, a much lower number of Ly6Clow monocytes from C57BL mice adhered to these compounds; hence, the maturation of monocytes in the circulation leads to a decline in the ability to attach to these integrin ligands. It is interesting that although expression of several integrins did not differ between corresponding NOD and C57BL monocyte populations (Fig. 1C and Table 2 ), Ly-6Clow monocytes in NOD mice preserved the high adhesion capacity to fibronectin and ICAM-1, comparable with that of Ly-6Chigh monocytes (Fig. 3) .
 View larger version (16K): [in a new window] |
Figure 3. High in vitro adherence of sorted Ly-6Clo monocytes from the NOD mouse. Blood monocytes were sorted and put to adhere to ICAM-1 or fibronectin for 60 min. Subsequently, cells were washed, stained with phalloidine, and quantified on the fluorescence microscope. A high number of Ly-6Chi monocytes from both mouse strains adhered to both substrates. In contrast, although Ly-6Clo monocytes from the C57BL mice down-regulated the adhesion capacity upon maturation, the Ly-6Clo monocytes from the NOD mice failed to do so. Graphs represent an average ± SEM from three independent pools of a minimum of five mice per pool, obtained in two independent experiments.
|
-like cells [3
4
5
]. To assess the capacity of the two subpopulations of monocytes to differentiate spontaneously in vitro, we sorted and incubated them in culture fluid for 24 h without adding additional cytokines.
The overnight culture of sorted, immature Ly-6Chigh monocytes from the C57BL mouse induced an up-regulation of the CD43 molecule, characteristic of mature monocytes, on the majority of the cells (Fig. 4A
). In addition, a fraction of the cells expressed F4/80 at high levels, typical for M. Unlike C57BL cells, NOD Ly-6Chigh immature monocytes did not acquire CD43 in vitro, but all cells had strongly up-regulated the F4/80 molecule (note that the NOD mouse fresh blood monocytes had somewhat higher F4/80 expression). Hence, a significantly higher percent of Ly-6Chigh monocytes of NOD mice spontaneously became F4/80high M-like cells (P<0.05; Fig. 4B
). Addition of LPS increased the percentage of F4/80high cells in both cultures.
 View larger version (24K): [in a new window] |
Figure 4. Ly-6Chi monocytes from NOD mice display a different spontaneous maturation. (A) Spontaneous maturation of Ly-6Chi monocytes upon in vitro cultivation for 24 h with or without LPS proceeds via different phenotypic stages in NOD mice when compared with C57BL. Dot-plots show a representative staining of Ly-6Chi monocytes prior to and after the sorting, as well as after 24 h incubation alone or with LPS. Numbers represent percent of cells in the quadrant. (B) Frequency (±SEM) of F4/80hi cells derived from Ly-6Chi monocytes in vitro. Data are derived from four independent samples per mouse strain obtained from a pooled blood of five to six mice per strain in three independent experiments.
|
-like phenotype. The DC-like cells were also CD11c+MHCIIhigh, and M-like cells were CD11clowMHCIIlow, as previously reported [4
, 5
]. Although all three populations were present in NOD as well as C57BL samples, they had different frequencies (Fig. 5B)
. More than half of the Ly-6Clow monocytes remained unchanged in the cultures from C57BL mice, and in the NOD cultures, only less than 20% of cells preserved the phenotype of Ly-6Clow monocytes (P<0.001, NOD vs. C57BL). The percentages of DC-like cells were similar in NOD and C57BL cultures, in nonstimulated and the LPS-stimulated samples, so the NOD monocytes did not mature excessively into DC. Instead, the NOD monocytes differentiated more readily and predominantly into M-like cells (P<0.05; Fig. 5B
). With LPS as an additional stimulus, the enhanced differentiation into M-like cells was evident in cultures of both mouse strains.
 View larger version (45K): [in a new window] |
Figure 5. Predominant maturation of Ly-6Clo monocytes from NOD mice into M. (A) Culture of isolated, mature Ly-6Clow monocytes for 24 h in the presence or absence of LPS results in three distinct populations of cells: monocytes (Mo; CD43medF4/80lo), DC (CD43loF4/80med), and M (CD43hiF4/80hi). These populations are phenotypically similar in cultures from NOD and C57BL mice but occur in different frequencies, as marked on representative dot-plots by ellipses ("gates"). (B) The average frequency ± SEM of cells phenotypically defined as monocytes (light, shaded bars), DC (dark, shaded bars), and M (solid bars) points to an increased, spontaneous maturation of the NOD Ly-6Clow monocytes into F4/80hi M in vitro. Data are calculated from the samples obtained in the experiments described in the legend of the Figure 4
. Pvalues are probabilities derived from the Student’s t-test.
|
-like phenotype.
