|
|
||||||||


* Inserm U404, Lyon Cedex, France; and
Schering-Plough, Laboratory for Immunological Research, Dardilly, France
Correspondence: Dr. C. Caux, Schering-Plough, Laboratory for Immunological Research, 27 chemin des Peupliers, BP 11, 69571 Dardilly, France. E-mail: christophe.caux{at}spcorp.com
| ABSTRACT |
|---|
|
|
|---|
i
proteins on the responding B lymphocytes. It is interesting that SNDC
was able to synergize with several chemokines to induce massive
migration of B lymphocytes. These observations show that DC
spontaneously produce factors that, alone or in cooperation with
chemokines, specifically regulate B-cell migration, suggesting a key
role of DC in the recruitment or localization of B lymphocytes within
secondary lymphoid organs.
Key Words: B lymphocytes germinal center T cells B cells chemokine
| INTRODUCTION |
|---|
|
|
|---|
Signals responsible for the recruitment of B cells into secondary lymphoid organs and controlling the migration of B cells from the extrafollicular area to the B-cell follicle or to the medullary cords are poorly characterized. However, chemokines have recently been shown to play a critical role in these events. In particular, CCL19 (ELC/MIP-3ß) and CCL21 (SLC/6Ckine) have been recognized to participate in the recruitment of naive T cells, DC, and B cells in the extrafollicular area [5 6 7 8 9 10 ]. These chemokines are produced by scattered cells in the extrafollicular area and act through CCR7, specifically expressed on activated T and B cells and on mature DC [5 , 7 , 9 , 11 12 13 ]. Recruitment of B cells into the follicle would rely on the chemokine CXCL13 (BCA-1/BLC), because disruption of the gene encoding its receptor CXCR5 abolished GC formation [14 ]. This chemokine is specifically expressed in the B-cell follicle [15 , 16 ], and its receptor CXCR5 has been detected on circulating B cells and some memory T cells [17 ]. Finally, the differentiation stage and the engagement of the B-cell receptor have been shown to modulate migration in response to chemokines such as CXCL12 (SDF-1) [18 ] or CCL19 [5 ].
The cells producing B-cell-active chemokines are not well characterized, but CCL19 has been shown to be produced by DC [5 , 19 ] and CXCL13, by follicular DC [16 ]. Others have shown that encounter of DC and T cells precedes that of T cells and B cells [20 ], but there is growing evidence that DC can directly provide help to B cells. First, a recent study has shown that DC can present the processed and the native form of the antigen to T and B cells, respectively [21 ]. Second, our own studies have demonstrated that DC stimulate several steps of B-cell differentiation through direct interactions [22 23 24 25 ]. Thus, DC are likely to represent a bridge allowing antigen-specific T cells to meet with antigen-specific B cells. To strengthen this concept, we investigated in the present study whether DC could produce soluble chemoattractants for B cells.
We found that DC spontaneously produce soluble chemoattractants, inducing potent and selective chemotaxis of naive and memory B cells but not that of other lymphocytes. It is important that these chemoattractants do not involve pertussis toxin (PTX)-sensitive pathways but can synergyze with several chemokines to induce potent B-cell chemotaxis. The present results demonstrate that DC constitute an important source of a novel class of B-cell chemoattractants, suggesting a key role of DC in the recruitment and migration of B cells within discrete areas of secondary lymphoid organs.
| MATERIALS AND METHODS |
|---|
|
|
|---|
(TNF-
; specific activity: 2x107 U/mg;
Genzyme, Boston, MA), rh stem cell factor (SCF; specific
activity: 4x105 U/mg; R&D Systems, Abington, UK), and rh
interleukin (IL)-4 (specific activity: 2x107 U/mg;
Schering-Plough Research Institute) were used at saturating
concentrations of 100 ng/ml, 2.5 ng/ml, 25 ng/ml, and 50 U/ml,
respectively. rh Chemokines CXCL12 (specific activity:
2x105 U/mg), CCL20 (specific activity: 4x105
U/mg), CCL19 (specific activity: 1x104 U/mg), and CCL21
(specific activity: 5x103 U/mg) were obtained through R&D
Systems. Lipopolysaccharide (LPS) was used at 10 ng/ml (Sigma Chemical
Co., St. Louis, MO), and anti-CD40 monoclonal antibody (mAb; Upstate
Biotechnology, Lake Placid, NY) was used at 10 µg/ml. PTX (Sigma) was
used at concentrations ranging from 5 ng/ml to 10 µg/ml.
