| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Adult Oncology, Dana-Farber Cancer Institute and
Division of Dermatology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA 02115; and
Department of Microbiology, University of Washington, Seattle, WA 98195
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Shortly after the discovery that the B7-1 molecule was a high affinity ligand for CTLA-4 and low affinity ligand for CD28 (2), several lines of evidence suggested that additional CD28/CTLA-4 counterreceptors existed. The B7-1 (CD80) molecule was initially defined by two mAb termed 133 (B7-1) and BB1 (3, 4). These mAb both bound to B7-1 transfectants, cross-blocked each other’s binding, and blocked CD28-mediated costimulation, indicating that they both bound to B7-1. Of note, the BB1 mAb was instrumental in the discovery that CD28 is a counterreceptor for B7-1 (2), and indeed for several years B7-1 was often referred to in the literature as B7/BB1. In spite of the common features shared by the B7-1 and BB1 mAbs, a number of conflicting observations were reported. As noted in our initial characterization of B7, the B7-1 and BB1 mAbs consistently differed in their binding to some cell lines (3). From our studies of CTLA4Ig binding to activated B cells, we proposed the existence of a second CTLA-4 counterreceptor that was present on B cells early following activation and did not stain with either anti-B7-1 or BB1 mAbs (5). These observations culminated in the cloning of B7-2, a second CD28/CTLA-4 counterreceptor (6, 7). B7-2 transfectants did not stain with either B7-1 or BB1 mAbs but did bind both CD28-Ig and CTLA4Ig (6). We also identified a population of activated B cells that did not stain with anti-B7-1 but did stain with the BB1 mAb and CTLA4Ig (5). We termed this molecule B7-3.
Although the discovery of B7-2 as well as the clustering of B7-1 into the CD80 cluster (8) and B7-2 into the CD86 cluster (9) significantly advanced our understanding of CD28 and CTLA-4 counterreceptors, the earlier phenotypic and functional studies with anti-B7-1 and BB1 Abs still suggested that another molecule might exist. In fact, the Fifth International Workshop in 1993 reported that, by cellular expression and functional analysis, it was not possible to exclude the existence of a distinct molecule (8). While clearly reacting with B7-1 protein, the BB1 mAb has been shown to differ from other anti-B7-1 mAbs in Western blotting (10), chromosomal location (11, 12), and reactivity with thymic epithelial cells (13). Activated keratinocytes bound the BB1 Ab but not other B7-1 mAbs and did not express B7-1 mRNA (13). Surprisingly, activated keratinocytes were reported to adhere to CD28 transfectants. These studies further supported our previous concern that the BB1 mAb was binding to an additional molecule. Therefore, our working hypothesis was that the BB1 mAb bound to an epitope on a distinct protein and that this epitope was also expressed on the B7-1 protein.
In the present report, we have cloned a BB1-binding molecule distinct from CD80 and demonstrate its identity to the cell surface form of CD74, MHC class II-associated invariant chain. Many of the puzzling observations in the literature concerning the expression of B7-1 are resolved by an understanding that BB1 staining is the summation of CD74 plus B7-1 expression. Since CD74 is coordinately expressed with MHC class II, staining with BB1 mAb has given the impression that B7-1 is more coordinately expressed with MHC class II than is actually the case. This observation requires that the field reconsider the clustering of anti-B7-1 and BB1 mAbs into the CD80 cluster and refine its characterization of the expression of B7-1.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