Normal restoration and transition time of monocytes in the NOD mice upon in vivo challenge
To follow the monocyte maturation in vivo, we made use of clodronate-loaded liposomes (lip-CL2MDP). A single injection of lip-CL2MDP caused an almost complete depletion of monocytes (SSClowCD11bhigh cells) from the blood within the first 18 h in NOD and C57BL mice (Table 3 ). Apparently, the NOD monocytes were equally able to phagocytose and fragment lip-CL2MDP as the C57BL monocytes. At approximately 48 h postinjection, monocytes started to reappear in the blood of NOD and C57BL mice, and the total number of monocytes did not differ significantly between these two mouse strains (Table 3) .
|
View this table: [in a new window] |
Table 3. Return of Blood Monocytes after i.v. Depletion with lip-CL2MDP
|
 View larger version (15K): [in a new window] |
Figure 6. Normal restoration and similar kinetics of the monocyte return after depletion with lip-CL2MDP. Numbers of Ly-6Chi (A) and Ly-6Clo monocytes (B) in the NOD ( ) and the C57BL ( ) circulation are shown at different time-points after an i.v. injection of lip-CL2MDP (clo-lipo). Upon depletion of all monocytes, C57BL and NOD mice showed similar kinetics in restoration of Ly-6Chi and Ly-6Clo monocyte populations. Data represent the average frequency ± SD of a minimum of five mice for each time-point for each mouse strain.
|
Before application of lip-CL2MDP, the frequency of Ly-6Clow monocytes in the spleen of NOD mice was lower than in the C57BL mice (P<0.02; Table 4 and Fig. 7A and 7B ) but similar to other tested mouse strains. In the peritoneal cavity, NOD mice and C57BL mice had similar monocyte frequency before depletion (Table 4) . One week after depletion, Ly-6Clow monocytes returned to the steady-state point in the spleen and peritoneal cavity (Table 4) and importantly, with similar kinetics in NOD and C57BL mice (not shown). Hence, in the absence of inflammation, we found no sign of a reduced efflux of monocytes from the circulation to the periphery in NOD mice.
|
View this table: [in a new window] |
Table 4. Frequency of Ly-6Clow Monocytes in Different Compartments before and after Depletion
|
 View larger version (21K): [in a new window] |
Figure 7. Different balance between mature (Ly-6Clow) monocytes and mature myeloid cells (DC+M) in the spleen of NOD mice. (A) The average frequency ± SEM of monocytes (Mo; light, shaded bars) and mature myeloid cells (DC+M; dark, shaded bars) within the total leukocyte pool of the spleen from different mouse strains was determined by phenotypic analysis. (B) P values derived from the statistical comparison (Students’s t-test) among all mouse strains for Mo or DC + M. (C) To compensate for the large interstrain variation, the monocyte:mature cell ratio was calculated. In this case, a significantly lower ratio was found in NOD mice as compared with all other mouse strains. Comparison of ratios between other mouse strains did not result in P values lower than 0.1. Data are calculated from 13 C57BL mice, 19 NOD mice, and five to eight mice of other mouse strains.
|
and DC) in different mouse strains (Fig. 7A)
and performed a detailed statistical analysis. As already mentioned, C57BL mice had higher monocyte frequency than any other mouse strain (Fig. 7B)
. In the case of M + DC, C57BL mice but also NOD and NOR mice had a significantly higher M + DC frequency than BALB/c or C3H mice (Fig. 7B)
. Hence, the strains unrelated to the NOD had the monocytes and M + DC values in balance; both were high or low. In contrast, there was a disproportion in the monocytes and M + DC values in the NOD and partially in NOR mice (Fig. 7A)
. Therefore, we calculated the monocyte:M + DC ratio in the spleen (Fig. 7C)
. Indeed, the ratios were similar in C57BL, C3H, and BALB/c mice. The NOR mice were in between other mouse strains, and NOD mice had a significantly lower ratio (more mature cells than monocytes) than any other mouse strain. This indicated that the balance between monocytes and their progeny in the spleen is in NOD mice, also tipped toward the more differentiated cells. |
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In this study, we found that NOD mice display an altered balance between Ly-6Chigh immature monocytes and Ly-6Clow mature monocytes; i.e., the mice contained an abnormally high number of circulating, mature monocytes. This phenomenon appeared to be intrinsic to the NOD background, as NOR and NODH2b mice also showed this altered balance. In addition, we found more DC and M in the NOD spleen relative to monocytes when compared with other mouse strains. When isolated from the blood, NOD monocytes demonstrated an increased tendency to mature into M spontaneously, a feature found in control monocytes only after stimulation with LPS. In addition, although in vivo depletion of the blood monocytes by lip-CL2MDP did not reveal significant differences in the release from the BM or the transition time of immature-to-mature NOD monocytes, the shifted monocyte ratio was restored rapidly in the NOD mice after depletion. These observations support a view that mice with a NOD background show a skewing of cells of the monocyte lineage toward more differentiated cells in the circulation and the periphery.