Generation of DC from peripheral blood monocytes or
CD34+ progenitors
Monocytes were purified by immunomagnetic depletion (Dynabeads,
Dynal, Oslo, Norway) after preparation of peripheral blood mononuclear
cells (PBMC) followed by a 52% Percoll gradient. The pellet was
eventually recovered for T-cell purification, and monocytes were
purified from the low-density cells. The depletion was performed with
anti-CD3 (OKT3), anti-CD19 (4G7), anti-CD8 (OKT8), anti-CD56 (NKH1;
Coulter, Hialeah, FL), and anti-CD16 (ION16, Immunotech, Marseille,
France) mAbs. Monocyte-derived DC were produced by culturing purified
monocytes for 67 days in the presence of GM-CSF and IL-4
[26
]. At days 67, between 85% and 95% of the cells
were immature CD1a+ CD14- DC.
Alternatively, DC were generated from cord blood CD34+
progenitors in the presence of SCF, GM-CSF, and TNF-
, as previously
described [27
].
Purification of blood CD3+
T cells and natural killer (NK) cells
T cells were purified from PBMC by immunomagnetic depletion
(Dynabeads, Dynal). CD3+ T lymphocytes were purified using
a cocktail of mAbs, MOP9 (CD14), ION16 (CD16), mAb 89, (CD40), ION2
(HLA-DR; Immunotech), NKH1 (CD56; Ortho Diagnostic System, Raritan,
NJ), and 4G7 (CD19). After two rounds of bead depletion, the purity of
CD3+ T cells was routinely higher than 95%.
NK cells were purified following a similar procedure using anti-CD3 (OKT-3) and anti-CD4 (Sigma) mAbs instead of ION16 (CD16) and NKH1 (CD56). After two rounds of bead depletion, the purity of CD56+ NK cells was routinely higher than 80%.
Isolation of B cells
Mononuclear cells from tonsils were isolated by a standard
Ficoll-Hypaque (density=1077 g/ml) gradient method. Partially purified
tonsillar B cells were first enriched in the E- fraction
(9095% CD19+) and were used for most experiments. For
some experiments, this preparation was submitted to anti-CD2, -CD4,
-CD8, -CD14, and -CD16 mAbs negative selection with magnetic beads
coated with anti-mouse IgG (Dynabeads, Dynal). CD19 was expressed on
>99% of the B cells, as assessed by fluorescence analysis using a
FACScalibur (Becton Dickinson, San Jose, CA).
Alternatively, B cells were isolated from blood samples. They were first enriched in the high-density fraction of a 52% Percoll and further purified by positive selection using CD19 microbeads and the MACS system (Myltenyi Biotech, Bergish Gladbach, Germany).
Chemotaxis assays
B cells
Before migration assays, B cells were preincubated in migration
medium [RPMI 1640 supplemented with 2.5% fetal calf serum (FCS)] for
2 h. Migration assays were carried out using transwells (6.5 mm
diameter and 5 µm pore size; Costar, Cambridge, MA). Briefly,
supernatant from DC (SNDC) or other cells, or rh chemokines (1001000
ng/ml) or combinations was added to the lower wells in 24-well plates.
Cells (5x105) were added to the transwell inserts. Plates
were incubated for 3 h at 37°C. After removal of the transwell
inserts, cells from the lower compartments were labeled with
fluorescein isothiocyanate (FITC) anti-IgD, phycoerythrin (PE)
anti-CD38, and PE-Cy5 anti-CD19 for 30 min on ice. Relative cell counts
were determined using a FACScalibur for 40 s under a constant
sheath pressure with appropriate gates on forward-scatter and CD19
profiles. Each assay was performed in duplicate, and the results were
expressed as the mean ± SD of migrated cells per
well. Alternatively, the chemotaxis index was calculated as the ratio
between the numbers of migrating cells in the sample and in the control
medium. For PTX treatment, B cells were incubated for 2 h at
37°C with various concentrations of PTX, washed twice, and
subsequently added to the top chamber of the chemotaxis assay.
