Immunity, Vol. 16, 465–477, March, 2002, Copyright 2002 by Cell Press
?1 Integrin Is Not Essential for Hematopoiesis
but Is Necessary for the T Cell-Dependent
IgM Antibody Response
toskeletal reorganization and changes in gene expres-
sion affecting proliferation, differentiation, and survival
on the other hand, can modulate the affinity/avidity of
integrins, which is crucial for the extravasation of leuko-
cytes (Gonzalez-Amaro and Sanchez-Madrid, 1999).
HSC express several integrins that play a prominent
role in their adhesion to ECM and nonhematopoietic
cells of the bone marrow (BM) and perhaps also in their
hesionof HSCandhematopoieticprogenitor cells(HPC)
to fibronectin is mediated by ?4?1 as well as ?5?1 (Wil-
liams et al., 1991; van der Loo et al., 1998). ?4?1 also
mediates binding to VCAM-1 expressed on BM stroma
cells (Oostendorp et al., 1995). The importance of these
of FN fragments and antibodies against ?4?1 or VCAM-1,
which mobilized HPC into the blood (Papayannopoulou,
1995; van der Loo et al., 1998). In addition to retention
of HSC and HPC in the BM, integrin-mediated adhesion
might also be crucial for the self-renewal and survival
of HSC, since ?4?1-mediated attachmentof HPC to fibro-
nectin promotes proliferation (Yokota et al., 1998; Scho-
field, 1998) and prevents apoptosis (Wang et al., 1998).
Genetic studies with ?4 null chimeric mice revealed
an important function of ?4 integrin in myelo- and lym-
phopoiesis. Since the ?4 subunit can associate with
either the ?1 or the ?7 subunit, ?4 null hematopoietic
in the development of erythroid, myeloid, and B cell
progenitors (Arroyo et al., 1996, 1999). In addition, T cell
precursors are unable to leave the BM. None of these
defects was observed in ?7 null mice, which lack ?4?7
but still express ?4?1 (Wagner et al., 1996). It was there-
fore concluded that ?4?1 and not ?4?7 is responsible
for the marked abnormalities in ?4 null chimeric mice.
A plausible explanation for the severe phenotype is a
defect of ?4 null progenitors in the transmigration
through the BM stroma and in proliferation (Arroyo et
Integrin expression furthermore affects the migration
and function of differentiated blood cells as highlighted
by the severe defect in granulocyte extravasation in
patients lacking ?2 integrin (Hogg and Bates, 2000). It
is not known whether ?1 integrin plays a role for the
normal circulation of lymphocytes, although it is crucial
for the extravasation of ?1 null HSC into hematopoietic
organs (Hirsch et al., 1996; Potocnik et al., 2000). With
respect to lymphocyte function, both costimulatory (Ya-
mada et al. 1991; Damle and Aruffo, 1991) and inhibitory
effects (Groux et al., 1989; Ticchioni et al., 1993) of ?1
integrin on T cell activation were demonstrated in vitro.
The exact function in vivo, however, is unclear.
We generated mice in which we induced a deletion
of the ?1 integrin gene restricted to the hematopoietic
system to directly study the role of ?1 integrin for HSC
tiated blood cells.
Cord Brakebusch,1,2,7Simon Fillatreau,3
Alexandre J. Potocnik,4Gerd Bungartz,1,2
Patricia Wilhelm,5Marcus Svensson,1
Phil Kearney,6Heinrich Ko ¨rner,5
David Gray,3and Reinhard Fa ¨ssler1,2
2Max Planck Institute of Biochemistry
3University of Edinburgh
4Basel Institute for Immunology
5Nikolaus-Fiebiger-Zentrum fu ¨r Molekulare Medizin
Several experimental evidences suggested that ?1 in-
tegrin-mediated adhesion ofhematopoietic stem cells
(HSC) is important for their function in the bone mar-
row (BM). Using induced deletion of the ?1 integrin
gene restricted to the hematopoietic system, we show
BM, hematopoiesis, and trafficking of lymphocytes.
However, immunization with a T cell-dependent anti-
gen resulted in virtually no IgM production and an
increased secretion of IgG in mutant mice, while the
response toa Tcell-independent type 2antigen showed
decreases in both IgM and IgG. These data suggest
that ?1 integrins are necessary for the primary IgM
which have the potential to self-renew and to differentiate
into all hematopoietic lineages. Cell-cell and cell-matrix
adhesion of hematopoietic cells are important to control
(Prosper and Verfaille, 2001). Integrins are an important
family of cell surface receptors mediating several of these
interactions.Integrins aretransmembrane moleculescon-
sisting of an ? and a ? subunit. They bind extracellular
The cytoplasmic domain is connected to the actin cy-
toskeleton. Ligand binding to integrin activates various
intracellular signaling pathways, which results in cy-
Figure 1. Efficient Deletion of the ?1 Integrin Gene in the Hematopoietic System
(A and B) Southern blot analysis of genomic DNA isolated from indicated organs of (fl/fl cre) mice or ?1 null BM chimera 1 week (A) or at
different time points (B) after the first polyIC induction. Liver DNA from (fl/?) mice was used as a control. A lacZ fragment was used to identify
the floxed and the null allele.
(C) Cells were isolated from various lymphoid organs of (fl/– cre) and control (fl/? cre) BM chimeras 4 months after induction of the knockout.
Cells were analyzed by FACS for expression of ?1 integrin and of Ly-5.1, which specifically recognizes cells derived from the host (n ? 3-6/3-6).
?1 Integrin and T Cell-Dependent IgM Response
Table 1. Normal Numbers of Polymorphonuclear and
Mononuclear Cells in the Peripheral Blood
Maintenance and Differentiation of ?1 Null HSC
?1 integrin function is essential for homing of adult HSC
to the BM (Potocnik et al., 2000). To test whether ?1
integrin is important for the function of HSC within the
BM, we induced the deletion of the ?1 integrin gene in
the murine hematopoietic system.
Mice carrying a ?1 integrin gene flanked by loxP sites
(fl/fl) (Potocnik et al., 2000) were mated with mice with
1995) and mice expressing the cre recombinase under
the control of the Mx promoter (?/? cre) (Ku ¨hn et al.,
1995) to obtain a mouse strain with a conditional and a
null allele for ?1 integrin as well as the cre transgene
(fl/– cre). These mice expressed ?1 integrin on hemato-
poietic cells and were phenotypically normal. Three in-
jections of polyIC at 2 day intervals 4–8 weeks after the
BM transfer resulted in an efficient deletion of the ?1
integrin gene in liver and BM within 1 week (Figures 1A
and 1B). The deletion in spleen was about 70%, while less
than 5% of the thymocytes had lost the ?1 integrin gene.
To restrict the deletion of the ?1 integrin gene to the
hematopoietic system, BM from noninduced (fl/– cre)
As control, BM from litter mates carrying a conditional
and a wild-type allele of ?1 integrin as well as the cre
stitution of the hematopoietic system, the mice were
treated with polyIC. Mice were analyzed by FACS 2, 4,
6, and 8 weeks and 3, 4, 6, 8, 10, and 12 months after
the polyIC treatment.
After 8 weeks, BM cells and thymocytes had lost
nearly the entire ?1 integrin expression, while spleen,
?1-expressing cells (Figures 1B and 1C).
The percentage of host-derived cells was determined
by host (Ly- 5.1)- and donor-specific (Ly-9.1, Ly-5.2)
cell contribution was 1% or less in thymus and bone
marrow and up to about 5% in spleen, lymph nodes,
and Peyer’s patches (Figure 1C and data not shown).