mAbs were used as purified Ig unless otherwise indicated. BB1 mAb (IgM) (4) was prepared by one of us (E.A.C.) and was also purchased as purified and FITC-conjugated BB1 from Ancell (Bayport, MN). The anti-B7-1 mAbs used were: 133 (IgM) (3); EW3.4B2.C4 (IgG2a), EW3.1A5.F3 (IgG2a), and EW3.6B9 (IgG2a) (Repligen, Cambridge, MA); B7g (IgG1), 49 (IgG1), and L307.4 (IgG1, this mAb was sold for a time by Becton Dickinson as "BB1" but has no relation to the original BB1 mAb and was generated by immunization with B7-1 transfectants) (Leukocyte Typing V Workshop, Boston, MA); and 104 (IgG1; Dr. J. Banchereau, Dardilly, France) (14). The anti-B7-2 mAb used was HA3.1F9 (IgG1, Repligen) (15). Anti-CD74 mAbs LN2 (IgG1), M-B741 (IgG2a), and M-B741-FITC were obtained from PharMingen (San Diego, CA). CD28-Ig was a generous gift from Dr. P. Linsley (Bristol-Myers Squibb, Seattle, WA) (16). CTLA4Ig and control Ig fusion proteins were obtained from Repligen. Other mAbs used include anti-MHC I (W6/32; IgG2a), anti-CD11a (Mo1; IgM), anti-CD14 (Mo2; IgM), and anti-CD56 (3B8; IgM). For flow cytometric analysis, where indicated, fluorochrome-conjugated mAbs were prepared. FITC-conjugated anti-CD10 (J5; IgG2a) was obtained from Coulter (Hialeah, FL). FITC-conjugated goat anti-mouse IgG and IgM, FITC-conjugated goat anti-human IgG, and FITC-conjugated streptavidin were purchased from Southern Biotechnology (Birmingham, AL).
Phenotypic analysis
Expression of cell surface molecules was determined by direct or indirect labeling using standard methodology. For lymphocytes, Fc receptors were blocked by incubation with mouse Ig (for direct staining) or goat Ig (for indirect staining) before the addition of the specific mAbs. Irrelevant isotype-matched Abs (IgG subclasses or IgM) or control fusion protein were used as negative controls. FITC-conjugated goat anti-mouse IgG or IgM and FITC-conjugated goat anti-human IgG were used for unconjugated mAbs and Ig-fusion proteins, respectively. Samples were analyzed in a Coulter Elite flow cytometer.
Ab-binding competition
Cells were incubated with 10 µg/ml of unconjugated competitor mAbs for 30 min at 4°C, washed twice with PBS/0.1% BSA, and then incubated with FITC-conjugated CD74, CD10, or CD19. After 30 min incubation at 4°C, the cells were washed once and analyzed by flow cytometry.
Ab absorption
BB1 mAb (0.5 ml of 10 µg/ml) was absorbed by incubation with 5 x 106 of either B7-1-transfected Chinese hamster ovary (CHO)3 cells (CHO/B7-1) or Neo vector-transfected CHO cells (CHO/Neo) (17) for 30 min at 4°C. Cells were pelleted by centrifugation and the absorption repeated with a fresh 5 x 106 cells.
cDNA library
RNA was prepared from NALM-6 cells and oligo(dT) selected twice.
An oriented cDNA library was constructed in the pAXEF vector with an
oligo(dT15)-Not primer and SalI adapter using
the superscript cDNA synthesis kit according to the manufacturer’s
instructions (Life Technologies, Gaithersburg, MD). The pAXEF mammalian
expression vector contains the elongation factor 1
(EF1) promoter
and intron, multiple cloning site, supF, ß-globin polyadenylation
site, SV40 origin, and ColEI origin. It was constructed using DNA
fragments from pEF-BOS (18), p142B, and pAX114 (19), and the sequence
is available upon request. The library was introduced into DH10B/P3
Escherichia coli, and a total of 7 x 108
independent transformants were generated.
COS cell transfection and molecular cloning
For the first round of selection, COS cells were transfected via the DEAE-Dextran procedure (20) with 0.2 µg of NALM-6 plasmid library DNA per 100-mm dish. After 40 h, cells were harvested, incubated with 5 µg/ml BB1 mAb, washed, and panned on anti-IgM-coated plates as described (20, 21). Episomal DNA was prepared from adherent cells, reintroduced into E. coli, and transfected into COS cells by polyethylene glycol-mediated fusion of spheroplasts (20). The panning with BB1 mAb was repeated. Individual plasmid DNAs were transfected into COS cells via the DEAE-Dextran procedure (4 µg per 100-mm dish) and analyzed after 72 h for cell surface expression by indirect immunofluorescence and flow cytometry.
RNA blot hybridizations
RNA was prepared and analyzed by blot hybridization as
previously described (22). The 1.3-kb CD74 cDNA insert was labeled by
random oligonucleotide priming using -32P-labeled dCTP
and the Klenow fragment of DNA polymerase. Hybridization, washing, and
autoradiography were performed as previously described (22).
Keratinocytes
Normal human keratinocytes were derived from neonatal foreskins
by trypsinization, were cultured in low calcium (0.15 mM) keratinocyte
growth medium (Clonetics, San Diego, CA), and were used at passages 2
to 4 at 70 to 80% confluency. Keratinocytes were treated with 100 U/ml
IFN-
(Genzyme, Cambridge, MA) or 10 ng/ml PMA (Sigma, St.
Louis, MO).
B cell precursor acute lymphoblastic leukemia samples
B cell precursor ALL cells (pre-B ALL) were obtained from the bone marrow of patients with leukemia with greater than 90% marrow infiltration. Both fresh and cryopreserved samples were used in these studies. Appropriate informed consent and Institutional Review Board approval were obtained for all sample collections. Samples were enriched by density centrifugation over Ficoll-Hypaque, then washed twice in RPMI 1640 supplemented with 10% FCS and 2 mM L-glutamine.
Stimulation of B cell precursor ALL cells by CD40 cross-linking
National Institutes of Health-3T3 cells stably transfected with human CD40 ligand (t-CD40L) or vector (t-mock), have been described (23, 24). The t-CD40L or t-mock cells were irradiated at 96 Gy and plated at 1 x 105 cells/well in 24-well plates or 4 x 105 cells/well in 6-well plates, in media without G418, and incubated overnight at 37°C in a 5% CO2 humidified atmosphere. The plates were washed twice with PBS, and pre-B ALL cells were added at 5 x 105 cells/ml in IMDM supplemented with 2% FCS, 0.5 mg/ml BSA, 2 mM L-glutamine, 50 µg/ml iron-saturated transferrin, 5 µg/ml bovine insulin, and 2 ng/ml of recombinant human IL-4 (a gift of Immunex, Seattle, WA). On day 5, CD40-stimulated pre-B ALL cells were harvested, washed in IMDM, and used for phenotypic analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The CD80 cluster was defined in the Fifth International Workshop on Leukocyte Antigens on the basis of the reactivity of CHO/B7-1 with five mAbs (8) including the original B7-1 mAb (clone 133) (3) and the BB1 mAb (4). Nevertheless, in the original characterization of the B7-1 mAb, we had distinguished B7-1 mAb from BB1 by the capacity of BB1 but not B7-1 to stain the B cell line, BALM4 (3). One possible source of this different reactivity is that the immunogen used to generate the BB1 mAb was a baboon B cell line (4), and, consequently, the binding site might not precisely fit the human B7-1 molecule in the same way as mAb generated using human B7-1. Furthermore, BB1 mAb is an IgM and presumably has only a low affinity since it has not undergone affinity maturation.
To resolve the controversy of whether the BB1 mAb recognizes an
additional protein, we analyzed an additional pre-B cell line and
primary pre-B cell tumors. We observed that the pre-B cell line NALM-6
and 10 of 12 pre-B ALL stained positively with the BB1 mAb, although
none of them stained with the four other mAbs from the CD80 panel of
the Fifth Workshop (133, B7g, 49, L307) or other anti-B7-1 mAbs
(EW3.4B2.C4, EW3.1A5.F3, 104). The BB1 mAb consistently stained pre-B
ALL and NALM-6 cells with low to intermediate intensity (range 29–85%
positive cells for pre-B ALL samples). Cell staining profiles for a
representative pre-B ALL (ALL-12) are shown in Figure 1
. CHO/B7-1 cells were also stained as
positive controls and reacted with the BB1 mAb and all seven
anti-B7-1 mAbs, confirming that BB1 mAb does indeed recognize the
B7-1 molecule. CHO/B7-1 cells bound CD28-Ig and CTLA4Ig but did not
stain with anti-B7-2 mAb. Although approximately 60% of pre-B ALL
samples stain with anti-B7-2 mAb, both ALL-12 and NALM-6 did not.
ALL-12 and NALM-6 bound BB1 but did not bind any of the other seven
B7-1 mAb. In contrast to CHO/B7-1, neither ALL-12 nor NALM-6 bound
CD28-Ig or CTLA4Ig. CD40 activation of ALL 12, known to induce B7-1 and
B7-2 expression, resulted in binding to BB1, all seven B7-1 mAbs,
anti-B7-2, and both CD28-Ig and CTLA4Ig. These findings demonstrate
that the reactivity of BB1 mAb was distinct from the staining profile
observed with other anti-B7-1 mAbs. Moreover, they demonstrate that
the additional BB1-binding molecule expressed on ALL 12 and NALM-6 is
not a CD28 or CTLA-4 counterreceptor.
|
Molecular cloning with BB1 mAb demonstrates that the BB1 mAb also recognizes CD74
A cDNA expression library was prepared from
BB1+B7-1- NALM-6 cells and transfected into
COS cells. The BB1 mAb was used to immunoselect a BB1 cDNA. Multiple
copies of a 1.3-kb cDNA were isolated. Sequencing and data base search
of this 1.3-kb cDNA revealed it to encode CD74, the invariant chain of
the MHC class II Ag presentation pathway (GenBank accession number
X00497; 25 . While primarily an intracellular protein, some CD74
is transported to the cell surface (26, 27, 28). COS cells transfected with
this CD74 cDNA stained with the BB1 mAb, but not the original
anti-B7-1 mAb (clone 133) (Fig. 2) or
other anti-B7-1 mAbs (data not shown). The reactivity of BB1 mAb
with CD74 was found with all samples of BB1 tested, including five
different samples of BB1 produced in the laboratory of E.A.C., the BB1
from the Fifth Workshop, and commercially produced BB1 (Ancell) (data
not shown). COS cells transfected with B7-1 cDNA bound both BB1 and the
133 mAbs, confirming the dual specificity of the BB1 mAb (Fig. 2). Cell
surface expression of CD74 was demonstrated on CD74-transfected COS
cells using a CD74 mAb. The CD74 mAb did not stain B7-1-transfected COS
cells.
|
), with a charge shift from
aspartic acid to lysine at amino acid 80 probably accounting for the
antigenicity of the human epitope in mice. In contrast, CD74 is a type
II membrane protein. The carboxyl-terminal 160 amino acids (amino acids
73–232) comprise the extracellular domain and do not contain the
characteristic sequence motifs of an Ig family member but instead
contain a high proportion of helical structure. Nevertheless,
sequence comparison of B7-1 and CD74 identifies a region of weak
homology that includes the region comprising the BB1 epitope. This
alignment of CD74 conservatively substitutes a glutamic acid for
aspartic acid at position 80 of human B7-1, is identical to B7-1 at
tryptophan 84, and conservatively substitutes phenylalanine for
tyrosine at position 87. Provocatively, tryptophan 84 and
tyrosine 87 have been shown to be the most critical amino acids for
binding of the B7-1 Ig-V domain to CD28 (30, 31). Whether this is the
region of CD74 that folds into a three-dimensional structure comprising
a BB1 epitope is uncertain.
|
Since BB1+B7-1- keratinocytes have been
reported to bind to CD28-transfected CHO cells, we examined the binding
of CD28-Ig and CTLA4Ig to CD74-transfected COS cells. Neither CD28-Ig
or CTLA4Ig bound to CD74-transfected COS cells though both bound to
B7-1-transfected COS cells (Fig. 4A). The failure of CTLA4Ig to
bind to CD74-transfected COS cells also eliminates the possibility that
CD74 transfection induces expression of the endogenous B7-1 of African
green monkey COS cells, which would bind both BB1 and CTLA4Ig. In
addition, CD74-transfected COS cells did not bind other B7-1 or B7-2
mAbs, which have been shown to bind monkey B7-1 or B7-2 (32) (data not
shown).
|
To examine whether the BB1 mAb preparation might contain separate
Abs to B7-1 and CD74, we examined whether absorption of BB1 on CHO/B7-1
cells removed its CD74 reactivity. BB1 absorbed on CHO/B7-1 cells did
not stain either CD74 or B7-1-transfected COS cells, demonstrating that
the same BB1 mAb recognizes both structures (Fig. 4
B).
Absorption of BB1 on CHO/Neo cells did not affect reactivity of BB1
with either B7-1 or CD74.
Anti-CD74 mAb binding to CD74 is blocked by BB1 mAb
To determine whether the BB1 binding seen on NALM-6 cells is to
CD74, we examined the capacity of the BB1 mAb to block anti-CD74
mAb binding. Figure 5A shows
that the BB1 and anti-CD74 mAbs, as well as control anti-CD10
and anti-CD19 mAbs, all bind to NALM-6 cells. Preincubation of
NALM-6 cells with BB1 mAb blocks binding of anti-CD74 mAb but not
anti-CD10 or anti-CD19 mAbs (Fig. 5B). In contrast,
preincubation with control IgM did not affect the binding of
anti-CD74, anti-CD10, or anti-CD19 mAbs (Fig. 5C). The inhibitory effect of BB1 on anti-CD74 binding
was specific since preincubation with other IgM mAbs, 3B8
(anti-CD56), Mo1 (anti-CD11b), or Mo2 (anti-CD14) did not
block binding of anti-CD74 mAb (Fig. 5D).
|
induces CD74 mRNA expression in keratinocytes
Keratinocytes stimulated with IFN- have been shown to bind BB1
mAb but not to bind other anti-B7-1 mAbs or express B7-1 mRNA (13, 33). Figure 6 shows that untreated or
PMA-treated keratinocytes do not express CD74 mRNA but that IFN-
induces keratinocytes to express CD74 mRNA weakly by 8 h and
strongly by 24 h. CD74 mRNA is also strongly expressed in NALM-6,
Raji, and BALM4 cells. Indeed, the binding of the BB1 but not the B7-1
mAb to the BALM4 cell line was the original basis for distinguishing
BB1 from B7-1 (3).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
There have been multiple reported molecular and phenotypic discrepancies depending on the use of BB1 mAb or other anti-B7-1 mAbs. Examination of these discrepancies shows that they are resolved by an understanding that BB1 staining is the summation of CD74 plus B7-1 expression. BB1 was first reported to Western blot a 37-kDa protein (34), more consistent with the 33 to 43 kDa range of CD74 isoforms than with the 44- to 70-kDa range of B7-1 (14, 21). The first reported phenotypic discrepancy, that BB1 mAb but not anti-B7-1 mAb (clone 133) binds to BALM-4 cells (3), is clearly resolved by the expression of CD74 on BALM-4 cells.
Turka et al. reported that BB1 stained thymic interlobular fibrous septa and some thymic epithelial cells (13, 35) whereas two reports using other anti-B7-1 mAbs both found B7-1 expression on scattered, mostly medullary cells with a dendritic or macrophage morphology but not on thymic epithelial cells or thymic interlobular fibrous septa (36, 37). The expression of the BB1 epitope on MHC class II+ thymic epithelial cells presumably reflects expression of CD74. The expression of MHC class II, CD74, and HLA-DM genes are coordinately controlled by the class II transcriptional trans-activator protein (CIITA) via shared sequence motifs in their promoter regions (38, 39). CIITA regulates both constitutive and inducible expression (40). Since CD74 is coexpressed with MHC class II, staining with BB1 mAb has given the impression that B7-1 is more coordinately expressed with MHC class II than is actually the case.
We reported that anti-Ig-activated human B cells expressed a CTLA-4 counterreceptor by 24 h, whereas the binding of BB1 and B7-1 mAbs was observed only after 48 h (5). Further work identified this early CTLA-4 counterreceptor as B7-2 (CD86) (6). B7-2-transfected cells were not stained by BB1, and their costimulatory activity was blocked by CTLA4Ig but not BB1 mAb (6). When 72-h-activated B cells were sorted into B7-1+ and B7-1- populations, the B7-1--activated B cells bound both BB1 mAb and CTLA4Ig. These B7-1--activated B cells modestly costimulated T cell proliferation, which was blocked by BB1 mAb, CD28 Fab, and CTLA4Ig but not by B7-1 mAb. Therefore, we hypothesized the existence of a third CD28/CTLA-4 counterreceptor, termed B7-3, that was capable of inducing modest T cell proliferation. Based on the the development of mAbs to B7-2 as well as the results of the present study, we now believe that the phenotypic and costimulatory characteristics of the B7-1--activated B cells were due to the combined expression on B7-1--activated B cells of B7-2, a CD28/CTLA-4 counterreceptor, and CD74, a BB1-binding molecule. Therefore, the hypothesized B7-3 is a BB1-binding molecule but is not a third CD28/CTLA-4 counterreceptor but rather a cross-reactive epitope on CD74.
Kulova et al., using the BB1 mAb, reported that, after MHC class II ligation of dense tonsillar B cells, B7-1 expression increased rapidly, peaked within 6 to 9 h, and then waned (41). In contrast, Boussiotis et al., using the B7-1 mAb, found that cross-linking MHC class II on resting splenic B cells induced B7-1 expression only after 48 h (5). These differences may be due to differences in the cell sources used (tonsillar B cells tend to be more activated than splenic B cells) or could be due to Koulova et al. detecting a rapid increase in CD74 expression using the BB1 mAb. It will be of interest to determine which B cell activation signals regulate CD74 expression and whether increases in cell surface CD74 reflect changes in Ag processing or presentation pathways. What is clear is that ligation of MHC class II does increase B7-1 expression (5, 42), implying that MHC class II and B7-1, as might be expected, work closely together to activate T cells via TCR and CD28.
Multiple investigators have reported that BB1 is not expressed on
unstimulated keratinocytes but is induced by IFN- (reviewed in 43 . However, no B7-1 epitopes are expressed nor is B7-1 mRNA
detectable (13, 33). IFN- induces MHC class II on keratinocytes
(44), and MHC class II and CD74 have been shown to be expressed by
keratinocytes during the tuberculin reaction and in diseases such as
lichen planus (45, 46). We show here that IFN- induces CD74 mRNA in
keratinocytes. Strikingly, immunohistochemical analysis of
keratinocytes showed that most of the BB1 staining was intracellular
(13), as would be expected for CD74 present in intracellular
Ag-processing compartments. Indeed, we have found that BB1 mAb exhibits
strong intracellular staining of BB1+B7-1-
pre-B ALL cells (data not shown). This leads to the conclusion that the
BB1 epitope expressed on keratinocytes and pre-B ALL cells is on CD74.
When BB1 mAb is used for immunohistochemistry, which can detect both
intracellular and cell surface Ag, the contribution of CD74 to the
total staining would be greater than when BB1 is used for cell surface
analysis by flow cytometry. Nickoloff et al. also found that
BB1+ keratinocytes adhere to CD28-transfected CHO cells,
and this was blocked by BB1 mAb, suggesting that the BB1 Ag on
keratinocytes is a CD28 counterreceptor. Nevertheless, BB1 mAb did not
block the capacity of keratinocytes to costimulate T cells (13). We do
not find that CD74-transfected cells bind CD28-Ig or CTLA4Ig though we
cannot exclude a very low affinity interaction. CD74 has been reported
to act as a costimulatory molecule via binding of CD44 to chondroitin
sulfate linked to CD74 (47). The binding of BB1 mAb to CD74 was
unaffected by the presence of 100 µg/ml chondroitin sulfate or
hyaluronic acid, excluding the possibility that these sugars are
involved in BB1 mAb recognition (data not shown). Some of the weak
costimulatory activity of B7-1--activated B cells and
IFN--treated keratinocytes might be due to chondroitin
sulfate-modified cell surface CD74 binding to CD44 on T cells (47).
Alternatively, the BB1 mAb might bind to yet an additional Ag.
The presence of a common structural determinant on both B7-1 and CD74 might not be purely coincidental. At the site of interaction between APC and T cell, critical molecules become clustered in an interaction cap. These have been shown to include TCR, CD4, LFA-1, CD28, and CTLA-4 on the T cell side and MHC class II-peptide, ICAM-1, and, presumably, B7 on the APC side (48, 49, 50, 51). The intracellular cytoskeletal protein talin also concentrates beneath the activation cap on the T cell, leading to orientation of the microtubule organizing center and focused delivery of cytokines to the bound APC (48, 49). Perhaps the common structural determinant recognized by the BB1 mAb on both B7-1 and CD74 represents a functionally important region involved in intermolecular interactions between MHC class II and B7-1 that facilitate B7-1 inclusion in the activation cap.
The difficulty in isolating the CD74 cDNA using BB1 mAb prolonged our inability to resolve these discrepancies. CD74 functions to prevent peptide binding to MHC class II, facilitates MHC class II transport, and enhances its localization to Ag-processing compartments (reviewed in 52 . Full-length CD74 cDNA contains two potential translation initiation sites at nucleotides 8–10 and 56–58 whose alternative utilization leads to the p35 and p33 isoforms of CD74, respectively. CD74 is a type II membrane protein, and the major p33 isoform has the NH2-terminal 31 amino acids intracytoplasmic, a 25-amino acid transmembrane region, and 160 amino acids extracytoplasmic. The CD74 cDNA isolated by expression cloning using the BB1 mAb contains nucleotides 18–1304 of the CD74 sequence (1304 bp, GenBank accession number X00497; 25 . An additional CD74 cDNA beginning at nucleotide 13 was also isolated. Thus, neither cDNA contains the first translation initiation site, and both encode exclusively the p33 isoform of CD74. This was probably critical for the successful BB1 cloning since the 16 additional amino acids found in the p35 but not the p33 isoform contain a signal for retention in the endoplasmic reticulum. In the absence of MHC class II, as is the case for transfected COS cells, CD74 trimers containing the p35 isoform transit poorly from the endosomal compartment and thus have reduced cell surface expression (53). This may explain the considerable difficulty experienced in isolating the BB1/CD74 cDNA by expression cloning since a full-length cDNA would encode both p33 and p35 isoforms and have reduced cell surface expression. Visually, COS cells transfected with the CD74 cDNA become rounded, with massively enlarged vacuolar compartments, as has been reported by others (54, 55). This toxicity makes the production of stable high level CD74 transfectants difficult.
CD74 is produced in molar excess over MHC class II, and, by a poorly understood pathway, some is expressed on the cell surface independent of class II although some cell surface CD74 is associated with MHC class II. Mutant B cell lines deleted for all MHC class II genes still express the same amount of cell surface CD74 (56). The half-life of CD74 on the cell surface is only 3 to 4 min, after which it is internalized (56). In contrast, peptide-MHC class II complexes on the cell surface are long-lived. It has been proposed that cell surface CD74 binds to MHC class II molecules that have lost their antigenic peptide, leading to internalization and reloading with antigenic peptide (28, 56). Binding of BB1 mAb to cell surface CD74 may have effects independent of blockade of B7-1 costimulation. These might include inhibition of Ag presentation or stimulation via cross-linking of associated MHC class II. Somewhat perplexingly, CD74 has been reported to be expressed at low levels on the surface of unstimulated B cells; however, these cells do not bind BB1 mAb (3, 4, 26). This may reflect the inability of the low affinity BB1 mAb to detect low level CD74 cell surface expression.
While the above results resolve many of the discrepancies concerning the expression of the B7-1 molecule, its temporal relationship to B7-2 expression, and whether a B7-3 exists, we find no supporting evidence for the existence of additional CD28/CTLA-4 counterreceptors. Mice deficient in B7-1 and B7-2 have a severe immune defect, and activated B cells isolated from them do not bind CTLA4Ig (57). Further characterization of these B7-1- and B7-2-deficient mice should definitively test whether additional CD28/CTLA-4 counterreceptors exist in the mouse.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Gordon J. Freeman, Dana-Farber Cancer Institute, 44 Binney Street, Boston, MA 02115. E-mail address:
3 Abbreviations used in this paper: CHO, Chinese hamster ovary; ALL, acute lymphoblastic leukemia.
Received for publication February 9, 1998. Accepted for publication May 12, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. J. Bell, R. Rigby, N. English, S. D. Mann, S. C. Knight, M. A. Kamm, and A. J. Stagg Migration and Maturation of Human Colonic Dendritic Cells J. Immunol., April 15, 2001; 166(8): 4958 - 4967. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. K. Kapsogeorgou, H. M. Moutsopoulos, and M. N. Manoussakis Functional Expression of a Costimulatory B7.2 (CD86) Protein on Human Salivary Gland Epithelial Cells that Interacts with the CD28 Receptor, but Has Reduced Binding to CTLA4 J. Immunol., March 1, 2001; 166(5): 3107 - 3113. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. J. Freeman, A. J. Long, Y. Iwai, K. Bourque, T. Chernova, H. Nishimura, L. J. Fitz, N. Malenkovich, T. Okazaki, M. C. Byrne, et al. Engagement of the Pd-1 Immunoinhibitory Receptor by a Novel B7 Family Member Leads to Negative Regulation of Lymphocyte Activation J. Exp. Med., October 2, 2000; 192(7): 1027 - 1034. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Van Rhijn, L. H. Van den Berg, W. M. J. Bosboom, H. G. Otten, and T. Logtenberg Expression of accessory molecules for T-cell activation in peripheral nerve of patients with CIDP and vasculitic neuropathy Brain, October 1, 2000; 123(10): 2020 - 2029. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Abrams, S. L. Kelley, E. Hayes, T. Kikuchi, M. J. Brown, S. Kang, M. G. Lebwohl, C. A. Guzzo, B. V. Jegasothy, P. S. Linsley, et al. Blockade of T Lymphocyte Costimulation with Cytotoxic T Lymphocyte-Associated Antigen 4-Immunoglobulin (Ctla4ig) Reverses the Cellular Pathology of Psoriatic Plaques, Including the Activation of Keratinocytes, Dendritic Cells, and Endothelial Cells J. Exp. Med., September 5, 2000; 192(5): 681 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-y. Murata and M. C. Dalakas Expression of the co-stimulatory molecule BB-1, the ligands CTLA-4 and CD28 and their mRNAs in chronic inflammatory demyelinating polyneuropathy Brain, August 1, 2000; 123(8): 1660 - 1666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-y. Murata and M. C. Dalakas Expression of the Costimulatory Molecule BB-1, the Ligands CTLA-4 and CD28, and their mRNA in Inflammatory Myopathies Am. J. Pathol., August 1, 1999; 155(2): 453 - 460. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Behrens, M. Kerschensteiner, T. Misgeld, N. Goebels, H. Wekerle, and R. Hohlfeld Human Muscle Cells Express a Functional Costimulatory Molecule Distinct from B7.1 (CD80) and B7.2 (CD86) In Vitro and in Inflammatory Lesions J. Immunol., December 1, 1998; 161(11): 5943 - 5951. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. T. Doty and E. A. Clark Two Regions in the CD80 Cytoplasmic Tail Regulate CD80 Redistribution and T Cell Costimulation J. Immunol., September 15, 1998; 161(6): 2700 - 2707. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