Other cells
For the other cell populations, migration assays were performed
without preincubation. Purified monocytes and monocyte-derived DC were
allowed to migrate for 1.5 h and processed for counting using
FITC-labeled anti-CD14 and anti-CD1a mAbs, respectively. T-cell
migration was performed using 3 µm transwells and revealed using
FITC-labeled anti-CD3, PE-labeled anti-CD8, and PE-Cy5-labeled anti-CD4
mAbs. NK cells were allowed to migrate for 3 h and stained using
PE-labeled anti-CD56 mAb.
Production of SNDC
DC (3.75x105) were cultured in medium alone (RPMI
1640, 2.5% FCS) or with 10 µg/ml anti-CD40 mAbs or 10 ng/ml LPS in a
final volume of 0.5 ml in 24-well culture plates. For some experiments,
DC were seeded at concentrations ranging from 1 x 105
to 1.5 x 106 per ml. After 2 days of culture,
supernatants were recovered, 0.22 µm-filtrated, and used for
chemotaxis assays. Supernatants were also recovered from tonsillar B
cells activated with anti-CD40 mAbs (10 µg/ml), from peripheral blood
T cells cultured in the presence of IL-2, or from blood monocytes.
Partial characterization of the chemotactic activity released by
DC
Sensitivity to protease was tested by incubating the SNDC with
agarose cross-linked-type VIII-A protease (Sigma) for 60 min at 37°C
followed by two centrifugations to remove the protease. Heat
sensitivity was determined by heating the SNDC at 60 or 90°C for 15
min.
| RESULTS |
|---|
|
|
|---|
(unpublished results) triggered a strong migration
of B lymphocytes, as shown by a mean chemotactic index (CI) of 50.
CCL19 and CCL21 were reproducibly found to induce B-cell chemotaxis but
with a lower magnitude (mean CI of 9.8 and 6.6, respectively, Table 1 ),
as previously shown [5
, 8
, 9
].
In addition, another CC chemokine, CCL20, induced B-cell chemotaxis
with a comparable efficiency, in accordance with the expression of its
receptor CCR6 on B lymphocytes [29
]. In contrast, CCL22
(MDC) and CCL17 (TARC) did not induce significant migration
(unpublished results). Thus, CXCL12 appears the most potent chemokine
tested by far, able to trigger B-cell chemotaxis, whereas CCL19, CCL20,
and CCL21 share lower but significant B-cell chemotactic activity. Of
note, all these chemokines are known to be produced by DC upon
activation [5
, 12
, 19
].
|
for 12 days (unpublished
results).
|
|
Finally, to distinguish between chemotaxis and chemokinesis, SNDC was added in the upper and lower chamber of the transwell (checkerboard analysis). As shown in Figure 2 , B-cell migration in response to SNDC was inhibited when the gradient was abolished, showing that DC induce directional migration of B cells.
|
SNDC induces migration of naive and memory
B cells
To determine whether SNDC attracted a particular subset of
tonsillar B cells, transmigrated cells were phenotyped by triple-color
staining using labeled anti-CD19, -CD38, and -IgD antibodies. Based on
IgD and CD38 expression, three major subsets of mature B cells can be
identified [30
]: naive
IgD+CD38low B cells,
IgD-CD38low memory B cells, and
IgD-CD38high germinal center (GC) B cells. The
tonsillar B-cell preparation used in one representative experiment
(shown in Fig. 3A
) contained 61% naive B cells, 21% memory B cells, and 14% GC B
cells. In marked contrast, transmigrated cells were composed of 35%
naive B cells, 59% memory B cells, and 4% GC B cells (Fig. 3B)
. A
compilation of five independent experiments, presented in Figure 3C
3D
3E
3F
, and expressed as % of input cells, showed that SNDC
preferentially attracted memory and to a lesser extent naive B cells
but not GC B cells. This pattern of chemotactic response of B
lymphocytes is similar to that observed with CXCL12 (Fig. 3
and ref.
[18
]). However, when chemotaxis indexes were considered,
naive B cells (mean CI=26, range 567, n=17) appeared as
efficiently attracted as memory B cells (mean CI=23, range 560,
n=17). This finding is explained by the higher propensity of
memory B cells to migrate spontaneously as compared with naive B cells
(Fig. 3D and 3E
, condition medium). Similar results were obtained for
blood B-cell subsets (unpublished results).