These host cells were almost exclusively T cells leading
to a host cell contribution to the CD3?T cells of about
15% in BM and 10% in spleen and lymph nodes. In all
these tissues, the host cell contribution to B cells was
lower than 1%. More than 95% of the BM cells express-
ing markers for the erythroid (Ter-119) and myeloid (Gr-1,
Mac-1) lineage lacked expression of ?1 integrin. The
population sizes were similar to normal mice (data not
shown). Differential blood counts revealed similar num-
bers of polymorphonuclear (PMN) and mononuclear
cells (M) in mutant and control mice 5 weeks and 4
months after induction of the knockout (Table 1). This
phenotype was also observed even 12 months after the
induction of the knockout, further corroborating that the
?1 integrin gene was deleted on HSC and that loss of
?1 integrin expression did not impair HSC function in
To assess the number of granulocyte/macrophage
and erythroid progenitor cells in the BM colony, assays
(GM) were carried out with BM leukocytes. 5 weeks and
4 months after induction of the knockout, the frequency
?1 Null BM Chimera
4 months 0.89 ? 0.29 1.39 ? 0.75 0.97 ? 0.61 1.77 ? 0.71 6/6
1.48 ? 0.61 2.15 ? 0.57 1.83 ? 1.14 2.4 ? 0.744/4
5 weeks and 4 months after induction of the ?1 integrin gene dele-
tion, whole blood of control and ?1 null BM chimera was diluted
n/n indicates the number of animals in control and mutant group.
(All numbers in 106cells/ml). PMN, polymorphonuclear; M, mononu-
(Table 2). FACS analysis of randomly picked colonies
indicated an efficient deletion of the ?1 integrin gene on
GM HPC. 36 out of 37 colonies derived from 4 different
mutant mice 5 weeks after knockout induction were ?1
deficient. After 4 months, all 49 colonies derived from
To test whether ?1 null BM chimera show an increased
release of HPC into the blood, we performed colony
assays (GM) with peripheral blood leukocytes isolated
from mice 2 weeks after induction of the knockout. At
this time, more than 98% of the platelets in the mutant
mice had lost ?1 integrin expression (data not shown).
The frequency of colony forming units in the peripheral
blood was similar in normal and mutant BM chimera
(Table 2), indicating an efficient retention of progenitor
cells in the BM of mutant mice.
B Cell Development in the Absence of ?1 Integrin
?4 integrin-deficient mice, which lack ?4?1 and ?4?7,
have a severe defect in B cell differentiation (Arroyo et
al., 1996, 1999). Since ?7 null mice have normal B cell
development (Wagner et al., 1996), ?4?1 integrin was
thought to cause the B cell defect. Therefore, we moni-
tored B cell development in BM, spleen, and lymph
nodes of ?1 null BM chimera using different B cell-
specific markers: B220 (pre-proB and later), CD19 (proB
and IgD (all mature B).
Table 2. Normal Number of Progenitor Cells in Blood and BM of ?1
Null BM Chimera
?1 Null BM
5.5 ? 4.0
37.7 ? 6.8
20.6 ? 9.5
49.2 ? 18.3
10.5 ? 1.6
3.8 ? 5.0
47.1 ? 11.7
22.3 ? 13.5
43.7 ? 2.8
13.1 ? 8.7
Methylcellulose colony assays for granulocyte, macrophage, and
erythrocyte progenitors (GM) were carried out with 200000 leuko-
cytes from peripheral blood (PB) 2 weeks after knockout induction
or with 20000 cells of BM 5 weeks and 4 months after knockout
induction. For preB cell assays, 50000 BM leukocytes were seeded
per dish. The assay was carried out in triplicates and colonies were
counted after 8–9 days. Presented is the number of colonies per
dish with standard deviation. n/n indicates the number of animals
in control and mutant group. Integrin expression was assessed by
FACS on randomly picked clones (see Results).
Figure 2. Normal B Cell Development in the Absence of ?1 Integrin
(A) Presence of ?1 null CD19?B cells in BM, spleen, and lymph nodes of ?1 null BM chimeras 4 months after induction of the knockout (n ?
(B) Normal sizes of B cell subsets characterized by the expression of the B cell markers B220, CD21, IgM, and IgD in BM, spleen, and lymph
nodes of ?1 null BM chimeras 4 months after induction of the knockout as determined by FACS (n ? 5-6/4-6).
The frequency of preB cells in BM was determined in
methylcellulose assays. Similar numbers of preB cells
were found in BM of mutant and control mice 5 weeks
and 4 months after induction of the knockout. FACS
analysis of randomly picked colonies indicated a loss
Out of 20 colonies analyzed from 4 different mutant
mice 5 weeks after induction of the knockout, 19 lacked
expression of ?1 integrin (Table 2). After 4 months, 18
out of 19 preB cell colonies derived from 5 different
mutant mice were ?1 deficient.
85% or more of the CD19?cells in the mutant BM,
spleen, lymph nodes, and Peyer’s patches showed no
detectable expression of ?1 integrin after 5 weeks and
at all later time points tested (Figure 2A and data not
shown). In most cases, similar numbers of B220medium
(immature), B220high(mature), CD25?, and IgM?cells
were found in BM of control and mutant mice, although
between different BM transplantations, both in control
and control mice had similar amounts of IgD?IgM?cells
in spleen and lymph nodes (Figure 2B). In spleen, the
levels of B220?CD21?CD23?and B220?CD21?CD23?
cells (the latter characterizing marginal zone B cells)
were similar in mutant and control mice 5 weeks and 4
months after knockout induction (data not shown).
in a compensatory upregulation of other integrin sub-
units on immature (B220medium) and mature (B220high)
B cells in the BM. Normal immature B cells expressed
?4, ?5, low amounts of ?6, ?1, ?2, but no ?1, ?2, ?3,
and ?7 (data not shown). Normal mature B cells ex-
pressed ?4, low amounts of ?6, reduced ?1, low and
high amounts of ?2, and ?7 integrin, but not ?1, ?2, ?5,
and ?3. Deletion of the ?1 gene on immature B cells
resulted in loss of ?4 and ?5 integrin expression,
whereas on mature B cells ?4 expression was only
slightly reduced (data not shown).
Transient Defect in Thymus Colonization but Normal
Previous reports suggested that ?4 integrin is essential
for the emigration of T cell precursors from the BM
(Arroyo et al., 1996), ?6?1 integrin participates in their
migration to the thymus (Ruiz et al., 1996), and ?4?1
and ?5?1 are important for intrathymic development
(Salomon et al., 1994; Crisa et al., 1996). Therefore, we
assessed T cell development in ?1 mutant mice.
One week after induction of the knockout, less than
5% of the thymocytes had lost the ?1 integrin gene
compared to more than 75% of the BM cells (Figures
1A and 1B). In the following weeks, the percentage of
?1 null thymocytes slowly increased to about 95% after
?1 Integrin and T Cell-Dependent IgM Response
tion of thymic cellularity in ?1 null BM chimera. After 8
were similar in control and knockout mice.
To confirm and further investigate the migration of
T cell precursors to the thymus and the transient reduc-
tion in thymocytes, we performed an additional experi-
ment. Fetal thymi were isolated, depleted of T cells, and
engrafted under the kidney capsule of ?1-deficient BM
chimera. After 4 weeks, thymocytes were isolated and
counted. ?1 null thymocytes were present in the en-
grafted thymi, confirming that ?1 null T cell precursors
from the BM can colonize the thymic tissue (Figure 3B).
In agreement with the transient reduction of thymocytes
per ectopic thymus was about 3-fold lower in ?1 null
BM chimera. Thymocytes were then analyzed by FACS
using different markers for the intrathymic T cell devel-
opment (CD25, CD44, CD4, CD8, TCR). CD4CD8 DN cells
were subdivided into (CD25–CD44?), (CD25?CD44?),
(CD25?CD44lo) and (CD25–CD44?) subsets. In addition
to CD4 CD8 DP, CD4 SP, and CD8 SP, also the TCRdull
and TCRhighpopulation sizes were analyzed. No differ-
ence was found in the relative population size of these
thymocyte subsets between normal and mutant mice
ment of T cells.
?1 integrins have a costimulatory role on T lymphocytes
in vitro (Yamada et al. 1991; Damle and Aruffo, 1991).
Therefore, we examined whether loss of ?1 integrin af-
fects the amount of activated lymphocytes in nonimmu-
markers CD62L, CD44, and CD69 on B cells and CD4?
and CD8?T cells in spleen, lymph nodes, and peripheral
blood. Activated lymphocytes express low levels of
CD62Landhigh levelsofCD44and CD69.Nosignificant
difference in the amount of activated lymphocyte sub-
sets was observed between mutant and normal mice
(data not shown).
Figure 3. T Cell Precursors Can Migrate to the Thymus and Develop
into T Cells in the Absence of ?1 Integrin
(A) Efficient deletion of ?1 integrin expression on thymocytes and
normal development to CD4CD8 DP, CD4 SP, CD8 SP cells in the
thymus of ?1 null BM chimeras 4 months after induction of the
knockout (n ? 3/3).
(B) T cell depleted fetal thymi from normal mice were engrafted
under the kidney capsule of normal and ?1-deficient BM chimera.
4–5 weeks after engraftment of the thymi into ?1 null or control BM
chimeras the mice were sacrificed and the cell numbers within the
transplanted thymi were counted. ?1 null T cell precursors could
migrate but less efficiently into thymi engrafted under the kidney
Homing of Lymphocytes to Spleen, Lymph Nodes,
and Peyer’s Patches
topoietic stem cells, we tested whether it also plays a
role in the migration of differentiated lymphocytes to
spleen, lymph nodes, and Peyer’s patches. In addition,
we analyzed the distribution of the ?1-deficient B and
T cells within these organs.
wild-type lymphocytes were injected into the tail vein
of recipient mice. After 4 and 16 hr, the relative amounts
of mutant and normal transferred cells in the different
organs were determined by FACS. B and T cells of con-
trol, knockout, and recipient mice were distinguished
by cell surface markers and by labeling cells with
CMTMR prior to injection.
?1 null B cells showed normal homing to spleen,
lymph nodes, and Peyer’s patches 4 hr as well as 16 hr
after injection (Figure 4A). Similarly, T cell migration to
secondary lymphoid organs was not significantly af-
fected by the loss of ?1 integrin (Figure 4B). Labeling
by CMTMR did not affectthe homing efficiency. Homing
about 6 weeks as determined by FACS analysis. The
subset sizes of CD4CD8 DN, CD4 SP, CD8 SP, and
CD4CD8 DP cells were similar in mutant and control
mice at all time points tested (Figure 3A and data not
shown). Interestingly, mutant mice displayed a transient
reduction of the number of thymocytes. 5 weeks after
induction of the gene ablation, mutant thymi contained
56.8 ? 6.0 ? 106cells, whereas thymi from control mice
Figure 4. Normal Homing of ?1 Null B and T Cells into Spleen, Lymph Nodes, and Peyer’s Patches
(A and B) Mixtures of equal amounts of control and ?1 null cells (both Ly 5.2), one of them labeled with the intracellular fluorescent dye
CMTMR, were injected into the tail vein of B6SJL recipient mice (Ly-5.1). After 4 and 16 hr the mice were sacrificed. The ratio of control to
?1-deficient cells in spleen, lymph nodes, and Peyer’s patches was determined by FACS (n ? 3-5). These numbers were divided by the ratio
of control to mutant cells in the injected cell mixture, as assessed by FACS, resulting in the relative homing of control to mutant cells. Error
bars indicate standard deviation.
?1 Integrin and T Cell-Dependent IgM Response
assays with enriched preparations of B and T cells
showed similar results.
Exchange of dyes or separation of B and T cells using
B220 or Thy-1.2 coupled magnetic beads showed simi-
To study the localization of transferred lymphocytes
within the spleen, B and T cell fractions of splenocytes
from control and mutant mice were injected into the tail
vein of B6SJL mice. Spleen sections were stained with
Ly-5.2, detecting specifically the transferred cells. To
reveal the tissue organization, sections with labeled B
and T cells were counterstained with Thy-1.2 or B220,
were mainly found outside the CD3-rich periarteriolar
lymphoid sheath (PALS) (Figures 4C and 4D; data not
shown). At the same time points, mutant and wild-type
T cells were found concentrated in the PALS region
(Figures 4E and 4F; data not shown).
To detect long-term defects or defects dependent on
the interactions of lymphocytes with other ?1-deficient
hematopoietic cells, we assessed the distribution of B
and T cells in spleen, lymph nodes, and Peyer’s patches
of normal and ?1 null BM chimera. Mutant mice showed
normal distribution of B and T cells within the lymphoid
organs (Figures 4G and 4H; data not shown).
Decreased Levels of Both IgM and IgG after
Immunization of Mutant Mice with a
T Cell-Independent Type 2 Antigen
Immunization with the type 2 T cell-independent (TI-2)
antigen DNP-Ficoll led to decreased levels of both IgM
treatment (Figures 6A and 6B). IgG subtype analysis
revealed that IgG3, the major isotype produced in re-
sponse to T cell-independent type 2 antigens, but also
IgG1 and IgG2a, are decreased in the mutant mice (Fig-
Normal Secretion of Both IgM and IgG by Mutant B
Cells after LPS Treatment In Vitro
mice is a B cell autonomous defect, splenic leukocytes
were stimulated in vitro with 10 ?g/ml lipopolysaccha-
ride (LPS), which directly activates B cells. After 3 and
6 days, normal levels of IgM and IgG were detected in
the supernatant of mutant splenocytes, indicating that
the impaired IgM response observed in vivo is caused
by defective B cell activation and not by an intrinsic
B cell defect that prevents the efficient production of
IgM (Figure 6D).
Decreased Amount of IgM but Increased Levels
of IgG after T Cell-Dependent Immunization
of Mutant Mice
To study whether the loss of ?1 integrin in the hemato-
poietic system affects the antigenic response, normal
and mutantmice were immunized withthe T cell-depen-
dent (TD) antigen DNP-OVA. The number of germinal
centers per splenic section and the accumulation of
T cells in B follicles was similar between mutant and
wild-type mice 7 days after immunization (data not
shown). However, the relative titer of IgM was dramati-
cally reduced in the ?1 null mice 7, 14, and 21 days after
immunization (Figure 5A). IgG levels were slightly higher
in the ?1 null BM chimera, indicating that lack of ?1
integrin was not associated with a general attenuation
of the immune response (Figure 5B). A more detailed
increased in the sera of mutant mice, while IgG3 was
decreased (Figure 5C). Immunofluorescence analysis of
spleen sections revealed a significant reduction in IgM
plasma cells in spleen of ?1 null BM chimera, but no
obvious difference in their localization (Figures 5D and
in NP-Ova immunized mice. We found small numbers
of NP-specific plasma cells in the ?1 null BM chimera
that were scattered in the red pulp and did not form foci
associated with splenic vasculature as seen in normal
mice (data not shown).
Retention and Self-Renewal of HSC
It has been suggested that ?1 integrins on HSC are
important forHSC function(Prosper andVerfaille, 2001).
We tested this hypothesis by inducing a deletion of
the ?1 integrin gene on HSC in the BM of adult mice.
Southern blot and FACS analysis revealed a swift and
efficient loss of the ?1 gene and protein, respectively.
sis, normal number of BM cells, normal retention of HSC
In contrast to our findings, antibody and peptide inhi-
bition experiments had previously suggested that ?4
integrin, and to a lesser degree also ?5 integrin (Papa-
yannopoulou, 1995; van der Loo et al., 1998) and CD44
(Verfaillie et al., 1994; Vermeulen et al., 1998), are in-
volved in the retention of HPC and HSC in the BM.
However, such blocking studies bear the risk of un-
wanted side effects, e.g., by steric hindrance of other
ligand interactions or by partial activation of the target
cell. Furthermore, antibodies and protein fragments in-
jected into the tail vein of mice can bind to many other
integrin-expressing cells before reaching the HSC in the
BM. It could be, therefore, that such interactions indi-
rectly favored the release of HSC from the BM.
Studies of ?4-, ?5-, or ?v null chimeric mice, respec-
tively, revealed that none of these integrins alone is
(C–F) B cells (C and D) and T cells (E and F) from control (C and E) and mutant mice (D and F; all Ly-5.2) were injected into the tail vein of
B6SJL recipient mice (Ly-5.1). 4 hr later, the mice were sacrificed and spleen sections stained with antibodies against Ly-5.2 detecting the
transferred cells (red) and counterstained with Thy-1.2 (green; C and D) or antibodies against B220 (green; E and F). ?1-deficient cells showed
a similar distribution as transferred normal cells (200? magn.).
(G and H) Spleen sections of normal (G) and ?1 null BM chimeric mice (H; 6 months after induction of the knockout) were stained with
antibodies against B220 (green) and Thy-1.2 (red), detecting B and T cells, respectively. B and T cells show a normal distribution (100? magn.).
Figure 5. Decreased IgM and IgG3, but Increased IgG, IgG1, and IgG2a Production after Immunization of Control and ?1 Null BM Chimera
with a T Cell-Dependent Antigen
(A–C) Control (n ? 5) and ?1 null BM chimera (n ? 5) were i.p. injected with DNP-OVA. Serum levels of DNP-specific IgM (A) and IgG (B) were
determined by ELISA after 7, 14, and 21 days. Serum levels of IgG1, IgG2a, and IgG3 were determined after 21 days (C).
(D and E) Spleen sections of normal (D) and ?1 null BM chimeric mice (H) 7 days after immunization with DNP-OVA stained for IgM (200? magn.).
essential for the retention of HSC in the BM (Arroyo et
al., 2000). We extend these studies and show that even
simultaneous loss of both fibronectin receptors ?4?1
and ?5?1 in ?1-deficient HSC did not impair HSC reten-
tion and function. Although ?1 integrin is apparently not
necessary for HSC retention, it cannot be excluded that
the retention is reduced in the ?1 null BM chimera lead-
ing to an increased amount of HSC leaving the BM.
However, since loss of ?1 integrin on HPC did not lead
to a decreased number of colony forming units (CFU) in
does not seem likely. These data suggest that mutant
HSC use additional adhesive mechanisms mediated for
example by the fibronectin and hyaluronan receptor
CD44, which itself is not crucial for HSC retention and
maintenance as revealed in mice lacking the CD44 gene
(Schmits et al., 1997). It is possible that these alternative
adhesive mechanisms are not important under normal
conditions but are used in the mutant mice to compen-
sate for the loss of ?1 integrins. Double mutants should
allow testing of whether such compensatory mecha-
nisms are activated in our mouse strain.
Several earlier reports suggested that hematopoietic
progenitor cell attachment to fibronectin mediated by
?4?1 integrin affects their proliferation and survival in
vitro, although differently in different assay systems
(Schofield et al., 1998; Yokota et al., 1998; Hurley et al.,
1995; Wang et al., 1998). It was speculated that these
effects also play an important role in the self-renewal and
proliferation of HSC. However, we did not observe any
?1 Integrin and T Cell-Dependent IgM Response
Figure 6. Decreased IgM, IgG, IgG1, and
IgG3 Production after Immunization of Con-
trol and ?1 Null BM Chimera with a T Cell-
Independent Type 2 Antigen
(A–C) Control (n ? 4) and ?1 null BM chimera
(n ? 5) were i.p. injected with DNP-Ficoll. Se-
rum levels of DNP-specific IgM (A) and IgG
(B) were determined by ELISA after 7, 14, and
21 days. Serum levels of IgG1, IgG2a and
IgG3 were determined after 21 days (C).
(D) Splenic leukocytes from control (n ? 6)
and mutant animals (n ? 6) were incubated
for 3 and 6d in the presence or absence of
10 ?g/ml LPS. Cell supernatants were tested
for IgM and IgG levels by ELISA. Error bars
indicate standard deviation.
changes in the ?1-deficient mice in vivo, where the pres-
ence of ?1 null HSC and progenitor cells was indicated
cytes, erythrocytes, and platelets even 12 months after
deletion of the ?1 gene. Although we could not detect
any change in the size of B cell subsets in the BM,
homeostatic mechanismsmight concealdifferent prolif-
eration rates in the mutant animals (Agenes et al., 2000).
differentiation of these cells. One explanation for this
finding could be that ?4?1 and ?4?7 have redundant
functions in early hematopoiesis and only the absence
of both molecules impairs hematopoiesis. A double
knockout of ?1 and ?7 should allow confirmation of
this hypothesis. However, differences in the expression
pattern (Voura et al., 1997) and function of ?4?1 and
?4?7 argue against this hypothesis. For example, ?4?1
but not ?4?7 can promote activation of T cells binding
naling responses lead to T cell activation (Lehnert et al.,
1998). Encephalitogenic T cells express both ?4?1 and
?4?7. However, only antibodies against ?4?1 or VCAM-1,
but not against ?4?7, diminished the severity of the
disease (Engelhardt et al., 1998).
An alternative explanation is that loss of ?1 integrin
receptors in addition to ?4?1 rescues the ?4 null pheno-
observed with ?4 null progenitor cells could for example
be due to an increased adhesion to fibronectin via ?5?1.
Numerous studies of integrin-integrin crosstalks have
been described in the literature (Blystone et al., 1999,
and references therein).
The migration of T cell precursors from the BM to the
thymus was proposed to be at least partially mediated
The analysis of ?4 null chimeric mice revealed an impor-
tant role for ?4 integrin in lymphopoiesis (Arroyo et al.,
integrins, showed an early block in B cell development
and had defective T cell precursors that were unable to
leave the BM. In addition, no erythrocytes derived from
found in adult mutants. Since ?7 knockout mice have a
normal hematopoiesis (Wagner et al., 1996), these data
ment of hematopoietic progenitors (Wagner and Mu ¨ller,
1998). We now show that ?1 integrins including ?4?1
are not essential for lymphopoiesis, erythropoiesis, or
myelopoiesis. Our analysis of B cell development in BM
and spleen did not reveal any gross abnormality in the
by ?6?1 integrin (Ruiz et al., 1995). Furthermore, adhe-
sion andmigration of thymocyteswithin thethymus was
suggested to be dependent on a coordinated engage-
ment of ?4?1 and ?5?1 (Salomon et al., 1994; Crisa et
to migrate to and develop normally within the thymus.
However, we observed a temporary reduction in the
number of thymocytes 5 weeks after induction of the
knockout, which was confirmed by a decreased number
of ?1 null T cells in the fetal thymi 4 weeks after en-
graftment under the kidney capsule. Since the relative
numbers of thymocyte subpopulations were similar in
mutant and control mice, thereduced number of thymo-
cytes could be caused by a less efficient production or
emigration of ?1-deficient T cell precursors in the BM
or migration to the thymus. It is possible, therefore, that
?4?1 or ?6?1 are important for efficient emigration and
migration of T cell precursors, respectively. The transi-
tory nature of the phenotype can be explained by a
reduced requirement of the thymus for precursors, but
also indicative of a compensatory mechanism “re-
pairing” the defect.
function of these homeostatic cytokines, as docu-
mented by the normal migration behavior and organiza-
tion of the secondary lymphoid organs in mice with ?1-
to determine whether specific subsets of B and T cells
are affected by the loss of ?1 integrin expression.
The normal development of a primary TD immune re-
sponse is associated with an early production of serum
IgM and a later rise in IgG. The isotype switch away
from IgM is controlled by cell-cell contact between T
helper and B cells, by cytokines secreted by the helper
cells and by the number of divisions the B cell under-
goes. In fact, IgG-expressing plasmablasts can be de-
tected quite early in extrafollicular foci (Toellner et al.,
1998). A number of mice have been described that are
unable to switch isotype and produce only IgM, includ-
ing mice deficient in CD40, OCA-B, and Vav1 (Kawabe
et al., 1994; Kim et al., 1996; Gulbranson-Judge et al.,
1999). In contrast, no selective defect in the IgM re-
sponse has been reported to date and, so the ?1 integ-
rin-deficient mice exhibit a unique phenotype, secreting
OVA), but instead an increased amount of IgG sub-
classes. The signals that drive the IgM response to pro-
tein antigens are not known; indeed, this response has
often been viewed as relatively T cell independent, al-
cate that this is not the case and that specific cell inter-
actions are required to generate the primary IgM. The
enhanced levels of IgG in these mice demonstrate that
switchsignalsareavailableandthat thelossof ?1integ-
rin does not cause a general attenuation of the T cell-
dependent immune response. Summarized simply the
data suggest that a cell interaction vital for the initiation
or maturation of the IgM response is impaired in the ?1
integrin-deficient mice. The notion of a simple coactiva-
tor role of ?1-integrin on T cells, which was proposed
based on in vitro experiments (Yamada et al. 1991;
Damle and Aruffo, 1991), seems to be ruled out by these
A number of models could explain this phenotype.
First, signals necessary for the IgM response, normally
delivered during T-B cell interaction in the T zones of
secondary lymphoid tissues (Liu et al., 1992; Jacob and
Kelsoe, 1992), are missing in the ?1 integrin null chime-
ras. This might include interactions with accessory cells
as well as direct T-B interactions. Second, the lack of
?1 integrin signals could lead to increased delivery of
switch signals (by T cells or dendritic cells [DC]). It
should be noted that both IgG1 (Th2) and IgG2a (Th1)
to Th1 or Th2 in these mice, suggesting no obvious
change in the cytokine switch signals. Finally, the differ-
entiation or survival of IgM secreting plasma cells is
impaired in these mice, either directly due to lack of ?1
integrin signaling in B cells or indirectly because of a
?1 integrin requirement in the control of cell interactions
or production of survival factors from T cell or DC. This
last hypothesis cannot explain the increased amount of
IgG observed in the mutant mice.
We favor the first interpretation, although it is not
mutually exclusive with the second. If correct, this
Lymphocytes extravasate by weakly binding to and roll-
ing on the endothelium, followed by integrin activation,
integrin-mediated firm adhesion, and transmigration
through the endothelial cell layer and the underlying
basal lamina ([BL]; Moser and Loetscher, 2001). Al-
though ?1 integrin is crucial for the adhesion of HSC to
the endothelium of hematopoietic organs (Hirsch et al.,
1996; Potocnik et al., 2000), it was not known whether it
plays a similar role on differentiated lymphocytes. Since
antibodies against the BL component laminin reduced
lymphocyte emigration to lymph nodes in vivo (Kupiec-
rin-mediated binding of extravasating cells to BL com-
ponents like laminin is crucial for this process. We could
suggesting that ?1 integrin on lymphocytes is dispens-
able for firm adhesion to and transmigration through
endothelium and BL. Competitive lymphocyte migration
assays to spleen, lymph nodes, and Peyer’s patches
did not reveal significant differences in the migration of
?1-deficient and normal cells. Apparently, ?1-mediated
adhesion is dispensable for lymphocyte adhesion to en-
through the endothelial cell layer. It is possible that the
migration through the BL does not require binding to
this structure, but rather local protease activity enabling
tors and BL molecules (Friedl and Brocker, 2000).
Chemokine-mediated upregulation of integrin affinity
plays an important role in lymphocyte trafficking (Moser
and Loetscher, 2001). Targeted disruptions of genes for
chemokines or chemokine receptors, as for example
CCR7 (Fo ¨rster et al., 1999) and CXCR5 (Fo ¨rster et al.,
1996), cause defects in the homeostatic traffic of lym-
phocytes but also in the location of lymphocytes within
the secondary lymphoid tissues. Conceivably, all these
?1 integrins, however, seem not to be essential for the
?1 Integrin and T Cell-Dependent IgM Response
Preparation of genomic DNA and Southern blotting were carried out
EcoRI and probed with a fragment of the lacZ gene, which detects
only the targeted allele. Scanning was carried out using the Alpha-
Imager program (Alpha Innotech, San Leandro, USA).
strongly predicts that the signals necessary for initial
B cell activation and IgM secretion are delivered in a
quite distinct microenvironment from those that control
isotype switching. This conflicts with recent data sug-
gesting that switch signals are delivered very early after,
and hence in the same sites as, initial B cell activation
(Toellner et al., 1998). Preliminary investigation of the
IgM defect by transfer of wild-type B cells and/or T cells
into nonirradiated ?1 null BM chimeras showed that
neither B nor T cells could rescue the impaired IgM
response (data not shown). This suggests that ?1 integ-
rin is not involved in the direct collaboration of B and
T cells but possibly is important for the interaction of B
or T cells with an accessory cell. In relation to this, it is
interesting to note that dendritic cells in the spleens of
?1-deficient chimeras showed a complete absence of
?1 integrin expression, in contrast to the incomplete
deletion in the lymphocyte compartment (see above).
Future adoptive transfer analyses will focus on reconsti-
tution of the dendritic cells in these mice.
When we studied the response of the ?1 integrin null
BM chimeras to a TI-2 antigen (DNP-Ficoll) we found
both the IgM and IgG3 responses were impaired, al-
though the loss of IgM was not as severe as in the
TD response. TI-2 antigens are large, repeating epitope
polysaccharides that stimulate antibody production by
extensive and prolonged crosslinking of B cell recep-
tors. It has been proposed that a second signal is re-
quired for optimal activation and differentiation to
plasma cells and some have been identified; e.g., inter-
feron-? (Snapper et al., 1992) and TACI (von Bulow et
al., 2001; Yan et al., 2001). Both B1 B cells and marginal
zone (MZ) B cells take part in TI-2 responses (Martin et
al., 2001), in sites distinct from the T-dependent re-
sponse. The results shown here could indicate that ei-
ther a ?1 integrin acts as a direct costimulus for B cells
in TI-2 responses or that a cell interaction is lacking in
the mutant mice due to aberrant adhesion/migration.
Interestingly, NK and NK-T cells have been implicated
in TI-2 responses (Mond et al., 1995) and both cell types
express high amounts of ?1 integrin (L. Fahle ´n and C.B.,
unpublished data). A deficiency in ?1 integrin may pre-
vent delivery of NK- or NK-T cell-derived help to CD1-
expressing B cells. It is also possible that the T-depen-
dent and T-independent defects have a common cause.
For instance, the expansion of B cell blasts and their
differentiation into plasma cells in the extrafollicular foci
of response (Garcia de Vinuesa et al., 1999). This is
certainly true for TD IgM and TI-2 IgM and IgG, while
TD IgG responses may be propagated elsewhere (e.g.,
BM). ?1 integrin may be important for the formation of
these foci. The ?1 integrin null chimeras are a unique
resource, allowing for the first time a systematic dissec-
tion of the lymphocyte-accessory cell interactions that
occur during the early T-dependent antibody response,
as well as those involved in TI responses.
Single cell suspension were prepared by gently pushing the dis-
sected organs through 70 ?m cell strainers (BD). FACS analysis
was performed as described (Potocnik et al., 2000). The following
antibodies were used: hamster anti-?1 integrin (H?2/5), rat anti-?2
integrin (C71/16), rat anti-?3 integrin (2C9.G2), rat anti-?4 integrin
(346-11A), rat anti-?7 integrin (M293), hamster anti-?1 integrin
(Ha31/8), hamster anti-?2 integrin (Ha1/29), rat anti-?4 integrin
(R1-2), rat anti-?5 integrin (5H10-27), rat anti-?6 integrin (GoH3; all
Erlangen, Germany), rat anti ?v-integrin (H9.2B8), rat anti-B220
(RA3-6B2), rat anti-CD19 (1D3), rat anti-CD25 (7D4), rat anti-IgM
(B3B4), rat anti-CD4 (H129.19), rat anti-CD8 (53-6.2), rat anti-TCR
(H57-597), rat anti-CD44 (IM7), rat anti-Gr-1 (RB6-8C5), rat anti-
Mac-1 (M1-70), Ter-119, rat anti-CD62L (MEL-14), rat anti-CD69
(H1.2F3), rat anti-CD3 (17A2), rat anti-Ly-5.1 (A20), rat anti-Ly-5.2
(104), rat anti Ly-9.1 (30C7; all Pharmingen). Antibodies were uncon-
jugated or conjugated with FITC, PE, or biotin and used at 1:200
dilution. Unconjugated antibodies were detected by FITC-conju-
gated rat anti-hamster IgG (G70-204,G94-56, Pharmingen, 1:200 di-
lution), FITC-conjugated goat anti-rat IgG or FITC-conjugated goat
anti-mouse IgG (both from Jackson Immunoresearch, 1:200 dilu-
tion). Biotinylatyed antibodies were detected by streptavidin PE
son Immunoresearch, 1:500 dilution).
Colony Formation Assay
Blood was collected from the retroorbital bulbus of mutant and
control mice. Erythrocytes were lysed with ACK buffer (Coligan et
al., 1995). 600000 (PB) or 60000 (BM) leukoytes in 300 ?l Iscove’s
MDM with 2% FCS were mixed with 3 ml MethoCult GF M3434 (Stem-
Cell Technologies, Vancouver, Canada), containing Epo, IL-3, IL-6,
and SCF, and plated into three 35 mm culture dishes. Colonies were
counted after 8–9 days incubation at 37?C, 5% CO2. preB cell colony
assays were performed similarly using 50000 BM leukocytes per
dish and MethoCult M3630 (StemCell Technologies) containing IL-7.
T Cell Migration Assay
BM cells from (fl/? cre) or (fl/– cre) mice (Ly-9.1?) were depleted by
MACS (Miltenyi Biotech) for lineage marker positive cells and in-
jected into lethally irradiated C57BL/6J-IghaThy1aGpi1a recipients
(10000 cells/mouse). 4 weeks after transfer, ?1 integrin deletion was
induced as described above. 4 weeks later, 3–4 alymphoid lobes
from E15.5 C57BL/6 Ly-5.1 mice (Jenkinson et al., 1982) were trans-
planted under the kidney capsule of ?1-deficient BM chimeras with
at least 95% knockout. 4–5 weeks after engraftment the cellular
composition of the thymi was analyzed by FACS.
Lymphocytes were isolated from spleen of polyIC treated (fl/– cre)
were purified by MACS (Thy-1.2 or B220 beads; Miltenyi Biotech).
The purity was about 90% in the enriched and 85% in the depleted
fractions. Lymphocyte fractions of control and knockout mice, one
of them labeled with the intracellular tracking dye CMTMR (Molecu-
lar Probes, USA), were mixed at a ratio of 1:1 and injected into the
tail vein of B6SJL (Ly-5.1) mice. After 4 or 16 hr, lymphocytes were
isolated from spleen, lymph nodes (inguinal, axial, paraaortic), and
Peyer’s patches. The ratio of control to mutant cells in these organs
was determined by FACS using Ly-5.2 staining to specifically detect
the transferred cells. In a second experiment, splenocytes were
directly labeled with CMTMR, mixed with unlabeled cells, and in-
neous staining with antibodies against Ly-5.2 and either B220
Generation of Mice with a Deletion of the ?1 Integrin Gene
in the Hematopoietic System
Mice with a deletion of the ?1 integrin gene in the hematopoietic
system were generated as described previously and bred in a con-
ventinal animal facility (Nieswandt et al., 2001).
either mutant or control lymphocyte fractions were injected. Trans-
ferred Ly-5.2?cells were detected on cryo-sections by immuno-
Strober, W. (1995). Current Protocols in Immunology, (Wiley and
Crisa, L., Cirulli, V., Ellisman, M.H., Ishii, J.K., Elices, M.J., and Salo-
mon, D.R. (1996). Cell adhesion and migration are regulated at dis-
tinct stages of thymic T cell development: the roles of fibronectin,
VLA4, and VLA5. J. Exp. Med. 184, 215–228.
Damle, N.K., and Aruffo, A. (1991). Vascular cell adhesion molecule
1 induces T-cell antigen receptor-dependent activation of CD4?T
lymphocytes. Proc. Natl. Acad. Sci. USA 88, 6403–6407.
Engelhardt, B., Laschinger, M., Schulz, M., Samulowitz, U., Vest-
weber, D., and Hoch, G. (1998). The development of experimental
autoimmune encephalomyelitis in the mouse requires ?4-integrin
but not ?4?7-integrin. J. Clin. Invest. 102, 2096–2105.
Fa ¨ssler, R., and Meyer, M. (1995). Consequences of lack of ?1 integ-
rin gene expression in mice. Genes Dev. 9, 1896–1908.
Fo ¨rster, R., Mattis, A.E., Kremmer, E., Wolf, E., Brem, G., and Lipp,
M. (1996). A putative chemokine receptor, BLR1, directs B cell mi-
gration to defined lymphoid organs and specific anatomic compart-
ments of the spleen. Cell 87, 1037–1047.
Fo ¨rster, R., Schubel, A., Breitfeld, D., Kremmer, E., Renner-Mu ¨ller, I.,
Wolf, E., and Lipp, M. (1999). CCR7 coordinates the primary immune
response by establishing functional microenvironments in second-
ary lymphoid organs. Cell 99, 23–33.
Friedl, P., and Brocker, E.B. (2000). T cell migration in three-dimen-
Immunol. 7, 249–266.
Garcia de Vinuesa, C., O’Leary, P., Sze, D.M., Toellner, K.M., and
MacLennan,I.C. (1999).T-independenttype 2antigensinduce Bcell
proliferation in multiple splenic sites, but exponential growth is con-
fined to extrafollicular foci. Eur. J. Immunol. 29, 1314–1323.
Giancotti, F.G., and Ruoslahti, E. (1999). Integrin signaling. Science
Gonzalez-Amaro, R., and Sanchez-Madrid, F. (1999). Cell adhesion
molecules: selectins and integrins. Crit. Rev. Immunol. 19, 389–429.
Gray, D., Dullforce, P., and Jainandunsing, S. (1994). Memory B cell
development but not germinal centre formation is impaired by in
vivo blockade of CD40-CD40 ligand interaction. J. Exp. Med. 180,
Groux, H., Huet, S., Valentin, H., Pham, D., and Bernard, A. (1989).
cule of CD4? lymphocytes. Nature 339, 152–154.
Gulbranson-Judge, A., Tybulewicz, V.L., Walters, A.E., Toellner,
K.M., MacLennan, I.C.,and Turner, M. (1999). Defective immuno-
globulin class switching in Vav-deficientmice is attributable to com-
promised T cell help. Eur. J. Immunol. 29, 477–487.
Hirsch, E., Iglesias, A., Potocnik, A.J., Hartmann, U., and Fa ¨ssler, R.
(1996). Impaired migration but not differentiation of haematopoietic
stem cells in the absence of ?1 integrins. Nature 380, 171–175.
in man: LAD-1 and other syndromes. Matrix Biol. 19, 211–222.
sion to BM stroma via fibronectin receptors inhibits hematopoietic
progenitor proliferation. J. Clin. Invest. 96, 511–519.
Hynes, R.O. (1992). Integrins: versatility, modulation, and signaling
in cell adhesion. Cell 69, 11–25.
Jacob, J., and Kelsoe, G. (1992). In situ studies of the primary im-
mune response to (4-hydroxy-3-nitrophenyl)acetyl. II. A common
clonal origin for periarteriolar lymphoid sheath-associated foci and
germinal centers. J. Exp. Med. 176, 679–687.
Jenkinson, E.J., Franchi, L.L., Kingston, R., and Owen, J.J. (1982).
Effect of deoxyguanosine on lymphopoiesis in the developing thy-
mus rudiment in vitro: application in the production of chimeric
thymus rudiments. Eur. J. Immunol. 12, 583–587.
Kawabe, T., Naka, T., Yoshida, K., Tanaka, T., Fujiwara, H.,
Suematsu, S., Yoshida, N., Kishimoto, T., and Kikutani, H. (1994).
The immune responses in CD40-deficient mice: impaired immuno-
globulin class switching and germinal center formation. Immunity
Kim, U., Qin, X.F., Gong, S., Stevens, S., Luo, Y., Nussenzweig, M.,
Immunofluorescence was performed as described (Ko ¨rner et al.,
ated rat anti-mouse CD90.2 (clone 53-2.1, Pharmingen; T24, ATCC),
sheep anti-mouseIgD (TheBinding Site,Birmingham, UK)and texas
redconjugated goatanti-mouse IgM(Southern Biotechnology)were
used. The staining was developed with goat anti-rat Alexa488, goat
anti-rat Alexa594, streptavidin-Alexa488, streptavidin-Alexa594 (all
Molecular Probes), streptavidin texas red (Southern Biotechnology),
FITC-conjugated Fab mouse anti-rat IgG (Jackson Immunore-
centers were identified by FITC conjugated PNA (Vector labora-
tories). NP(nitrophenyl)-specific plasma cells were detected on ace-
tone-fixed cryostat sections using NP conjugated to sheep IgG (a
gift from Prof. Ian MacLennan, University of Birmingham, UK), which
in turn was detected with donkey anti-sheep IgG (The Binding Site,
Birmingham, UK) and then streptavidin-FITC (Southern Biotech-
Mice were immunized via the intraperitoneal route with 100 ?g of
alum precipitated dinitrophenylated (DNP) ovalbumin (OVA) (Sigma)
or with 100 ?g of DNP-Ficoll (Biosearch Technologies, USA). ELISA
assays were carried out as described (Gray et al., 1994).
In Vitro Stimulation of B Cells
Splenic leukocyteswere cultured in 24-wellplates at 1.5 ?106cells/
well, 1 ml/well in RPMI 1640, 10% FCS, 50 ?M 2-mercaptoethanol,
10 mM Hepes (pH 7.4), 2 mM glutamin, 0.1 mM nonessential amino
acids, 1 mM sodium pyruvate, 100 ?g/ml penicillin, 100 ?g/ml strep-
Cell supernatants were analyzed by ELISA.
We would like to thank Mrs. Alison Lundquist for expert technical
son and Shohreh Issazadeh for initial help with the FACS analysis,
Drs. Ralf Ku ¨hn and Werner Mu ¨ller for the Mx-cre mice, and Dr.
Antonio Iglesias for critically reading the manuscript and excellent
advice. This work was supported by the Swedish Research Council,
the inflammation program of the SSF, and the Go ¨ran Gustafsson
Institute of Immunology was founded and is supported by F. Hoff-
man-La Roche, Ltd., Basel, Switzerland.
Received August 10, 2001; revised January 25, 2002.
Agenes, F., Rosado, M.M., and Freitas, A.A. (2000). Considerations
on B cell homeostasis. Curr. Top. Microbiol. Immunol. 252, 68–75.
Arroyo, A.G., Yang, J.T., Rayburn, H., and Hynes, R.O. (1996). Differ-
esis. Cell 85, 997–1008.
Arroyo, A.G., Yang, J.T., Rayburn, H., and Hynes, R.O. (1999). ?4
eage hematopoietic progenitors in vivo. Immunity 11, 555–566.
Arroyo, A.G., Taverna, D., Whittaker, C.A., Strauch, U.G., Bader,
B.L., Rayburn, H., Crowley, D., Parker, C.M., and Hynes, R.O. (2000).
In vivo roles of integrins during leukocyte development and traffic:
insights from the analysis of mice chimeric for ?5, ?v, and ?4 integ-
rins. J. Immunol. 165, 4667–4675.
Blystone, S.D., Slater, S.E., Williams, M.P., Crow, M.T., and Brown,
E.J. (1999). A molecular mechanism of integrin crosstalk: ?v?3 sup-
pression of calcium/calmodulin-dependent protein kinase II regu-
lates ?5?1 function. J. Cell Biol. 145, 889–897.
Coligan, J.E., Kruisbeek, A.M., Margulies, D.H., Shevach, E., and
?1 Integrin and T Cell-Dependent IgM Response
and Roeder, R.G. (1996). The B-cell-specific transcription coactiva-
noglobulin isotypes. Nature 383, 542–547.
Ko ¨rner, H., Cretney, E., Wilhelm, P., Kelly, J.M., Rollinghoff, M.,
Sedgwick, J.D., and Smyth, M.J. (2000). Tumor necrosis factor sus-
tains the generalized lymphoproliferative disorder (gld) phenotype.
J. Exp. Med. 191, 89–96.
Ku ¨hn, R., Schwenk, F., Aguet,M., and Rajewsky, K. (1995). Inducible
gene targeting in mice. Science 269, 1427–1429.
is modified in vivo by anti-laminin antibody. Immunology 72,
Lehnert, K., Print, C.G., Yang, Y., and Krissansen, G.W. (1998).
MAdCAM-1 costimulates T cell proliferation exclusively through in-
tegrin ?4?7, whereas VCAM-1 and CS-1peptide use ?4?1: evidence
for “remote” costimulation and induction of hyperresponsiveness
to B7 molecules. Eur. J. Immunol. 28, 3605–3615.
Liu, Y.-J., Lane, P.J.L., Chan, E.Y.-T., and MacLennan, I.C.M. (1992).
Sites of specific B cell activation in primary and secondary re-
sponses to T cell-dependent and T cell-independent antigens. Eur.
J. Immunol. 21, 2951–2962.
Martin, F., Oliver, A.M., and Kearney, J.F. (2001). Marginal zone and
B1 B cells unite in the early response against T-independent blood-
borne particulate antigens. Immunity 14, 617–629.
Mond, J.J., Vos, Q., Lees, A., and Snapper, C.M. (1995). T cell inde-
pendent antigens. Curr. Opin. Immunol. 7, 349–354.
Moser, B., and Loetscher, P. (2001). Lymphocyte traffic control by
chemokines. Nat. Immunol. 2, 123–128.
Nieswandt, B.,Brakebusch, C.,Bergmeier, W., Schulte,V., Bouvard,
D., Mokhtari-Nejad, R., Lindhout, T., Heemskerk, J.W., Zirngibl, H.,
tial for platelet interaction with collagen. EMBO J. 20, 2120–2130.
Oostendorp, R.A., Reisbach, G., Spitzer, E., Thalmeier, K., Diene-
mann, H., Mergenthaler, H.G., and Dormer, P. (1995). VLA-4 and
VCAM-1 are the principal adhesion molecules involved in the inter-
action between blastcolony-forming cells and BM stromal cells. Br.
J. Haematol. 91, 275–284.
Papayannopoulou, T., Craddock, C., Nakamoto, B., Priestley, G.V.,
and Wolf, N.S. (1995). The VLA4/VCAM-1 adhesion pathway defines
contrasting mechanisms of lodgement of transplanted murine he-
mopoietic progenitors between BM and spleen. Proc. Natl. Acad.
Sci. USA 92, 9647–9651.
Potocnik, A.J., Brakebusch, C., and Fa ¨ssler, R. (2000). Fetal and
adult hematopoietic stem cells require ?1 integrin function for colo-
nizing fetal liver, spleen, and BM. Immunity 12, 653–663.
Prosper, F., and Verfaillie, C.M. (2001). Regulation of hematopoiesis
through adhesion receptors. J. Leukoc. Biol. 69, 307–316.
Ruiz, P., Wiles, M.V., and Imhof, B.A. (1995). ?6 integrins participate
in pro-T cell homing to the thymus. Eur. J. Immunol. 25, 2034–2041.
Salomon, D.R., Mojcik, C.F., Chang, A.C., Wadsworth, S., Adams,
of integrin ?4?1 defines a unique stage of human thymocyte devel-
opment. J. Exp. Med. 179, 1573–1584.
Schmits, R., Filmus, J., Gerwin, N., Senaldi, G., Kiefer, F., Kundig,
T., Wakeham, A., Shahinian, A., Catzavelos, C., Rak, J., et al. (1997).
CD44 regulates hematopoietic progenitor distribution, granuloma
formation, and tumorigenicity. Blood 90, 2217–2233.
Schofield, K.P., Humphries, M.J., de Wynter, E., Testa, N., and Gal-
lagher, J.T. (1998). The effect of ?4?1-integrin binding sequences
tors. Blood 91, 3230–3238.
Sixt, M., Engelhardt, B., Pausch, F., Hallmann, R., Wendler, O., and
Sorokin, L.M. (2001). Endothelial cell laminin isoforms, laminins 8
and 10, play decisive roles in T cell recruitment across the blood-
brain barrier in experimental autoimmune encephalomyelitis. J. Cell
Biol. 153, 933–946.
Snapper, C.M., McIntyre, T.M., Mandler, R., Pecanha, L.M., Finkel-
man, F.D., Lees, A., and Mond, J.J. (1992). Induction of IgG3 secre-
tion by interferon gamma: a model for T cell-independent class
switching in response to T cell-independent type 2 antigens. J. Exp.
Med. 175, 1367–1371.
Ticchioni, M., Aussel, C., Breittmayer, J.P., Manie, S., Pelassy, C.,
and Bernard, A. (1993). Suppressive effect of T cell proliferation via
and phosphatidic acid levels in activated T cells. J. Immunol 151,
Toellner, K M., Luther, S.A., Sze, D.M., Choy, R.K., Taylor, D.R.,
MacLennan, I.C., and Acha-Orbea, H. (1998). T helper 1 (Th1) and
Th2 characteristics start to develop during T cell priming and are
switching. J. Exp. Med. 187, 1193–1204.
van der Loo, J.C., Xiao, X., McMillin, D., Hashino, K., Kato, I., and
Williams, D.A. (1998). VLA-5 is expressed by mouse and human
long-term repopulating hematopoietic cells and mediates adhesion
to extracellular matrix protein fibronectin. J. Clin. Invest. 102, 1051–
Verfaillie, C.M., Benis, A., Iida, J., McGlave, P.B., and McCarthy, J.B.
(1994). Adhesion of committed human hematopoietic progenitors to
synthetic peptides from the C-terminal heparin-binding domain of
fibronectin: cooperation between the integrin ?4?1 and the CD44
adhesion receptor. Blood 84, 1802–1811.
Vermeulen, M., Le Pesteur, F., Gagnerault, M.C., Mary, J.Y., Sain-
teny, F., and Lepault, F. (1998). Role of adhesion molecules in the
tor cells. Blood 92, 894–900.
von Bulow, G., van Deursen, J.M., and Bram, R.J. (2001). Regulation
of the T-independent humoral response by TACI. Immunity 14,
Voura, E.B., Billia, F., Iscove, N.N., and Hawley, R.G. (1997). Expres-
sion mapping of adhesion receptor genes during differentiation of
individual hematopoietic precursors. Exp. Hematol. 25, 1172–1179.
Wagner, N., and Mu ¨ller, W. (1998). Functions of ?4- and ?7-integrins
in hematopoiesis, lymphocyte trafficking and organ development.
Curr. Top. Microbiol. Immunol. 231, 23–32.
Wagner, N., Lohler, J., Kunkel, E.J., Ley, K., Leung, E., Krissansen,
G., Rajewsky, K., and Mu ¨ller, W. (1996). Critical role for ?7 integrins
in formation of the gut-associated lymphoid tissue. Nature 382,
Wang, M.W., Consoli, U., Lane, C.M., Durett, A., Lauppe, M.J.,
apoptosis in early (CD34-selected) versus late (non-CD34-selected)
human hematopoietic cells by very late antigen 4- and vascular cell
adhesion molecule (VCAM) 1-dependent adhesion to bone marrow
stromal cells. Cell Growth Differ. 9, 105–112.
Williams, D.A., Rios, M., Stephens, C., and Patel, V.P. (1991). Fibro-
nectin and VLA-4 in haematopoietic stem cell-microenvironment
interactions. Nature 352, 438–441.
Yamada, A., Nojima, Y., Sugita, K., Dang, N.H., Schlossman, S.F.,
and Morimoto, C. (1991). Cross-linking of VLA/CD29 molecule has
a co-mitogenic effect with anti-CD3 on CD4 cell activation in serum-
free culture system. Eur. J. Immunol. 21, 319–325.
Yan, M., Wang, H., Chan, B., Roose-Girma, M., Erickson, S., Baker,
T., Tumas, D., Grewal, I.S., and Dixit, V.M. (2001). Activation and
accumulation of B cells in TACI-deficient mice. Nat. Immunol. 2,
H., Matsumura, I., Tsai, S., Tomiyama, Y., Kanakura, Y., and Matsu-
ment for fibronectin activities of CS1 and cell-binding domains.
Blood 91, 3263–3272.