The reported finding here of an enlarged subpopulation of mature Ly-6Clow monocytes in the NOD mouse circulation raises the question of its functional significance for the aberrant development of islet autoimmunity in the NOD mouse and the mechanism(s) that leads to the shift toward the more mature forms.
The true nature and function of the mature Ly-6Clow monocytes in the circulation are not yet clear. The cells correspond to the Gr1–CCR2–CX3CR1high cells, which have been proposed to represent precursors for resident M and DC [8
]. Indeed, mouse Ly-6Cmed monocytes (included in the Ly-6Clow population in our gating) have been found to contain direct precursors for DC, which upon antigen uptake, migrate to lymph nodes [29
]. It is interesting that MCP-1R/CCR2 (CCR2–/–) deficient mice used in the latter study show a shifted monocyte ratio toward mature cells, similar to that found by us in the NOD mouse. Moreover, MCP-1/CCL2-deficient mice have a similarly shifted monocyte ratio (Douglas A. Drevets, Oklahoma University Health Sciences Center, Oklahoma City, OK, personal communication). These observations point to a defective CCR2/MCP-1 signaling as a potential cause for the shifted monocyte ratio in the NOD mouse. Indeed, we recently observed a deficient chemotactic response of NOD leukocytes toward CCL2/MCP-1 [28
]. However, even if true, CCR2 signaling probably is not responsible for all functional aberrations of the NOD monocytes reported here, such as the increased adherence to fibronectin or the poor migratory response to inflammatory stimuli (including MCP-1), which we reported. Moreover, there are no reports of autoimmune processes in the CCR2–/– and MCP-1 (CCL2)-deficient mice. However, as NOD and autoimmunity-prone SJL mice (not shown) have a shifted ratio toward mature monocytes, autoimmunity and the shifted ratio in circulating monocyte subpopulations might be linked indirectly.
The Ly-6Clow monocyte population has been proposed to represent the mouse equivalents of human CD14+CD16+ monocytes [8 ]. In humans, these monocytes are generally considered to act as an important proinflammatory effector subset based on their inflammation-related amplification in the blood and high-level production of proinflammatory cytokines upon stimulation [30 ]. Hence, a parallel can be drawn between the enlarged CD16+ pool during acute and chronic inflammations in the human [15 16 17 18 19 20 ] and the enlarged pool of Ly-6Clow monocytes in the NOD mouse that suffers from various (autoimmune) chronic inflammations. This further implies that the imbalance toward mature forms of circulating monocytes may relate to the proneness to develop (autoimmune) chronic inflammations rather than to the presence of such inflammations per se.
Is the higher percentage of fibronectin-adherent blood monocytes in the NOD mouse relevant for the autoimmune process? If taken irrespectively of degree of Ly-6C expression, we also found a raised adhesion to fibronectin of monocytes in patients with T1D [14 ], as reported here for the NOD mouse. However, as in the case of the increased Ly-6Clow monocyte pool, our data do not provide a formal proof of a causal link between fibronectin adherence and the autoimmune pathogenesis.
Also, the reasons for the increased adherence of the NOD monocytes to fibronectin are not obvious from our data. We found a similar expression of adhesion molecules and several chemokine receptors in NOD as compared with C57BL mice. Perhaps the modified Ly-6C molecule in myeloid cells of the NOD mouse plays a role in this abnormality [31 ]. Cross-linking of Ly-6C molecules on the surface of T cells induces integrin expression and has been associated with cell adhesion [32 ]. Expression of integrins involved in fibronectin and ICAM-1 binding were not modified in the NOD monocytes. Still, the recombination in the Ly-6C gene might have changed the function of this molecule in such a way that it aberrantly influences the activation of integrins and not their expression.
Direct comparison of our findings in the mouse with the human monocytes points to the possibility that the Ly-6Chigh/Ly-6Clow subdivision of mouse monocytes might not correspond completely to the CD14/CD16-based division in the human.