Together, these results show that DC produce soluble factors inducing selective migration of naive and memory B lymphocytes.
DC constitute the preferential source of
B-cell chemoattractants
We next determined whether the ability to produce B-cell
chemotactic factors was shared by other populations of the immune
system. For this purpose, three purified cell populations were tested
under the same experimental conditions: tonsillar B cells activated by
anti-CD40 mAbs, IL-2-stimulated CD3+ T cells, and
monocytes. Culture supernatants were recovered after 2 days and tested
for their capacity to induceB-cell chemotaxis. As shown in
Figure 4A
, only SNDC induced important migration of B cells, and a very
weak effect of supernatants from other cell populations was observed
(CI<3).
|
Migration of B cells in response to
SNDC is not sensitive to PTX
With the notable exception of CCL8 (MCP-2) [31
],
the chemotactic activity of most chemokines described to date is
inhibited by treatment of cells with Bordetella PTX, which interfere
with G
i proteins. As expected, low
concentrations of PTX (20 ng/ml) totally abolished migration of B cells
in response to CXCL12 (Fig. 5
) or CCL19 (unpublished results). In contrast, even at high PTX
concentrations, the migration of B cells in response to SNDC was only
affected slightly, suggesting that SNDC-induced B-cell chemotaxis does
not involve use of G
i proteins.
|
Altogether, these results demonstrate that DC produce nonprotein, soluble factors, inducing B-cell chemotaxis in a PTX-insensitive way.
SNDC synergizes with defined chemokines to induce
B-cell migration
DC produce many chemokines following activation [19
,
32
], but the absence of a significant PTX-sensitive
chemotactic activity for T and B cells suggests that our in vitro
experimental conditions might not allow production of chemokines in
sufficient concentrations to display in vitro chemotactic activity. To
examine whether the chemotactic factors produced by DC are able to
cooperate with chemokines to induce B-cell migration, SNDCs were
supplemented with exogenous recombinant chemokines. As shown in
Figure 6A
, the migration induced by optimal concentrations of CXCL12 or
CCL19 was strongly potentiated in the presence of SNDC. In particular,
the number of cells that migrated under those conditions was always
much more important than that expected by an additive effect of the two
activities (arrows in Fig. 6
), indicating that SNDC synergized with
CXCL12 and CCL19. SNDC increased by 1.65-fold (range 1.31.9,
n=5) and 2.0-fold (range 1.72.2, n=5) the
number of cells migrating in response to CXCL12 and CCL19,
respectively. In contrast, few if any effects were observed when
optimal concentrations of recombinant CXCL12, CCL19, and CCL21 were
combined together (unpublished results). Better synergism was observed
when experiments were performed with suboptimal concentrations of
CXCL12 (Fig. 6B , compare 0.2 and 1 µg/ml CXCL12).
|
| DISCUSSION |
|---|
|
|
|---|
Using in vitro-generated DC, we found that SNDC allows directional migration of naive and memory B cells, whereas its activity on other cell populations, although significant, remains marginal. The level of B-cell migration elicited by SNDC is comparable with that observed with the most efficient B-cell chemokine tested, CXCL12, but much more stronger than the migration induced by other B-cell-active chemokines, such as CCL19, CCL20, or CCL21. Of note, most of these chemokines can be produced by DC upon activation [5 , 12 , 19 ]. That B-cell migration in response to SNDC is marginally inhibited by PTX, in contrast to the chemotactic response elicited by recombinant CCL19, CCL20, CCL21, or CXCL12, indicates that these chemokines did not contribute to the observed activity. SNDC was found to induce low but significant migration of other cell types including CD4+ and CD8+ T cells. This finding might somehow appear in contradiction with the recognized role of DC in the induction of T lymphocytes recruitment within T-cell-rich areas and their capacity to produce T-cell-attractive chemokines such as CCL19 [5 ] or CCL20 (personal observation). This result is probably a result of the low concentrations of chemokines reached in the conditioned medium as compared with the relatively high amounts of chemokine required to observe in vitro chemotaxis. In addition, production of T-cell-active chemokines by DC requires their activation [19 , 32 , 33 ], and only resting SNDCs were tested in T-cell migration assays in the present study (Fig. 1) .
