| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Pathology, Massachusetts General Hospital, and
Physiology Program, Harvard School of Public Health, Boston, MA 02129
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DC in peripheral tissues, including the lung, act as sentinels by engulfing foreign Ags that have traversed epithelial barriers (9, 10) and transporting these via lymphatics to local lymph nodes for presentation to T cells (1). Their turnover in airway epithelium is rapid (11), approaching that of DC in the intestinal tract (12), thus necessitating their continuous replenishment by DC precursors released from the bone marrow. In addition to migrating via lymphatics, DC may also emigrate from peripheral tissues via the blood stream to the spleen (1, 13) and liver (6, 13). This trace population of MHC class II+ DC constitutes <0.1% of circulating leukocytes (14).
Accumulation of leukocytes in the pulmonary vascular bed is a consequence of a number of factors including the structurally complex capillary network of the lung as well as the expression of selectins, integrins, and members of the Ig-like superfamily on both the endothelium and leukocytes (15). Langerhans cells express the E-selectin ligand, cutaneous lymphocyte-associated Ag (16), and DC circulating in the blood express the P-selectin glycoprotein ligand-1 that binds to P- and E-selectins (17). DC and their precursors also express ICAM-1 (CD54), ICAM-2 (CD50), ICAM-3 (CD102) (14), as well as LFA-1 (CD11a/CD18) (5).
We have previously shown that in vitro generated bone marrow-derived DC at varying stages of differentiation, when administered i.v., do not home to the lung (18). Further studies showed that monocyte-like cells are sequestered in the vascular bed of the lung (19). When these cells are harvested exclusively from the lung vasculature and cultured in the presence of GM-CSF, significantly greater numbers of DC are generated than from an equivalent number of monocytes derived from the systemic circulation. The present study was designed to identify those adhesion molecules that are important for the migration of monocyte-derived DC precursors into the lung under steady-state conditions. For this purpose, mutant mice deficient in E-selectin (E), P-selectin (P), E/P, ICAM-1, or CD18 and their respective controls were screened for evidence of reduced DC precursor margination and migration to the lung. Of these, only mice deficient in CD18 had significantly reduced numbers of DC in the walls of alveoli, veins, and airways of the lung. In ICAM-1 mutant mice, while the number of DC was reduced in alveolar walls, the reduction did not reach statistical significance and no reduction was observed in the walls of veins and airways. The results suggest that the absence of CD18, but not its ligand ICAM-1, significantly inhibits the margination and migration of DC precursors into the lung.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mice with homozygous mutations in genes for E, P, E/P, ICAM-1, or CD18 (kindly provided by Dr. A. L. Beaudet) were generated by gene targeting in embryonic stem cells as previously described (20, 21, 22, 23). Wild-type (WT) mice, serving as controls for CD18-/- mutants, were from the same mixed genetic background (129/Sv and C57BL/6) as CD18-/- mutant mice. C57BL/6 mice served as controls for all the other mutant mice, which have been back-crossed to this background at least six generations. All mice were between 6 and 9 wk of age.
Reagents and Abs
Monoclonal Abs M5/114 (24) (anti-MHC class II), PS/2 (25) (anti-VLA4), and M/K-2.7 (26) (anti-VCAM-1) were produced from hybridomas (TIB 120, CRL1911, and CRL-1909, respectively) obtained from American Type Culture Collection (Manassas, VA). The mAb MOMA-2 (27) (anti-monocytes/macrophages) was purchased from Biosource International (Camarillo, CA). Biotinylated rabbit anti-rat IgG was obtained from Harlan Bioproducts for Science (Indianapolis, IN). Other reagents included Gill’s hematoxylin No.2, n-n-dimethylformamide (Sigma, St. Louis, MO), 3-amino-9-ethylcarbazole (Aldrich, Milwaukee, WI), Tissue-Tek OCT compound embedding medium (Sakura Finetek, Torrance, CA), glycergel (Dako, Carpinteria, CA), hydrogen peroxide, LeukoStat kit (Fisher Scientific, Fair Lawn, NJ), and halothane (J. A. Webster, Sterling, MA).