The Ly-6Chigh monocytes from C57BL mice have a selective capability to adhere to fibronectin and ICAM-1. In contrast, CD16– and CD16+ human monocytes similarly adhere to fibronectin [14 ]. Discrepancy between the populations is also present in autoimmune situation: The frequency of CD16+ circulating, mature monocytes in human cases was not raised [14 ], and we here show an increased presence of Ly-6Clow monocytes in the NOD mouse blood. Alternatively, homology can be proposed between the mouse Ly-6Chigh monocytes and the P-monocytes in humans [11 ], as they both have a clearly raised ability to adhere to fibronectin. However, the fibronectin-adhering monocytes in the NOD mouse blood were not only in the Ly-6Chigh population. Therefore, the overlap between the Ly-6Chigh and the P-monocytes also might not be absolute. Taken together, a word of caution is necessary in trying to define and compare monocyte subpopulations among species; the functional flexibility of the cells might make such a definition troublesome.
With regard to the capability of blood monocytes of the NOD mouse to differentiate into cells with a DC- or M-like phenotype, we observed a tendency of immature and mature monocytes to differentiate in vitro preferentially into cells with a M-like phenotype, i.e., F4/80highCD11clowMHCIIlow cells. This strengthens our previously expressed view, based on in vitro development of BM precursors, that a differentiation into the DC direction is hampered in NOD mice and skewed into the M direction [33
]. This abnormally skewed production of M-like cells could be a result of dysfunctional signaling pathways, such as the hyperactivity of the nuclear factor-
B and extracellular signal-regulated kinase-1/2 pathways, which have been found previously in NOD mouse DC and M [34
35
36
].
In conclusion, mice with the NOD background show larger numbers of mature monocytes in the circulation and a preferential development of M-like cells from immature and mature monocytes. We can only speculate on the contribution of these described results to the proneness of NOD mice to develop various autoimmune conditions. We excluded the possibility that they are the consequence of the autoimmune inflammations, as they were also present in prediabetic NOD mice and inflammation-protected NOD-H2b and NOR mice. Conversely, if they are causally related to autoimmunity, we can envisage several mechanisms. First, the larger number of Ly-6Clow monocytes that preferentially mature into M might support the destructive character of the spontaneous autoimmune inflammation. In addition, the imbalance between the generation of DC versus M from monocytes in favor of the M lineage may play a role, particularly as DC are essential in tolerance induction and maintenance. Indeed, transfers with optimally functioning DC have proven to prevent the development of autoimmune diabetes in the NOD mouse [37
38
39
]. Therefore, an insufficient production of high-quality DC as a result of developmental aberrancies in their direct precursors could form a key to the initiation of autoimmunity.
Received November 16, 2004; revised February 15, 2005; accepted February 23, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B hyperactivation has differential effects on the APC function of nonobese diabetic mouse macrophages J. Immunol. 170,1770-1780This article has been cited by other articles:
 |
|
 |
|
  Y. Peng, Y. Latchman, and K. B. Elkon Ly6Clow Monocytes Differentiate into Dendritic Cells and Cross-Tolerize T Cells through PDL-1 J. Immunol., March 1, 2009; 182(5): 2777 - 2785. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M E Wildenberg, J M C Welzen-Coppens, C G van Helden-Meeuwsen, H Bootsma, A Vissink, N van Rooijen, J P van de Merwe, H A Drexhage, and M A Versnel Increased frequency of CD16+ monocytes and the presence of activated dendritic cells in salivary glands in primary Sjogren syndrome Ann Rheum Dis, March 1, 2009; 68(3): 420 - 426. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Timmerman, M. G. Flynn, P. M. Coen, M. M. Markofski, and B. D. Pence Exercise training-induced lowering of inflammatory (CD14+CD16+) monocytes: a role in the anti-inflammatory influence of exercise? J. Leukoc. Biol., November 1, 2008; 84(5): 1271 - 1278. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. McDuffie, N. A. Maybee, S. R. Keller, B. K. Stevens, J. C. Garmey, M. A. Morris, E. Kropf, C. Rival, K. Ma, J. D. Carter, et al. Nonobese Diabetic (NOD) Mice Congenic for a Targeted Deletion of 12/15-Lipoxygenase Are Protected From Autoimmune Diabetes Diabetes, January 1, 2008; 57(1): 199 - 208. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. M. Choi, J. Zhu, G. J. Stoltz, S. Vernino, M. Camilleri, J. H. Szurszewski, S. J. Gibbons, and G. Farrugia Determination of gastric emptying in nonobese diabetic mice Am J Physiol Gastrointest Liver Physiol, November 1, 2007; 293(5): G1039 - G1045. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Saxena, J. K. Ondr, A. F. Magnusen, D. H. Munn, and J. D. Katz The Countervailing Actions of Myeloid and Plasmacytoid Dendritic Cells Control Autoimmune Diabetes in the Nonobese Diabetic Mouse J. Immunol., October 15, 2007; 179(8): 5041 - 5053. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ziegler-Heitbrock The CD14+ CD16+ blood monocytes: their role in infection and inflammation J. Leukoc. Biol., March 1, 2007; 81(3): 584 - 592. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