To our knowledge, CCL8 (MCP-2) is the only chemokine described to date that elicits a chemotactic response involving PTX-insensitive G proteins [31 ], suggesting that a DC-derived B-cell chemoattractant(s) might be different from classical chemokines. That SNDC-induced B-cell migration is not merely affected by protease digestion demonstrates that the DC-chemotactic factor(s) is not a protein, definitely excluding a role for chemokines. Besides chemokines, other mediators involved in the recruitment and trafficking of cells have been recognized, such as formyl-Met-Leu-Phe (fMLP), platelet-activating factor (PAF), complement fragments (C3a, C5a), or leukotrienes, which are potent chemoattractants for neutrophils and other leukocytes. Complement fragments are unlikely involved in SNDC-induced B-cell chemotaxis, as indicated by resistance to protease treatment and by the fact that up to 1 µg/ml rhC5a did not induce significant migration of B cells nor did a blocking anti-C5aR (anti-CD88) antagonized, SNDC-dependent B-cell migration (unpublished results). DC were recently shown to express the 5-lipoxygenase [34 ] necessary for synthesis of leukotrienes, which display chemotactic activity. That SNDCs, generated in the presence of various 5-lipoxygenase (5-LO) inhibitors [nordihydroguaiaretic acid (NDGA), DEC, and hydrocortisone] [35 ], induced potent B-cell migration (unpublished results) suggests that leukotrienes are not mandatory for SNDC-dependent B-cell chemotaxis. PAF-R is one of the rare receptors described to date that mediates at least part of its activity on cell migration through PTX-insensitive G proteins [36 37 38 39 ]. PAF-R is expressed on B cells [40 , 41 ], and engagement of this receptor by PAF induces their activation [42 , 43 ]. However, preliminary results, obtained with rhPAF and chemical antagonists of PAF-R, indicate that PAF is unlikely involved in DC-induced B-cell attraction (unpublished results).
That DC chemotactic activity is not inhibited by PTX does not exclude
that G-protein-coupled receptors (GPCR) might be involved on B cells.
In line with this possibility, many orphan GPCR have been described,
some of these receptors being expressed in the immune system and in
particular on B cells, such as Epstein-Barr-induced 2 (EBI-2)
[44
]. Conversely, other factors, acting through surface
receptors different from GPCR, such as cytokines, have been shown to
display chemotactic activity in some instances. For example, IL-15,
which can be produced by DC [45
46
47
], has been shown to
induce migration of T cells [46
, 48
] and
TNF-
, that of B cells [49
].
All the nonchemokine chemoattractants mentioned above (fMLP, complement fragments, PAF, and leukotrienes) participate in the development of inflammatory responses during injury and infection. A group of chemokines induced upon inflammation [i.e., macrophage-inflammatory protein (MIP), GRO, and monocyte chemoattractant protein (MCP)] also participate in the selective recruitment of leukocyte subpopulations at a site of injury. Another group of chemokines that are expressed constitutively (i.e., CXCL12, CCL19, and CXCL13) are involved in the regulation of lymphocyte trafficking in lymphoid organs. The chemoattractants produced spontaneously by DC, selectively attracting B cells, are likely to be involved in the regulation of B-cell trafficking and recruitment within secondary lymphoid organs. If this holds true, this mediator would represent the first example of a factor, different from a chemokine, involved in the control of lymphocyte trafficking in lymphoid organs.
In the last part of this study, we showed that this DC chemoattractant synergizes with defined chemokines to induce B-cell migration. In particular, SNDC was found to allow a robust B-cell chemotactic response when combined with suboptimal doses of CXCL12 or CCL19, probably by lowering the threshold of sensitivity of B cells to chemokines. It is interesting that these chemokines active on B cells are produced by DC following activation [5 , 12 , 19 ]. Limited studies have documented cooperative effects between chemoattractants on cell migration. Recently, a cooperative interaction between CXCL12 and SCF in the chemotaxis of CD34+ hematopoietic progenitors has been shown [50 , 51 ]. In addition, thrombin, which acts through a GPCR [52 , 53 ], has been shown to prime responsiveness to chemoattractants such as IL-8, C5a, and fMLP on transfected cells [54 ]. More recently, leukotrienes C4 and D4 were found to promote optimal chemotaxis of DC to the chemokine CCL19 [55 ]. Furthermore, surface molecules such as proteoglycans or GPCR themselves (Duffy), by acting as presenting molecules, have been proposed to enhance the activity of chemokines [56 ]. Soluble forms of such molecules putatively produced by DC could be involved in this synergistic activity. More likely, DC chemoattractant, which displays potent chemotactic activity per se, might transduce signal to B cells, lowering the number of receptors needed to be engaged by chemokine to reach biological activity.
