| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
The G. W. Hooper Foundation, Department of Microbiology and Immunology and
Departments of Biopharmaceutical Sciences and Pharmaceutical Chemistry, University of California, San Francisco, CA 94143;
Laboratory of Intracellular Parasites, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, Hamilton, MT 59840; and
§
Sir William Dunn School of Pathology, Oxford, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CHS is a rare, autosomally inherited immunodeficiency in which numerous cell types from affected patients have fewer and considerably larger lysosomes than usual. Lysosomal degradation occurs normally in CHS fibroblasts (6), but in hematopoietic cells and melanocytes, secretion of lysosomes and lysosome-like organelles is impaired (7, 8). Thus, these patients are hypopigmented and have nonfunctional cytotoxic T cells and NK cells incapable of secreting lytic granules, as well as defects in enzyme secretion by macrophages and neutrophils (9). The mutations responsible for the CHS defect and the murine version in the beige mouse have all been mapped to the same gene (10, 11, 12). The gene product has predicted sequence homology to the yeast protein VPS15, which is involved in protein trafficking to the yeast vacuole, a process thought to be analogous to lysosomal targeting. The beige gene produces a 400-kDa cytosolic protein that promotes lysosome fission when overexpressed (13). Mutations at various locations in the gene in both mouse and human result in swollen lysosomes characteristic of the disease. Enlarged lysosomes are also characteristic of cells infected with the rickettsia C. burnetii (14). Coxiella entry occurs by host cell phagocytosis, after which bacteria-containing vacuoles fuse and acquire a characteristic lysosomal phenotype with low internal pH and typical lysosomal markers, becoming engorged with replicating organisms (14). Infection by this obligate intracellular bacterium often presents as a prolonged atypical pneumonia called Q fever. Despite an Ab response, infections may become chronic, resulting in relapse of symptoms years later (15). Thus, both the genetic and infectious lysosomal diseases we examined are associated with an impaired immune response, as well as altered lysosome morphology.
Here we demonstrate that the HLA-DR human class II molecules and the related HLA-DM molecules are localized to enlarged lysosome-like vacuoles in B lymphoblastoid cells from CHS patients and in Coxiella-infected cells. In both disease states, we can detect enhanced interaction between HLA-DR and HLA-DM by coimmunoprecipitation. Peptide loading of class II molecules is dependent on HLA-DM, which is normally localized to the loading compartment and which functions by catalytic removal of the endogenous CLIP peptide from the class II peptide binding site to allow peptide derived from exogenous protein to bind (16, 17, 18). CLIP is a proteolytic remnant of the invariant chain, which is responsible for targeting class II molecules to the endocytic pathway (19). In the CHS B cell lines, we observe that the peptide loading process is skewed toward more efficient removal of CLIP from HLA-DR molecules. While our manuscript was under revision, Faigle et al. (20) reported that CHS B cells have an altered peptide repertoire and exhibit delayed Ag presentation to T cells. These properties, as well as reduced CLIP association, can be explained by the mechanism of enhanced HLA-DM/HLA-DR association that we describe here. Thus, we find that the morphology and function of the class II peptide loading compartment is affected by processes that affect lysosome morphology, indicating common mechanisms in their biogenesis and suggesting a role for compartment morphology in Ag presentation. Furthermore, the immune defects in CHS patients and the poor immune response to Coxiella infection may be exacerbated and/or partially explained by this morphologically induced perturbation of the class II MHC pathway.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
EBV-transformed B cell lymphoblastoid lines, JY, Jest Hom, AVL, EA, 9.5.3 (21), and CHS-GG (7), were cultured in RPMI 1640 supplemented with 2 mM L-glutamine and 10% FBS (Gemini Bio-Products, Calabasas, CA) in 5% CO2 at 37°C. Pala cells were grown in Iscove’s medium with 5% FBS, and GM03365 and GM02431A (Human Genetic Mutant Cell Repository, Camden, NJ) were grown in RPMI 1640 with 20% FBS. The HLA-DR types of the cells are as follows: AVL, DRb1*03011; EA, DRb1*1501; CHS-GG, DRb1*1501/0701; GM02431A, DRb1*0301; GM03365, DRb1*1501/1001; Jest Hom, DRb1*0101; and JY, DRb1* 1301/0402. Cell cycle synchronization was achieved by starving the cells in medium with 0.1% FBS for 32 h followed by a 20-h recovery in normal medium before analysis. HeLa 229 cells (American Type Culture Collection, Manassas, VA) were cultured in Eagle MEM with Earle’s balanced salt solution (Mediatech, Washington, DC) with 10% FBS in 5% CO2 at 37°C.
Antibodies
The specificities of anti-HLA-DR-
ß mAb L243 (22, 23),
anti-HLA-DR -chain mAb DA6.147 (24), anti-class I MHC mAb
W6/32 (25), anti-CLIP-HLA-DR complex CerCLIP.1 mAb (26),
anti-LAMP-1 mAb H4A3 (27), and anti-HLA-DM rabbit serum (28)
have been previously described. The antiserum DMBS1, against
DMß, used for immunoblotting was produced by rabbit
immunization with a peptide (CYTPLPGSNYSEGWHIS) from the cytoplasmic
domain of the DM ß-chain.
Bacterial infections
Infections were done as previously described (14). HeLa 229
cells were washed with HBSS and overlaid with C. burnetii,
Nine Mile strain, in phase II or Chlamydia trachomatis
serovar L2 diluted into SPG (0.22 M sucrose, 20 mM sodium phosphate,
and 5 mM L-glutamic acid) and incubated at 37°C for
1 h. Cells were then washed with HBSS and supplemented with medium
and 0.5 µg/ml L-tryptophan (29). HeLa cells were
incubated in the presence of 100 U/ml human IFN-
either 24 h
before infection and during infection or simultaneous with infection,
as specified. Infections were allowed to progress for 24–48 h before
cell lysis in Nonidet P-40 lysis buffer. Experiments were performed
with HeLa cultures, with infection efficiencies >65%.
Immunofluorescent microscopy
The indirect immunofluorescence procedure for the EBV-B cell lines was modified from a published procedure (26). B cells (200 µl) suspended at 5 x 105 cells/ml medium were plated onto 12-mm poly-L-lysine-coated coverslips and incubated for 1 h at 37°C. Coverslips were then washed in PBS; fixed for 10 min in 4% paraformaldehyde/PBS; permeabilized for 10 min with 0.04% saponin; and blocked in a solution of 5% goat serum, 0.02% SDS, and 0.1% Nonidet P-40 in PBS with 0.02% sodium azide. Coverslips were incubated with primary Ab diluted in blocking solution for 1 h at room temperature, washed with PBS-sodium azide, and incubated for 1 h with the appropriate Cy3- or FITC-conjugated secondary Abs (Jackson Immunoresearch Laboratories, West Grove, PA), then mounted in 0.1% p-phenylenediamine (Sigma Chemical, St. Louis, MO) in Fluromount G (Fisher Scientific, Pittsburgh, PA), and viewed with a Zeiss Axiophot fluorescence microscope. Indirect immunofluorescence for the adherent HeLa cells was conducted in the same manner, except that the cells were grown on uncoated coverslips; infected with bacteria (see infection protocol); and then fixed, permeabilized, and stained as described above. Paired immunofluorescence/Nomarski imaging was done on a Nikon fluorescence microscope.
Flow cytometry
B cells (5 x 105) were washed in ice-cold PBS-sodium azide containing 2% BSA, resuspended in 200 µl of primary Ab diluted into PBS/BSA, and incubated on ice for 45 min. Cells were washed in PBS/BSA and resuspended in FITC-conjugated goat anti-mouse Ig (Dako, Carpinteria, CA) diluted in PBS/BSA for 30 min at 4°C. After washing, cell samples were analyzed on a Becton Dickinson FACScan (Mountain View, CA). The CLIP-HLA-DR to total HLA-DR ratios were calculated by taking the ratios of the geometric means of the CerCLIP.1 peak and the L243 peak. Geometric means were calculated by CellQuest software (Becton Dickinson) and are a measure of both mean values of all events and the range of distribution of the events. Staining with secondary Ab only was equivalent in all cells. Concentrations of the L243 mAb and CerCLIP.1 mAb used were established to be saturating by independent titration experiments.
Immunoprecipitations
To isolate CLIP-HLA-DR complexes, cells were lysed in Nonidet
P-40 (NP-40) lysis buffer (1% NP-40 50 mM Tris, pH 7.2, 150 mM sodium
chloride, 0.2 mM PMSF, 2.5 µg/ml aprotinin, and 1 µg/ml leupeptin)
for 1 h on ice. After pelleting the nuclei, lysates were incubated
with protein G-bound CerCLIP.1. Immunoprecipitates were washed, boiled
in SDS-sample buffer, and analyzed by SDS-PAGE (12%). The gels were
transferred to nitrocellulose and immunoblotted with DA6.147. Total
HLA-DR levels were obtained by quantifying DR -chain in whole
lysates in the same manner. Bands were detected and quantified using an
alkaline phosphatase-conjugated goat anti-mouse Ig and enhanced
chemifluorescence (ECF) substrate (Amersham, Arlington Heights, IL) on
the Storm system (Molecular Dynamics, Sunnyvale, CA). To detect levels
of HLA-DR/HLA-DM complexes, cells were lysed in 1%
3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid (CHAPS)
or 1% digitonin in 150 mM NaCl and 50 mM sodium acetate, pH 5.0. mAbs
L243 and DA6.147, respectively, were used to immunoprecipitate HLA-DR,
as above. Samples were analyzed by SDS-PAGE (12%), transferred to
nitrocellulose, and immunoblotted with DMBS1 (anti-DM ß-chain),
and signals were detected by ECF on film, as above. Blots were then
stripped of Ab by incubation in 100 mM 2-ME, 2% SDS, and 62.5 mM Tris,
pH 6.7 at 55°C for 20 min and then reprobed with DA6.147 to quantify
the level of DR -chain immunoprecipitated. Blotting signals were
scanned and then quantitated using the National Institutes of Health
Image 1.61 program.
Cell surface biotinylation
The biotinylation procedure was modified from a published report (30). Infected cells were washed three times with ice-cold PBS and then biotinylated for 10 min on ice with 0.5 mg/ml sulfo-NHS-LC-biotin (Pierce, Rockford, IL) in cold PBS. After washing the cells, the reaction was quenched with 50 mM NH4Cl in PBS for 10 min. Nonidet P-40 lysates were made as described earlier. Immunoprecipitations were conducted using L243 mAb and analyzed by SDS-PAGE. Equivalent lanes were blotted with either DA6.147 or streptavidin conjugated to horseradish peroxidase (Zymed) followed by detection with the enhanced chemiluminescence substrate (Amersham).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In human B cell lines, HLA-DM is a characteristic marker of the
class II loading compartment (31). To establish whether morphology of
the loading compartment is altered in cells of CHS patients, we
examined the distribution of HLA-DM relative to the enlarged CHS
lysosomes. B lymphoblastoid cell lines from three unrelated CHS
patients were double-stained for HLA-DM and the lysosomal marker LAMP-1
and analyzed by immunofluorescent microscopy (Fig. 1
). The three CHS cell lines all
contained significantly enlarged LAMP-1-staining vacuoles (Fig. 1, F, H, and J), the majority of which
costained for HLA-DM (Fig. 1, E, G, and
I). The wild-type B cells, JY and Jest Hom, contained
numerous small cytoplasmic LAMP-1-positive vesicles, indicated by a
punctate staining pattern (Fig. 1, B and D), a
subset of which also stained for HLA-DM (Fig. 1, A and
C). In CHS cells, HLA-DM-staining vacuoles were consistently
larger and fewer in number than those seen in wild-type cells.
Double-staining for HLA-DR and HLA-DM revealed that HLA-DR accumulated
in the enlarged HLA-DM-containing vesicles, although staining of
cell-surface HLA-DR was also observed in CHS cells (Fig. 2). The vesicular HLA-DR staining in
wild-type cells was much more diffuse (Fig. 2, B and
D). A similar immunofluorescent staining pattern of
accumulated HLA-DM, HLA-DR, and LAMP-1 was recently reported for two
additional B cell lines derived from CHS patients (20). We observed no
significant difference in the distribution of class I MHC molecules
between the CHS and wild-type B cells (data not shown), which
demonstrates that the CHS mutation specifically affects molecules
trafficking to the endocytic pathway, such as HLA-DM and HLA-DR,
without disrupting the secretory pathway used by class I molecules.
These data are compatible with earlier observations (6) that attribute
the CHS pathology to disruption of late endocytic compartments.
|
|
The swollen lysosomal compartments of Coxiella-infected
cells are similar to the enlarged lysosomes seen in CHS cells. To
determine whether alterations in lysosome morphology by mechanisms
other than the CHS mutation would also affect HLA-DM and HLA-DR
distribution, we analyzed their distribution in
Coxiella-infected HeLa cells. HeLa cells do not normally
express HLA-DR, invariant chain, or HLA-DM, so expression of these
molecules was induced by 48-h treatment with IFN-. Induction was
done simultaneously with Coxiella infection, and although
IFN- treatment has been reported to slow the growth of
Coxiella (32), it did not alter the vacuole morphology or
growth rate of the bacteria in our experience (data not shown).
Uninfected and Coxiella-infected IFN--treated HeLa 229
cells were then stained for HLA-DM and HLA-DR (Fig. 3). Immunofluorescent microscopy showed
that the enlarged vacuoles of Coxiella-infected HeLa cells
accumulated both HLA-DR (Fig. 3B) and HLA-DM (Fig. 3D). These vacuoles were identified as Coxiella
vacuoles by paired Nomarski imaging of the same field (Fig. 3, A and C). In contrast, uninfected cells displayed
a punctate cytoplasmic staining for both HLA-DR and HLA-DM indicative
of numerous, small vesicles (data not shown). To control for possible
nonspecific effects of infection by intracellular bacteria on class II
distribution, the same experiment was done with Chlamydia
trachomatis, an organism that resides in a nonlysosomal vacuole
(14). The chlamydial vacuoles (Fig. 3, E and G)
did not label with Abs against either HLA-DR (Fig. 3F) or
HLA-DM (Fig. 3H). Thus, distortion of lysosome morphology by
two independent mechanisms, the CHS mutation and Coxiella
infection, correlated with distortion of the class II loading
compartment. These observations indicate that the biogenesis of the
class II loading compartment shares mechanisms with the biogenesis of
lysosomes, suggesting that the class II compartment is derived from the
normal endocytic pathway rather than generated as a specialized
compartment.
|
Cytotoxic T cells from CHS patients exhibit defective exocytosis
of lysosomally derived lytic granules (7). Morphological studies have
led to the suggestion that the exocytosis of the class II loading
compartment, in a manner similar to T cell degranulation, might be a
mechanism of delivering peptide-loaded class II molecules to the cell
surface (33, 34). If this is the case, the CHS mutation should affect
the surface expression of class II molecules. The concentrated staining
of HLA-DR in Coxiella vacuoles also raised the possibility
that the export of HLA-DR molecules might be hindered. Surface
expression of HLA-DR was analyzed in both disease states to investigate
these issues (Fig. 4). Levels of HLA-DR
on the surface of CHS B cells were compared with HLA-DR levels on a
number of wild-type B cells by flow cytometry of cells labeled with Ab
under saturating conditions. Since lysosome size has been observed to
vary within the cell cycle of CHS cells (35), cell division of both CHS
and wild-type B cells was synchronized by serum starvation followed by
recovery in normal growth medium before analysis. HLA-DR staining with
mAb L243 indicated that CHS B cells express wild-type levels of class
II molecules on their surface (Fig. 4A) about twice as much
as the HLA-DR-hemizygous cell line 9.5.3 (21). According to this
steady-state measurement, the CHS mutation does not inhibit the
expression of class II molecules on the cell surface. The steady-state
surface levels of HLA-DR were measured biochemically in
Coxiella-infected cells, since infected cells were not
amenable to FACS analysis. HeLa cells were exposed to IFN- and
simultaneously either infected with Coxiella or
mock-infected. After 48 h in culture, surface molecules were
labeled by biotinylation and HLA-DR molecules were isolated from
solubilized cells by immunoprecipitation with the mAb DA6.147,
recognizing the -chain of HLA-DR. These immunoprecipitates were
analyzed for the proportion of HLA-DR derived from the cell surface by
blotting with streptavidin and for the total level of HLA-DR by
immunoblotting with mAb DA6.147. No substantial difference in the
amount of surface HLA-DR was detected when Coxiella-infected
cells were compared with uninfected cells, with total levels of HLA-DR
being equivalent in both samples (Fig. 4B). These results
demonstrate that although HLA-DR appears to accumulate in the enlarged
lysosomes present in both disease states, neither condition reduces the
steady-state surface levels of these class II molecules.
|
While the CHS mutation or Coxiella infection did not
decrease steady-state surface levels of HLA-DR, it is possible that the
accumulation of HLA-DR with HLA-DM in enlarged loading compartments
might affect their interaction and consequently have an effect on Ag
presentation. To assess whether HLA-DM/HLA-DR interactions were
altered, HLA-DR was immunoprecipitated from the GM02431A CHS B cells,
and the level of associated HLA-DM was measured, in comparison with the
level of HLA-DM/HLA-DR complexes in the haplotype-matched wild-type
cell line Pala (Fig. 5). When either of
two anti-HLA-DR mAbs was used to isolate these complexes, it was
found that the HLA-DR from GM02431A had more HLA-DM associated with it
than the HLA-DR from Pala. This is demonstrated quantitatively by the
ratios of the HLA-DM/HLA-DR blotting signals in the immunoprecipitates
from each cell line (Fig. 5B). The total levels of HLA-DM
and HLA-DR were comparable in the cell lysates analyzed (data not
shown), so the altered ratios in the immunoprecipitates represent
skewed association of HLA-DM with HLA-DR in the CHS cells. A similar
enhancement of HLA-DM/HLA-DR association was observed in cells infected
with Coxiella. When HLA-DR was immunoprecipitated from
infected cells, more HLA-DM was bound than the amount that
coimmunoprecipitated with an equivalent amount of HLA-DR from
uninfected cells (Fig. 6).
|
|
Before binding antigenic peptide, class II molecules must first
have the residual invariant chain peptide, CLIP, removed from the
peptide binding site. This process is catalyzed when CLIP-HLA-DR
complexes interact with HLA-DM, which stabilizes the unloaded form of
HLA-DR (16, 17, 28). To evaluate whether the enhanced interaction of
HLA-DM with HLA-DR in CHS cells has a functional consequence, their
levels of CLIP-HLA-DR complexes were analyzed in comparison with
wild-type cells and with the HLA-DM-deficient cell line, 9.5.3, on
which the majority of surface HLA-DR molecules are loaded with CLIP
(36, 37, 38). The CerCLIP.1 mAb, an Ab recognizing CLIP bound to HLA-DR
molecules, was used to measure surface levels of CLIP-HLA-DR complexes
(Fig. 7). Wild-type, homozygous B cells
with HLA-DR allelic types partially or fully matching the CHS types
were used as controls due to the unavailability of fully matched
heterozygous cells. Allele matching was attempted because products of
different HLA-DR alleles have been shown to have varying affinities for
CLIP (39). Flow cytometry (Fig. 7A) of synchronized cells
stained with CerCLIP.1 indicated that the levels of CLIP-HLA-DR
complexes on the surface of the three CHS B cell lines were lower than
on the wild-type Jest Hom cell line, which has a comparable level of
HLA-DR on its surface (Fig. 4
A). In comparing the levels of
CLIP on CHS cells with other control cell lines, it was necessary to
account for varying levels of HLA-DR on the surface of these cells.
Therefore, the proportion of HLA-DR loaded with CLIP to total HLA-DR
was calculated by taking the ratio of the geometric mean of the
CerCLIP.1 staining intensity to the geometric mean of the
anti-HLA-DR L243 staining intensity. As determined from the average
ratios from multiple experiments (n = 3), levels of
CLIP-HLA-DR on the CHS cell lines were among the lowest in the panel of
cells analyzed (Fig. 7B), with the CHS line GM03365 always
having the lowest ratio. Surface levels of CLIP on wild-type B cells
were quite variable and on one cell line (Jest Hom) approached the
level on the HLA-DM-deficient 9.5.3 cell line, while on another (EA)
they approached CHS cell levels. The finding of high CLIP levels on
wild-type cells was unexpected, since very little CLIP-HLA-DR is
present on the surface of HLA-DM-deficient B cells that have been
genetically rescued by transfection with HLA-DM (26, 40). However, it
may be that when HLA-DM is present at wild-type levels and not
overexpressed due to transfection, its unloading of CLIP is not
particularly efficient, a process that might also vary for different
HLA-DR types. The reduction of cell surface CLIP on CHS cells suggests
that an altered peptide repertoire is presented by these cells.
|
). The GM03365 or GM02431A cell
lysates had hardly any detectable CLIP-associated HLA-DR compared with
the wild-type JY cell line or the HLA-DM-deficient 9.5.3 cell line.
Thus, by two measures, the CHS B lymphoblastoid cells have fewer
CLIP-HLA-DR complexes than their wild-type counterparts, a difference
seen most strikingly in the GM03365 line. The decrease in CLIP-HLA-DR
on the cell surface of CHS cells, without a concomitant decrease in
overall HLA-DR surface levels, suggests that the HLA-DR molecules
observed in swollen lysosome-like compartments are transported through
these compartments and out to the cell surface. Thus, the swollen
HLA-DR/HLA-DM-rich compartments appear to function as class II loading
compartments and are not merely dead-end lysosomes. In addition, these
results indicate that the necessary conditions for HLA-DM to function
are present in the enlarged lysosomes, including low pH, as would be
predicted from previous studies demonstrating normal degradative
function in lysosomes of CHS fibroblasts (6).
|
50% of infections (data not shown). Thus,
diseases that expand lysosomes cause the accumulation of HLA-DR and
HLA-DM in enlarged compartments and thereby enhance their association,
which can affect the peptide repertoire displayed by the diseased
cells, without a cumulative effect on their surface HLA-DR stability. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The fact that diseases that induce enlargement of lysosomes can also affect the morphology of the class II loading compartment indicates that these two compartments share mechanisms of biogenesis. This suggests that the class II loading compartment is part of the endocytic pathway, rather than an entirely specialized compartment. In two different cell lines, which are not specialized for Ag presentation, it has been previously demonstrated that the expression of class II molecules along with invariant chain and HLA-DM is sufficient to generate compartments that resemble class II loading compartments (3, 4). These experiments were interpreted to mean that class II molecules and their associated proteins contained signals that allowed the de novo generation of a specialized compartment. An alternative interpretation could be that class II, invariant chain, and HLA-DM traffic to pre-existing endocytic compartments and simply accumulate there by virtue of retention or targeting signals in their cytoplasmic tails. This explanation would account for the observed distribution of class II molecules throughout a number of endosomal compartments (41, 42). Recent electron microscopy data, identifying a progression of endocytic, class II-containing structures, also suggests an intimate relationship between pre-existing endocytic compartments and those used for Ag processing and presentation (5), a relationship corroborated by the findings reported here.
Although HLA-DR molecules accumulate in the CHS and Coxiella-induced vacuoles, a decrease in their steady-state surface levels was not detected. We also found that the total levels of HLA-DR molecules in CHS cells were equivalent to those in normal B cells (data not shown). Thus, the intense staining of the vacuoles in diseased cells does not reflect an increase in intracellular HLA-DR and HLA-DM, merely a redistribution, so they are are concentrated together in larger, but fewer, structures than in normal cells. This correlates with the report that the gross cellular volumes of lysosomes in normal and CHS fibroblasts are similar (6). The fact that HLA-DR molecules can reach the cell surface in CHS cells also indicates that secretion of class II molecules is not exclusively via the same mechanism as regulated secretion of lytic granules, a process that is defective in CHS cytotoxic lymphocytes (7). It is likely that the primary mechanism for class II expression is via transport in small vesicles that bud from the loading compartment (43), and the expression of class II on CHS cells is consistent with this mechanism. In CHS cells, enlarged lysosomes are apparently the consequence of a defective fission mechanism (13). Therefore, fission of class II-containing compartments themselves does not seem to be required for class II export to the cell surface. This fission mechanism is presumably distinct from budding of transport vesicles.
The colocalization of HLA-DM and HLA-DR in abnormally large structures has several measurable effects. Here we show that in both disease states, elevated levels of HLA-DM/HLA-DR complexes can be detected relative to levels in cells with a normal endocytic pathway. Faigle et al. (20) have recently shown that the transit of HLA-DR through the endocytic pathway of CHS cells is slowed, suggesting that altered compartment morphology affects intracellular trafficking. Slower transit could be responsible for the enhanced association that we observe between the HLA-DM residing in enlarged vacuoles and the HLA-DR molecules trafficking through the vacuoles. Faigle et al. also described a slower acquisition of SDS stability by HLA-DR molecules in CHS cells, with an accompanying delay in HLA-DR-restricted presentation of processed Ag to T cells (20). These phenotypes can be explained by the enhanced HLA-DM/HLA-DR interaction that we report. Together these data support previous predictions that in vivo, as well as in vitro, HLA-DM does not actively load HLA-DR molecules, but rather stabilizes the unloaded conformation (16, 17, 28). Enhancement of this predicted HLA-DM activity is consistent with the reduction in associated CLIP, shown here, and slower acquisition of stability and antigenic peptides by HLA-DR molecules in CHS cells (20).
Enhanced interaction of HLA-DM with HLA-DR also alters the peptide
repertoire that is ultimately expressed on the CHS cell surface, as
indicated by the reduced proportion of CLIP-loaded class II on the
surface of these cells. It was not possible to test this effect for
Coxiella-infected HeLa cells induced with IFN-, because
CLIP-HLA-DR complexes are undetectable in IFN--treated uninfected
HeLa cells. This is likely due to the absence of induction of HLA-DO
expression by IFN- and the consequent heightened activity of HLA-DM
(44, 45, 46). However, independent evidence of an altered peptide
repertoire in CHS cells is indicated by the fact that their peptides
are, on average, shorter than the peptides bound by class II on
wild-type cells (20). This property, as well as the reduced CLIP, could
reflect enhancement of peptide-editing function of HLA-DM, as a result
of increased interaction with class II molecules (47). Here we report
that, as also described by Faigle et al. (20), the surface HLA-DR
molecules on CHS cells are stably loaded with peptide at steady state.
This is not surprising, because only peptide-loaded molecules will
accumulate over time, even if HLA-DR stability takes longer to achieve
in these cells. Interestingly, in a subset of Coxiella
infections, it was possible to detect a decrease in the steady-state
stability of HLA-DR molecules (data not shown). The fact that unloaded
complexes were even detectable at steady state may have been due to the
combined effect of enhanced interaction of HLA-DM with HLA-DR and the
hyperactivity of HLA-DM in these IFN--induced HeLa cells, which lack
HLA-DO. Thus, both lysosomal diseases display defects that could result
in altered peptide presentation to T cells.
B cells from the three CHS patients tested showed variability in their
levels of surface CLIP-HLA-DR complexes, although low relative to most
normal cells. This result hints at a range of CHS phenotypes and
suggests that the more dramatic phenotypes may cause serious impairment
in Ag presentation. Of the CHS lines examined, GM03365 showed the
largest reduction in CLIP relative to wild-type B cell lines, and by
microscopy, GM03365 cells generally had the largest vacuoles (Figs. 1 and 2). Genetic analysis of GM03365 has revealed a nonsense mutation in
the CHS gene, resulting in premature termination and loss of two-thirds
of the protein (12). While only a handful of CHS patients have been
genetically characterized, all have been shown to have homozygous
nonsense mutations at various positions along the entire length of the
3801-amino acid gene, resulting in null alleles (48). Interestingly,
early termination of the transcript does not correlate with an earlier
age of onset or increased severity of the clinical disease, suggesting
that other factors may be involved in the progression of CHS pathology.
In conclusion, our data demonstrate that in both a hereditary and an infectious disease characterized by swollen lysosomes, the common phenotype correlates with accumulation of HLA-DM and HLA-DR in enlarged vacuoles. HLA-DR molecules apparently exit this vacuole and appear on the cell surface with decreased levels of CLIP peptide. Thus, these enlarged vacuoles seem to represent morphologically expanded class II loading compartments, indicating that the biogenesis of these compartments utilizes the same fusion and fission mechanisms that control lysosome formation. The interaction between HLA-DM and HLA-DR was measurably enhanced in these cells and appears to be a function of the altered morphology, the only feature common to both disease states. This enhanced HLA-DM/HLA-DR interaction suggests a mechanism for delays in antigenic peptide presentation, characteristic of CHS cells (20). Thus, the morphological effects on the class II transport pathway occurring in these lysosomal diseases contributes to the immune deficiencies seen in CHS patients and may very well do so in Coxiella-infected individuals.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Frances M. Brodsky, The G. W. Hooper Foundation, Box 0552, University of California, San Francisco, San Francisco, CA 94143-0552. E-mail address:
3 Abbreviations used in this paper: CHS, Chediak-Higashi syndrome; ECF, enhanced chemifluorescence; CHAPS, 3-[(3-cholamidoprophyl)dimethylammonio]-1-propanesulfonic acid; NP-40, Nonidet P-40.
Received for publication March 30, 1998. Accepted for publication September 14, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß dimers and facilitates peptide loading. Cell 82:155.[Medline]
and ß chains. Eur. J. Immunol. 12:942.[Medline]
-interferon inhibition of Chlamydia trachomatis growth. Curr. Microbiol. 16:9.
and Igß subunits in MHC class II-restricted antigen presentation. Immunity 3:335.[Medline]
inhibits growth of Coxiella burnetii in mouse fibroblasts. Infect. Immun. 45:781.This article has been cited by other articles:
|
|
 |
|
  J. M. Martin-Fernandez, J. A. Cabanillas, M. Rivero-Carmena, E. Lacasa, J. Pardo, A. Anel, P. R. Ramirez-Duque, F. Merino, C. Rodriguez-Gallego, and J. R. Regueiro Herpesvirus saimiri-transformed CD8+ T cells as a tool to study Chediak-Higashi syndrome cytolytic lymphocytes J. Leukoc. Biol., May 1, 2005; 77(5): 661 - 668. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Huynh, D. Roth, D. M. Ward, J. Kaplan, and N. W. Andrews From the Cover: Defective lysosomal exocytosis and plasma membrane repair in Chediak-Higashi/beige cells PNAS, November 30, 2004; 101(48): 16795 - 16800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Gutzmer, W. Li, S. Sutterwala, M. P. Lemos, J. I. Elizalde, S. L. Urtishak, E. M. Behrens, P. M. Rivers, K. Schlienger, T. M. Laufer, et al. A Tumor-Associated Glycoprotein That Blocks MHC Class II-Dependent Antigen Presentation by Dendritic Cells J. Immunol., July 15, 2004; 173(2): 1023 - 1032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. T. A. Marques Jr., P. Chikhlikar, L. B. de Arruda, I. C. Leao, Y. Lu, J. Wong, J.-S. Chen, B. Byrne, and J. T. August HIV-1 p55Gag Encoded in the Lysosome-associated Membrane Protein-1 as a DNA Plasmid Vaccine Chimera Is Highly Expressed, Traffics to the Major Histocompatibility Class II Compartment, and Elicits Enhanced Immune Responses J. Biol. Chem., September 26, 2003; 278(39): 37926 - 37936. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. van Lith, M. van Ham, A. Griekspoor, E. Tjin, D. Verwoerd, J. Calafat, H. Janssen, E. Reits, L. Pastoors, and J. Neefjes Regulation of MHC Class II Antigen Presentation by Sorting of Recycling HLA-DM/DO and Class II within the Multivesicular Body J. Immunol., July 15, 2001; 167(2): 884 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-W. Wang, J. Howson, E. Haller, and W. G. Kerr Identification of a Novel Lipopolysaccharide-Inducible Gene with Key Features of Both a Kinase Anchor Proteins and chs1/beige Proteins J. Immunol., April 1, 2001; 166(7): 4586 - 4595. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Raposo, D. Tenza, D. M. Murphy, J. F. Berson, and M. S. Marks Distinct Protein Sorting and Localization to Premelanosomes, Melanosomes, and Lysosomes in Pigmented Melanocytic Cells{image} J. Cell Biol., February 19, 2001; 152(4): 809 - 824. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-J. Ullrich, W. L. Beatty, and D. G. Russell Interaction of Mycobacterium avium-Containing Phagosomes with the Antigen Presentation Pathway J. Immunol., December 1, 2000; 165(11): 6073 - 6080. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
65%.