Procedure
Mice were euthanized by inhalation of a lethal overdose of halothane. Peripheral blood was withdrawn from the inferior vena cava. After lysis of erythrocytes in a 10-µl aliquot of blood, the total number of leukocytes was calculated from hemocytometer counts. Erythrocytes in the remaining sample were sedimented on 4% dextran in PBS, and cytospin slides of the leukocytes were prepared. Differential white blood cell counts were obtained from cytospin preparations stained with LeukoStat. The remaining cytospin preparations were processed for immunoperoxidase staining. Lungs were removed, inflated with 0.8 ml of a 1:1 mixture of OCT and 0.9% NaCl, and snap-frozen on dry ice.
Immunoperoxidase
Cytocentrifuged leukocytes or frozen sections of lung, 4-µm thick, were stained using the avidin-biotin immunoperoxidase technique to identify MHC class II Ag (M5/114), monocytes/macrophages (MOMA-2), VLA-4 (PS/2), or VCAM-1 (M/K-2.7) as previously described (28). Briefly, after blocking with normal rabbit serum, sections/cells were incubated with optimal dilutions of the mAbs for 60 min at room temperature. Incubations with PBS, instead of the primary mAb or an isotype-matched irrelevant mAb, served as controls. Endogenous peroxidase was inhibited with 0.3% hydrogen peroxide in PBS for 20 min. This was followed by sequential incubations with biotinylated rabbit anti-rat IgG (1:100) and avidin-biotinylated peroxidase complex (1:50) according to the manufacturer’s instructions (ABC Elite kit; Vector Laboratories, Burlingame, CA) for 30 min each. Each incubation was followed by three washes with PBS. Reaction product was generated by incubation with 0.03% H2O2, 0.03% 3-amino-9-ethylcarbazole, 5% n-n-dimethylformamide in 0.1 M acetate buffer, pH 5.0. Sections were then counterstained with Gill’s hematoxylin No.2.
Morphometry of immunolabeled lung sections
The number of DC and monocytes was counted in each of three tissue compartments of the lung, including the walls of alveoli, veins, and airways. The walls of arteries were largely devoid of DCs and monocytes; they were excluded from the compartments measured. Owing to the low resolution inherent in frozen sections, no distinction was made between cells located within and outside of capillary lumens in alveolar walls. DC were identified as large cells with lobulated nuclei that strongly expressed MHC class II Ag, whereas monocytes were smaller, round cells that were immunolabeled with the MOMA-2 mAb. Using these criteria, DCs and monocytes were counted in two randomly selected lobes from each of nine animals per group. VLA-4+ cells in alveolar walls were similarly counted. Cells were counted in five fields using a x40 objective and 1-cm2 graticule divided into 10 x 10 squares. Because the relative proportion of tissue to airspace varied, cell counts were corrected for the fraction of airspace included in the area counted. This was accomplished by counting the fraction of the graticule squares devoid of tissue and subtracting this value from 100. The number of positively stained cells was then divided by the number of squares that contained tissue, and the quotient was multiplied by 100. To obtain the number of cells per cm2, the mean cell number was divided by 0.000625 (1 cm/40)2. To obtain the number of immunolabeled cells along the perimeter of airways and veins, cell counts were divided by the external perimeter of each airway or vessel, as measured with a 1-cm graticule, and the product was divided by 0.025 (1 cm/40).
Expression of VCAM-1 was evaluated on immunolabeled sections by counting the number of arteries and veins that had continuous (+), focal (±), or no (-) endothelial staining. The counts were conducted on coded slides by two observers who did not know the origin of the sections.
Statistical analysis
Each group consisted of six to nine mice. Data are presented as mean ± SEM. Data from CD18-/- and WT control mice and from ICAM-1-/- and C57BL/6 control mice were compared by an unpaired t test. Differences were considered statistically significant when p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because of the labor-intensive nature of these morphometric studies, we sought to identify limited but robust criteria that would enable us to identify DC and monocyte/macrophages in frozen sections of the lung without requiring the use of multiple immunological reagents. The criteria used to identify DC included intense staining for MHC class II, a lobulated nucleus and abundant cytoplasm. Although a number of anti-murine DC mAb have been developed, many of these recognize specific subsets of DC and some react with other cells in addition to DC (1). Unlike the anti-human CD14 mAb, the commercially available anti-murine CD14 mAb binds not only to monocytes but to a number of other leukocytes including DC and, therefore, was not suitable to identify monocytes. Finally, because the macrophage-specific F4/80 mAb produced weak and variable staining, the MOMA-2 mAb was selected to identify monocytes/macrophages.
Studies on adhesion molecule mutant mice
A preliminary study was conducted on groups of six of the
following mutant mice: P-/-,
E-/-, P/E-selectin-/-,
ICAM-1-/-, CD18-/-, and
their respective C57BL/6 and WT controls. Relative to their controls,
only CD18-/- and, to lesser extent,
ICAM-1-/- mutant mice showed a reduction in the
number of MHC class II+ cells in alveolar walls
(Fig. 1
A). Their number in
venous and airway walls was reduced only in
CD18-/- and not in
ICAM-1-/- mutant mice (data not shown).
Leukocyte counts were elevated to different degrees in all of the
mutant mice examined (Fig. 1B). These results suggested that
the margination and migration of monocyte-derived DC precursors into
the lung are, in part, CD18 dependent and less dependent on one of its
ligands, ICAM-1. Therefore, we elected to examine the distribution
of MHC class II+ DC and
MOMA-2+ monocytes/macrophages in the lungs of six
additional CD18-/- and
ICAM-1-/- mutant mice and six genetically
matched control mice.
|
Leukocytosis occurred in both CD18-/- and
ICAM-1-/- mutant mice and was greater in
CD18-/- than in
ICAM-1-/- mutants (Table I). Total white blood cell counts were
six times higher in CD18-/- mutants than in WT
control mice. Although the leukocytosis was primarily due to a 22-fold
increase in neutrophils, there was also an 8- and 6-fold increase in
the number of mature MHC classes II+ DC and
monocytes, respectively. In ICAM-1-/- mutants,
total white cell counts were 
3-fold higher than those of normal
controls. As in CD18-/- mice, leukocytosis was
primarily due to a 6-fold elevation in neutrophils, and mature DC and
monocytes were increased by 5- and 4-fold, respectively.
|
As expected, immunolabeling for CD11a/CD18 and ICAM-1 was absent
in the lungs of CD18-/- and
ICAM-1-/- mutant mice, respectively. The number
of MHC class II+ DC in the alveolar walls of
CD18-/- mutant mice was reduced by 68% as
compared with WT controls (p < 0.05) (Fig. 2, a and d and Fig. 3). The percent reduction in the number
of DCs in the walls of pulmonary veins (28%) and large (30%) and
small (41%) airways was less, but statistically significant (Fig. 2, b, c, e, and f and Fig. 3).
However, it is noteworthy that the observed percent reduction
significantly underestimates the magnitude of inhibition when the
elevated number of circulating leukocytes, including monocytes and DC,
is taken into account. The number of DC in the lungs of C57BL/6 and WT
control mice was not significantly different
(p > 0.09).
|
|
). While their
number appeared higher in the walls of veins and large and small
airways than in controls, the difference was also not statistically
significant. The number of MOMA-2+ monocytes/macrophages is reduced in the walls of alveoli and veins in CD18-/-, but not ICAM-1-/-, mutant mice
A statistically significant reduction in the number of
MOMA-2+ monocytes/macrophages in
CD18-/- mice, relative to controls, was
observed only in the walls of alveoli (35%) and pulmonary veins
(26%), but not around large and small airways (Fig. 4). In contrast, no difference was
observed in the number of MOMA-2+
monocytes/macrophages in the lungs of ICAM-1-/-
as compared with C57BL/6 control mice (Fig. 4). The number of
MOMA-2+ cells in the lungs of C57BL/6 and WT
control mice was not significantly different (p
> 0.6).
|
Ablation of a single gene in an animal may be associated with the
up-regulated expression of other gene products (29),
either as a compensatory response or as a reflection of a redundancy in
function of other molecules that may be protective. Therefore, we
examined pulmonary arteries and veins for either continuous (+),
discontinuous (±), or absent (-) immunolabeling for VCAM-1. In
CD18-/- mutant mice, there was a statistically
significant increase in the number of arteries that exhibited
continuous luminal labeling and a concomitant decrease in the number of
vessels that showed segmental labeling compared with WT mice (Fig. 5). In ICAM-1-/-
mutants, the extent of staining was increased in arteries and to a
lesser degree in veins, as compared with controls (Fig. 5). VCAM-1 was
not detected in pulmonary capillary endothelium in any of the animals
examined.
|
To determine the extent to which the leukocytosis present in these
mutant mice resulted in an augmented sequestration of leukocytes in the
pulmonary vasculature, sections of lungs were immunostained for VLA-4,
the ligand for VCAM-1. There was a 5-fold increase in the number of
VLA-4+ cells in alveolar walls of
CD18-/- mutants than in WT controls (Fig. 6). This corresponds well with the 6-fold
increase in circulating leukocytes observed in these mutant mice.
However, there was no increase in the number of
VLA-4+ cells in the walls of veins and airways
(data not shown). Similar to the CD18-/-
mutants, there was a 3-fold increase in the number of
VLA-4+ cells in alveolar walls in
ICAM-1-/- mice as compared with C57BL/6
controls, which agrees well with the 3-fold expansion in the number of
circulating leukocytes. The number of VLA-4+
cells was not increased in the walls of veins and airways.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The leukocyte integrins form a group of adhesive molecules that share a
common ß2 (CD18) subunit that is associated
with at least four 
subunits (CD11a, -b, -c -d) (33).
Mutant mice deficient in the ß2 subunit
(CD18-/-) express none of these integrins on
circulating leukocytes. Monocytes express CD11a, CD11b, and CD11c
subunits, whereas DCs express CD11a but not CD11b and variable amounts
of CD11c (14). In contrast to CD18, ICAM-1 is a member of
the Ig-like superfamily and is one of the ligands for CD11a/CD18 and
CD11b/CD18. While ICAM-1 is expressed on endothelium, it is also
present, together with ICAM-2 and ICAM-3, on DC and monocytes
(14). Because ablation of the CD18
ß2 subunit affects the function of more than
one integrin, and CD11/CD18 has multiple ligands, the lack of CD18
might be expected to have a greater impact than a deficiency of ICAM-1
on the emigration of DC precursors to the lung.
A subset of DC precursors is derived from monocytes (3), and it has been suggested that these give rise to tissue DC, including lung DC. Little is known about the adhesion molecules used by DC precursors entering the lung in the unperturbed state and whether they are similar to those used by monocytes. In vitro studies using IL-1-activated human umbilical vein endothelium suggest that monocyte migration is substantially CD18 independent (34) and that it uses a VLA-4-dependent pathway. By contrast, during LPS-induced lung inflammation in vivo, a large fraction of monocytes migrate through CD11/CD18 and VLA-4 independent pathways (35). As discussed below, these observations do not exclude the possibility that under unperturbed, steady-state conditions the trans-endothelial migration of DC precursors is to a large extent CD18 dependent, particularly because pulmonary capillary endothelial cells do not express VCAM-1.
The present study indicates that the accumulation of DC precursors within the walls of alveoli, veins, and large and small airways requires, in part, the adhesion molecule CD11/CD18. Because of the large body of evidence describing the role of CD11/CD18 in leukocyte adhesion to endothelium in preparation for migration, the most likely explanation for the defect in DC accumulation is a decrease in adhesion of blood borne DC precursors and their subsequent migration. In fact, the reduction in the number of DC at each site likely underestimates the role of CD11/CD18, as the observed reduction occurred in the presence of a 6-fold increase in the number of circulating monocytes, a population that harbors within it DC precursors. The proportion of DC precursors within this pool of circulating murine monocytes and more specifically within those sequestered in the lung vasculature is not known. However, if the 6-fold increase in the number of circulating monocytes is taken into account in predicting the expected number of DC precursors that migrate to the four tissue compartments measured, the inhibition is estimated to be between 91 and 95%. Predicting the number of cells expected to migrate into tissue in the presence of high circulating counts is problematic. First, there are spatial constraints in the number of circulating cells that contact the endothelium. Second, in the lungs the increase in the marginated pool does not linearly parallel the increase in circulating cells so that the increase in emigrating cells is almost certainly not linearly related to the number of circulating cells. Third, the organ itself is likely to control the number of cells that enter, in part by the effect of the entering cells on either host defense or the clearance of the inflammatory stimulus.
While a defect in adhesion is a likely mechanism for the observed decrease in the accumulation of DC, the numerical density of migratory cells in a given tissue depends on the rates of both influx and efflux of these cells. At present there are no data indicating that CD18 is necessary for DC efflux from the lung. Ligation of CD18 has been suggested to enhance apoptosis of neutrophils that respond to an inflammatory stimulus, although no data examining DC are available. However, a defect in apoptosis would serve to increase the accumulation of DC, while the data clearly show that deficiency of CD18 results in a defect in the accumulation of DC in all regions of the lungs. However, a role for CD18 in the efflux of DC or in the migration of these cells within the extra-cellular matrix cannot be excluded.
In the initial group of six mice (Fig. 1), there appeared to be a
reduction in the number of DC in alveolar walls of
ICAM-1-/- mice that was not, however,
statistically significant (p = 0.1409), and
there was no difference in the other three tissue compartments
examined. In the second group of nine ICAM-1-/-
mice (Fig. 3), the reduction in the number of DC in alveolar walls was
marginally statistically significant (p =
0.0599). In view of the 4-fold increase in circulating monocytes, this
observation may underestimate a partial defect in DC precursor
migration. Nevertheless, the defect in DC precursor migration is
smaller than that observed in the absence of CD18. This discrepancy
could be due to several possibilities. First, ICAM-1 mutant mice may
have adopted alternative pathways that compensate for their genetic
deficiency. Studies comparing neutrophil emigration following blockade
of ICAM-1 using several approaches during LPS and
Pseudomonas pneumonias (36, 37)
raise this possibility. Second, the ICAM-1-deficient mice used in this
study are known to express low levels of alternatively spliced ICAM-1
(38). Although this ICAM-1 has never been shown to be
functionally active, this possibility cannot be excluded. Third, other
molecules, including ICAM-2, may mediate CD11/CD18 adhesion
(39). Fourth, other adhesion systems may mediate DC
precursor migration. Our studies addressed the possibility that
VLA-4/VCAM-1 adhesion mediates DC precursor emigration and demonstrated
an increase in VCAM-1 expression in arteries. However,
VLA-4/VCAM-1-mediated adhesion is unlikely to be contributing to DC
precursor emigration in alveolar walls, as VCAM-1 is not expressed in
capillaries. Moreover, the increase in VCAM-1 expression is similar in
both ICAM-1-/- and
CD18-/- mutant mice, but the latter and not the
former have a significant defect in the accumulation of DC in the
lung.
The observation that the number of circulating DCs was increased 8-fold in CD18-/- mice poses an interesting question. It has been assumed that this trace population of circulating, mature DC has emigrated from peripheral tissues and is en route to the spleen (1). If the reduced number of DC in the lungs of these mutant mice is representative of other peripheral tissues, a reduction in the number of circulating mature DC would be expected. Possible mechanisms underlying the observed increase in the number of circulating, mature DC include a partial inhibition of a CD18-dependent migration of mature DC into the spleen. A second contributory factor includes the altered hemopoiesis that many of these mutant mice develop (40). Both CD18-/- and ICAM-1-/- mutants have increased numbers of circulating neutrophils, monocytes, and DC, all of which share a common progenitor cell (2). A third, albeit speculative, mechanism might include the presence of elevated circulating growth factors, such as GM-CSF, in these mutant mice that promotes the maturation of DC in circulation. However, regardless of the mechanisms involved relative to the other leukocytes, mature DCs remain a minor circulating leukocyte population.
In summary, these observations suggest that under nonperturbed conditions CD18-mediated adhesion is necessary for the emigration of normal numbers of DC precursors into the lung. However, the defect is partial, suggesting that CD18-independent pathways are also required, possibly involving VLA-4/VCAM-1. The smaller defect observed in ICAM-1-/- mice suggests that the CD11/CD18 complex recognizes ligands other than ICAM-1.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Eveline E. Schneeberger, Molecular Pathology Unit, Massachusetts General Hospital East, 149 13th Street, Charlestown, MA 02129. E-mail address:
3 Abbreviations used in this paper: DC, dendritic cell; VLA-4, very late activation Ag-4; E, E-selectin; P, P-selectin; WT, wild type.
Received for publication September 23, 1999. Accepted for publication December 13, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. J. Exp. Med. 184:695.. II. Functional analysis. Blood 90:1458.
and dexamethasone on dendritic cell (DC) traffic to the lung. FASEB J. 10:A1007.
reduces Ia+ dendritic cell traffic to the lung. J. Leukocyte Biol. 60:1.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  J. McGill, N. Van Rooijen, and K. L. Legge Protective influenza-specific CD8 T cell responses require interactions with dendritic cells in the lungs J. Exp. Med., July 7, 2008; 205(7): 1635 - 1646. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Ichiyasu, J. M. McCormack, K. M. McCarthy, D. Dombkowski, F. I. Preffer, and E. E. Schneeberger Matrix Metalloproteinase-9-Deficient Dendritic Cells Have Impaired Migration through Tracheal Epithelial Tight Junctions Am. J. Respir. Cell Mol. Biol., June 1, 2004; 30(6): 761 - 770. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. T. Pribila, A. A. Itano, K. L. Mueller, and Y. Shimizu The {alpha}1{beta}1 and {alpha}E{beta}7 Integrins Define a Subset of Dendritic Cells in Peripheral Lymph Nodes with Unique Adhesive and Antigen Uptake Properties J. Immunol., January 1, 2004; 172(1): 282 - 291. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Y. Vermaelen, D. Cataldo, K. Tournoy, T. Maes, A. Dhulst, R. Louis, J.-M. Foidart, A. Noel, and R. Pauwels Matrix Metalloproteinase-9-Mediated Dendritic Cell Recruitment into the Airways Is a Critical Step in a Mouse Model of Asthma J. Immunol., July 15, 2003; 171(2): 1016 - 1022. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Sakurai, H. Taguchi, A. Anand, B. K. Ambati, E. S. Gragoudas, J. W. Miller, A. P. Adamis, and J. Ambati Targeted Disruption of the CD18 or ICAM-1 Gene Inhibits Choroidal Neovascularization Invest. Ophthalmol. Vis. Sci., June 1, 2003; 44(6): 2743 - 2749. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Fiorini, W. Vermi, F. Facchetti, D. Moratto, G. Alessandri, L. Notarangelo, A. Caruso, P. Grigolato, A. G. Ugazio, L. D. Notarangelo, et al. Defective migration of monocyte-derived dendritic cells in LAD-1 immunodeficiency J. Leukoc. Biol., October 1, 2002; 72(4): 650 - 656. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Gironella, M. Molla, A. Salas, A. Soriano, M. Sans, D. Closa, P. Engel, A. Salas, J. M. Pique, and J. Panes The role of P-selectin in experimental colitis as determined by antibody immunoblockade and genetically deficient mice J. Leukoc. Biol., July 1, 2002; 72(1): 56 - 64. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. MAUS, J. HUWE, L. ERMERT, M. ERMERT, W. SEEGER, and J. LOHMEYER Molecular Pathways of Monocyte Emigration into the Alveolar Air Space of Intact Mice Am. J. Respir. Crit. Care Med., January 1, 2002; 165(1): 95 - 100. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Bouvard, C. Brakebusch, E. Gustafsson, A. Aszodi, T. Bengtsson, A. Berna, and R. Fassler Functional Consequences of Integrin Gene Mutations in Mice Circ. Res., July 30, 2001; 89(3): 211 - 223. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