The putative physiological relevance of these observations is presented in the following model. After antigen capture in the periphery, DC reach the lymphoid organs through the lymph stream in response to inflammatory stimuli. DC, now homing in the T-cell area, produce chemoattractant(s), allowing recruitment of B cells within the extrafollicular area. Upon cognate interaction with T cells, DC are able to produce chemokines such as CCL19 [5 ] or CCL20 (personal observation), which can synergize with DC chemoattractant to amplify B-cell recruitment. Antigen-specific B cells might be preferentially attracted through the production of CCL19, because this chemokine potently attracts B cells that have been stimulated through their B cell receptor (BCR) [5 , 8 ]. In this context, it is noteworthy that in contrast to CXCL12 [18 ], B-cell chemotaxis in response to SNDC is not impaired by BCR engagement (unpublished results). By favoring recruitment of immunocompetent lymphocytes and subsequently delivering critical signals to T and B lymphocytes, DC appear as the key antigen-presenting cells (APC) in the orchestration of the humoral response.
That production of B-cell chemotactic factor(s) is not conditioned by maturation stimuli would indicate that immature DC might also contribute to B-cell trafficking. Indeed, besides a role in B-cell recruitment during initiation of immune responses, DC may have a role in B-cell translocation from one lymphoid compartment to another. In particular, DC (GCDC) with a relatively immature phenotype have been described within primary and secondary B-cell follicles [57 ] and were proposed to contribute to the GC reaction [25 ]. Thus, GCDC may contribute to the constitutive homing of B cells within follicle and/or attraction of GC precursors generated during the extrafollicular reaction. In addition, immature DC of other sites might contribute to the localization of B cells within mucosal areas (such as crypts) or marginal zones.
Altogether, this study shows that DC are likely involved in the early steps of B-cell recruitment and localization into discrete areas within secondary lymphoid organs through the production of nonclassical chemoattractant(s) selective for B cells.
| ACKNOWLEDGEMENTS |
|---|
Received February 22, 2000; revised May 14, 2001; accepted May 16, 2001.
| REFERENCES |
|---|
|
|
|---|
J. Exp. Med. 184,695-706This article has been cited by other articles:
![]() |
K. Shimizu and R. N. Mitchell The Role of Chemokines in Transplant Graft Arterial Disease Arterioscler. Thromb. Vasc. Biol., November 1, 2008; 28(11): 1937 - 1949. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. W. Smith, R. L. Gamelli, S. B. Jones, and R. Shankar Immunologic responses to critical injury and sepsis. J Intensive Care Med, May 1, 2006; 21(3): 160 - 172. [Abstract] [PDF] |
||||
![]() |
N.-N. Huang, S.-B. Han, I.-Y. Hwang, and J. H. Kehrl B Cells Productively Engage Soluble Antigen-Pulsed Dendritic Cells: Visualization of Live-Cell Dynamics of B Cell-Dendritic Cell Interactions J. Immunol., December 1, 2005; 175(11): 7125 - 7134. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. G. Cormier, R. J. Durso, F. Tsamis, L. Boussemart, C. Manix, W. C. Olson, J. P. Gardner, and T. Dragic L-SIGN (CD209L) and DC-SIGN (CD209) mediate transinfection of liver cells by hepatitis C virus PNAS, September 28, 2004; 101(39): 14067 - 14072. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. A. Bajer, D. Garcia-Tapia, K. R. Jordan, K. M. Haas, D. Werling, C. J. Howard, and D. M. Estes Peripheral blood-derived bovine dendritic cells promote IgG1-restricted B cell responses in vitro J. Leukoc. Biol., January 1, 2003; 73(1): 100 - 106. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Beyer, M. Meyer-Hermann, and G. Soff A possible role of chemotaxis in germinal center formation Int. Immunol., December 1, 2002; 14(12): 1369 - 1381. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |