вход по аккаунту


Disparate folding and stability of the ankylosing spondylitisassociated HLAB.24045.pdf1403 and B2705 proteins

код для вставкиСкачать
Vol. 58, No. 12, December 2008, pp 3693–3704
DOI 10.1002/art.24045
© 2008, American College of Rheumatology
Disparate Folding and Stability
of the Ankylosing Spondylitis–Associated
HLA–B*1403 and B*2705 Proteins
Elena Merino, Begoña Galocha, Miriam N. Vázquez, and José A. López de Castro
Objective. To investigate the folding, assembly,
maturation, and stability of HLA–B*1402 and B*1403,
which differ by 1 amino acid change and are differentially associated with ankylosing spondylitis (AS), and
to compare these features with those of B*2705.
Methods. Stable transfectants expressing B*1402,
B*1403, and B*2705 were used. Folding rates were
estimated from the ratio of unfolded heavy chains to
folded heavy chains that had been immunoprecipitated
with specific antibodies in pulse–chase experiments.
Heavy chain misfolding was measured as the half-life of
endoglycosidase H (Endo H)–sensitive ␤2-microglobulin–
free heavy chains. Maturation/export rates were measured by acquisition of Endo H resistance. Association
with calnexin or tapasin was analyzed by coprecipitation
with chaperone-specific antibodies, and surface expression was estimated by flow cytometry. Thermostability of
HLA–peptide complexes was assessed by immunoprecipitation after incubation at various temperatures. Heavy
chain expression was quantified by Western blotting.
Results. The folding rates of B*1402 and B*1403
were similar, and both were faster and more efficient
than B*2705, but some unfolded heavy chains from both
B14 subtypes remained in the endoplasmic reticulum
(ER) with a long half-life. The export rates of B*1402
and B*1403 were slow, and the heterodimers partially
dissociated after exiting the ER, as revealed by significant
amounts of Endo H–resistant and surface-expressed free
heavy chains. Both interaction with tapasin and thermostability were higher for B*2705 than for B*1402 and
higher for B*1402 than for B*1403, suggesting that the
repertoires of the B*1402-bound peptide and especially
the B*1403-bound peptide were less optimized than that
of B*2705.
Conclusion. Our results indicate that the folding,
maturation, and stability of B*1403 differ more from
B*2705 than from B*1402. Thus, these features cannot
account for the fact that only the 2 former allotypes are
associated with AS.
Ankylosing spondylitis (AS) is a chronic rheumatic disease of unknown etiology that is strongly associated with HLA–B27 (1). The notion of direct involvement of this molecule in the pathogenesis of AS is
supported by the ubiquity of this association across
ethnic groups (2) and by the development of a disease
with many similarities to the human spondylarthritides
in HLA–B27–transgenic rats (3). The molecular properties underlying the pathogenetic role of HLA–B27 remain unclear. Current research is focused mainly on 4
possibilities. First, recognition of a specific self peptide
showing molecular mimicry with an external antigen
might trigger disease through an autoimmune mechanism (4). Another possibility is that the inefficient folding
of HLA–B27 and its tendency to misfold and accumulate
in the endoplasmic reticulum (ER) may trigger an
autoinflammatory reaction (5). Alternatively, surface
expression of heavy-chain homodimers may lead to
immunomodulation through their recognition by leukocyte receptors (6,7) or to aberrant immune responses
(8). Finally, it is possible that ␤2-microglobulin (␤2m)
release, resulting from dissociation of HLA–peptide
complexes, may elicit an inflammatory disease if trapped
within the synovium (9). We have recently suggested
that a more integrative approach, emphasizing the whole
Supported by grant SAF2005/03188 from the Spanish Ministry of Science and Technology and an institutional grant from the
Fundación Ramón Areces to the Centro de Biologı́a Molecular Severo
Elena Merino, PhD, Begoña Galocha, MD, PhD, Miriam N.
Vázquez, PhD, José A. López de Castro, PhD: Centro de Biologı́a
Molecular Severo Ochoa (Consejo Superior de Investigaciones Cientı́ficas and Universidad Autónoma de Madrid), Universidad Autónoma, Madrid, Spain.
Address correspondence and reprint requests to José A.
López de Castro, PhD, Centro de Biologı́a Molecular Severo Ochoa,
Calle Nicolás Cabrera 1, Universidad Autónoma de Madrid, 28049
Madrid, Spain. E-mail:
Submitted for publication June 2, 2008; accepted in revised
form August 15, 2008.
biology of HLA–B27 and the mutual dependence of its
molecular properties, might be needed to explain the
pathogenesis of AS (10).
When the nascent HLA class I heavy chain enters
the ER, it is glycosylated and bound to the lectin-like
chaperone calnexin (11). Upon incorporation of ␤2m,
calnexin is replaced by calreticulin (12). Folding and
assembly of the class I molecule proceed within the
peptide-loading complex, which also includes the covalently bound ERp57-tapasin subunit (13) and the
transporter associated with antigen processing (TAP)
(14). ERp57 catalyzes disulfide bond formation (15),
whereas tapasin mediates the optimization of the peptide cargo (16). Peptides reach the ER through TAP,
and may be trimmed to their optimal size for class I
binding by the ERAP1/ERAP2 aminopeptidase complex
(17). Heavy-chain polypeptides that fail to assemble
remain bound to calnexin or are dislocated to the cytosol
for proteasomal degradation (18,19). Heavy chain–␤2m–
peptide complexes exit the ER and migrate to the cell
surface. During transport, the high-mannose glycan moiety of the heavy chain is transformed to complex-type
oligosaccharide through successive cleavage of mannose
and progressive incorporation of other sugars (20,21).
Cell surface–expressed class I molecules recycle through
the endosomal compartment. For HLA–B27, this process is critical for the formation of surface heavy-chain
homodimers (22). Presumably, some of the molecules
dissociate when exposed to the lower pH of the endosome, to an extent that depends on the stability of the
peptide cargo.
HLA–B*1403 is found almost exclusively in subSaharan African (23,24) and African American populations (25,26), with a prevalence of ⬍1%. In sub-Saharan
Africa, AS is also very uncommon (27). Thus, the finding
that 4 of 8 unrelated Togolese patients with AS, an
exceptionally large series for an African population,
carried B*1403 constituted strong evidence of the association of this allele with AS (23). In the ethnically unrelated Bantu population of Zambia, where both B*1403
and AS are also very rare, 2 of the 3 unrelated patients
with AS who were available for study were B*1403
positive (28). Recently, a B*1403–positive patient with
AS was identified in Sardinia (29). This individual also
carried B*2709, an allotype not otherwise associated
with AS in this population (30). Taken together, these
findings strongly support the hypothesis that B*1403 is a
susceptibility factor for AS.
In contrast, B*1402, which differs from B*1403
only at position 156 (L in B*1402, R in B*1403), does
not predispose to AS. Since B*1402 shows a more general
distribution and is found in sub-Saharan and Caucasoid
populations, it is unlikely that its lack of association with
AS is haplotype-dependent, as has been suggested for
B*2709 (31). B*1402 and B*1403 share ⬃32–35% of
their peptide repertoires with each other and share ⬃3%
of their peptides and T cell epitopes with B*2705 (32),
which is compatible with the possibility that presentation
of a common arthritogenic peptide might underlie the
association of B*1403 and B*2705 with AS. In the
present study, we compared the folding, assembly, export, and stability of B*1402, B*1403, and B*2705 in
order to determine whether these features are also
compatible with the association of B*1403 and B*2705,
but not B*1402, with AS.
Cell lines, antibodies, and flow cytometric analysis.
HMy2.C1R (C1R) is a human lymphoid cell line with low
expression of its endogenous HLA class I molecules (33). C1R
transfectants expressing B*1402, B*1403, and B*2705 have
been described previously (32,34). B*0702 C1R transfectants
were also used. Analogous transfectants in the HLA class I–
negative human lymphoid 721.221 (.221) cell line (35) were
obtained as on C1R, but with electroporation at 260 mV and
500 ␮F. Cells were cultured in RPMI 1640 supplemented with
2 mM L-glutamine and 10% fetal bovine serum (FBS) (Gibco
Life Technologies, Paisley, UK). The monoclonal antibodies
(mAb) used were W6/32 (IgG2a, specific for a monomorphic
HLA class I determinant) (36), HC10 (IgG2a), which recognizes free class I heavy chains in monomeric and oligomeric
forms (37), AF8 (IgG1) (kind gift from Dr. Michael B.
Brenner, Brigham and Women’s Hospital, Boston, MA), which
recognizes human calnexin (38), PaSta-1 (kind gift from Dr.
Peter Cresswell, Howard Hughes Medical Institute, Yale University Medical School, New Haven, CT), which is specific for
human tapasin (15), and the anti–␥-tubulin mAb GTUBB
(IgG1) (Sigma, Steinheim, Germany). Cells were analyzed by
flow cytometry as previously described (32).
Western blot analysis. HLA–B14 and B27 protein
expression was estimated by Western blot analysis as previously described (39). For detecting heavy-chain homodimers,
⬃107 cells were lysed in Nonidet P40 (NP40) lysis buffer (0.5%
NP40, 50 mM Tris HCl [pH 7.4], and 5 mM MgCl2) containing
a cocktail of protease inhibitors (Complete Mini tablets;
Roche, Mannheim, Germany) and 10 mM iodoacetamide
(Calbiochem, Schwalbach, Germany). The lysates were incubated for 30 minutes at 4°C in the presence of the alkylating
agent and centrifuged for 10 minutes at 16,000g at 4°C. The
supernatant was subjected to 7.5% sodium dodecyl sulfate–
polyacrylamide gel electrophoresis (SDS-PAGE) under nonreducing or reducing conditions. Blots were performed with
HC10 as previously described (39).
Metabolic labeling, pulse–chase analysis, and immunoprecipitation. Cells were incubated with L-methionine–free
and L-cysteine–free Dulbecco’s modified Eagle’s medium supplemented with 10% FBS and 2 mM L-glutamine for 45 minutes
at 37°C. Cells were pulse-labeled with 500–1,000 ␮Ci/ml of
S-methionine/cysteine (Amersham, Buckinghamshire, UK)
at 37°C and chased for various times with complete RPMI 1640
supplemented with 1 mM cold L-methionine and L-cysteine at
37°C. At each time point, cells were spun down, resuspended in
50 ␮l of PBS, frozen in liquid nitrogen, and stored at ⫺80°C.
Cells were lysed in NP40 lysis buffer (0.5% NP40, 50 mM Tris
HCl [pH 7.4], and 5 mM MgCl2) containing a cocktail of
protease inhibitors (Complete Mini tablets). Lysates were
centrifuged at 16,000g for 10 minutes at 4°C and precleared 3
times at 4°C for 60 minutes with Sepharose CL-4B beads
(Sigma) and 3 ␮l of normal mouse serum, followed by immunoprecipitation with specific mAb and protein A–Sepharose
beads (Sigma) for 60 minutes. Immunoprecipitates were normalized to trichloroacetic acid–precipitable 35S-labeled protein. The immunoprecipitates were washed 3 times with NP40
washing buffer (0.5% NP40, 50 mM Tris HCl [pH 7.4], 150 mM
NaCl, and 5 mM EDTA) and analyzed by 10% SDS-PAGE
under reducing conditions. Endoglycosidase H (Endo H) was
added to the immunoprecipitates, according to the recommendations of the manufacturer (New England Biolabs, Beverly,
MA). Samples were visualized by fluorography as previously
described (39).
Fast folding of HLA–B14 subtypes. C1R transfectants were labeled for 15 minutes and chased for up
to 4 hours. One half of each sample was immunoprecipitated with W6/32 and the other half with HC10, and
both were analyzed by SDS-PAGE. HLA–B14 and B27
assembly was examined by measuring the disappearance
of free heavy chains and the appearance of assembled
heavy chain–␤2m–peptide complexes with time. In all
cases, the heavy chain was in complex with ␤2m at the
onset of the chase (Figure 1A, lane 1). However, the recovery of mature molecules decreased over time in both
B14 subtypes, while it increased in B*2705 (Figure 1A,
lanes 1–5). A significant amount of ␤2m-free heavy chain
from B*1402, B*1403, and B*2705 was precipitated with
HC10 along the chase (Figure 1A, lanes 6–10), suggesting that free heavy chains of the 3 allotypes accumulated in the ER. A characteristic decrease in electrophoretic mobility, which occurs when N-linked glycans
are modified to complex oligosaccharides during transit
through the Golgi, was observed along the chase in the
B14 and B27 heavy chains precipitated with W6/32. The
same pattern was observed in the samples immunoprecipitated with HC10 from both B14 subtypes, but not
from B27, after 4 hours of chase (Figure 1A, lane 10).
These results suggested that a fraction of the assembled
B*1402 and B*1403 molecules may dissociate after exiting
the ER.
Quantitative analysis of the pulse–chase data
indicated that there was no difference in the ratio of
unfolded heavy chains, which were precipitated with
HC10, to assembled heavy chains, which were precipi-
Figure 1. Folding rates of HLA–B27 and HLA–B14 subtypes. A, Sodium
dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) analysis
of folded HLA–B14 and B27 complexes and free heavy chains immunoprecipitated (IP) with W6/32 (lanes 1–5) and HC10 (lanes 6–10). C1R
transfectants were labeled for 15 minutes and chased for the indicated
times. Asterisks denote a nonspecific polypeptide observed in nontransfected C1R cells (not shown). Arrows indicate the heavy chains before
(left) and after (right) modification of the glycan moiety to complex
oligosaccharide. B, Ratio of free heavy chains (precipitated by HC10) to
assembled heavy chains (precipitated by W6/32) in B*1402, B*1403, and
B*2705 at the indicated time points. Radioactive signals of HC10 and
W6/32 were quantified by densitometry. Values are the mean ⫾ SD of 3
experiments with B*2705 and 5 experiments with the B14 subtypes. C,
Western blot of whole cell lysates of nontransfected (⭋), B*2705, B*1402,
B*1403, and B*0702 (B7) C1R cells, obtained in the presence of 10 mM
iodoacetamide. The lysates were fractionated by SDS-PAGE under nonreducing or reducing conditions and blotted with HC10. The bands
around 45 kd and 80 kd correspond to monomeric and oligomeric heavy
chains, respectively. Each blot corresponds to a single gel, which was cut
to remove unrelated samples run alongside.
tated with W6/32 (hereafter designated as the HC10:
W6/32 ratio), between the B14 subtypes. This ratio was
higher for B*2705, particularly at early chase times
Figure 2. Intracellular dissociation and incomplete folding of HLA–B14 subtypes. A, Sodium dodecyl sulfate–polyacrylamide gel electrophoresis
(SDS-PAGE) analysis of endoglycosidase H (Endo H)–resistant (HC⫹CHO) and Endo H–sensitive (HC⫺CHO) heavy chains (HC) in B*1402,
B*1403, and B*2705. C1R cells were labeled for 15 minutes and chased for the indicated times. Free heavy chains were immunoprecipitated (IP)
with HC10. Half of the sample was digested with Endo H prior to analysis. Results are representative of ⱖ3 experiments. Asterisks denote a
nonspecific polypeptide observed with normal mouse serum (not shown). B, Percentage of Endo H–resistant heavy chain (HCr), which was
immunoprecipitated with HC10, in B*1402 and B*1403 at the indicated time points. Values are the mean ⫾ SD of 3 experiments. C, Half-life (T ⁄ )
of the Endo H–sensitive heavy chain (HCs) in B*1402, B*1403, and B*2705, as calculated from the pulse–chase data. Bars show the mean and
SD of 3 experiments. Values above the bars are the mean. D, Ratio of heavy chains to calnexin (CNX) in B*1402 (squares), B*1403 (triangles),
and B*2705 (diamonds) at the indicated time points. C1R transfectants were labeled and chased as described in A. Unfolded heavy chains were
recovered by coimmunoprecipitation with the anticalnexin monoclonal antibody AF8, and analyzed by SDS-PAGE. Results of a representative
experiment are shown. Values are the mean ⫾ SD of 3 experiments with the B14 subtypes and 7 experiments with B*2705.
(Figure 1B). Thus, the folding rates of the 2 B14
subtypes were similar and were significantly faster than
the folding rate of B*2705. The same results were
obtained with .221 transfectant cells (data not shown).
Formation of covalent heavy-chain homodimers
by B*1402 and B*1403. The occurrence of disulfidelinked heavy-chain homodimers in B*1402 and B*1403
C1R cells was analyzed by Western blotting of whole
lysates obtained in the presence of iodoacetamide using
HC10 (Figure 1C). Both B14 subtypes showed an ⬃80-kd
band under nonreducing conditions that was susceptible
to reducing agents. The high molecular weight B14 bands
had less heterogeneity and a slightly faster electrophoretic mobility than those of B*2705. We do not know
the reason for this difference, but it might be related to
a differential Cys residue between HLA–B27 and B14.
Both allotypes share, besides the canonical disulfide
bonds, the free Cys67 and Cys310 residues, but B*2705
has Cys325, which is changed to Ser in HLA–B14.
Whether the B14 homodimers arise in the ER, the cell
surface, or both was not analyzed.
Intracellular dissociation and incomplete folding
of HLA–B14 subtypes. The decrease in HLA–B14 heterodimers with time and the extra band of lower electrophoretic mobility that precipitated with HC10 at late
chase times (Figure 1A) suggested that B14 might
dissociate after exiting the ER. To examine this possibility, we analyzed the presence of Endo H–resistant
␤2m-free heavy chains on the material immunoprecipitated with HC10 from pulse-labeled cells. Endo
H–resistant heavy chains were observed at 0.5 hours and
later chase times for B*1402 and B*1403, but not at all
or only marginally for B*2705 (Figure 2A). The percentage of Endo H–resistant ␤2m-free heavy chains increased with time, reaching significant levels at 4 hours
(mean ⫾ SD 36 ⫾ 4% and 37 ⫾ 9% for B*1402 and
B*1403, respectively) (Figure 2B). The significant amount
of Endo H–sensitive heavy chain precipitated with HC10
even at 4 hours of chase time (Figure 2A) indicated that,
like B*2705, some of the B*1402 and B*1403 heavy
chains failed to assemble and remained in the ER. The
half-life of the Endo H–sensitive heavy chain was similar
for B*1402, B*1403, and B*2705 (Figure 2C). Endo H–
resistant heavy chain was also immunoprecipitated with
HC10 from B*1402 and B*1403 .221 transfectants (data
not shown).
To analyze the association of the B*1402,
B*1403, and B*2705 heavy chains with calnexin, C1R
cells were labeled for 15 minutes, chased up to 4 hours,
and immunoprecipitated with the anticalnexin mAb AF8
(Figure 2D). Calnexin coimmunoprecipitated with the
heavy chain along the chase in all 3 cases, but the
amount of calnexin-bound heavy chain was higher for
B*2705. The 3 molecules were released from calnexin
with similar kinetics. These results confirmed that a
fraction of the B*1402 and B*1403 heavy chains remained unfolded in the ER.
High surface levels of free heavy chains in HLA–
B14 subtypes. Cell surface expression of HLA–B27
heavy-chain homodimers occurs upon endosomal recycling (22), presumably following dissociation of the
canonical heterodimers. In B*2705 C1R cells, the surface expression of ␤2m-free heavy chains is ⬃5% of the
heterodimer (40). Thus, we reasoned that if HLA–B14–
peptide complexes were less stable than HLA–B27 and
dissociated to a larger extent after exiting the ER, the
surface expression of HC10-reactive molecules should
be higher in B*1403 and B*1402 than in B*2705. That
this was the case was shown by flow cytometry of C1R
transfectants with W6/32 and HC10 (Figure 3). For
B*1402 and B*1403, the expression of ␤2m-free heavy
chains, relative to heterodimers, as measured by the
HC10:W6/32 fluorescence ratio, was similar (⬃0.30 and
0.35, respectively) and significantly higher than for
HLA–B27 (⬃0.07). Similar results were obtained with
B*1402 and B*1403 .221 transfectants (data not shown).
The amount of free B*2705 heavy chain was too small to
show up among the Endo H–resistant material immunoprecipitated with HC10 (Figure 2A) unless the autoradiograms were heavily overexposed (results not shown).
Slower export rates for HLA–B14 subtypes than
for HLA–B27. Endo H cleaves only the heavy-chain
glycan moieties that have not been processed by medial
Figure 3. Cell surface expression of free heavy chain from HLA–B14
and HLA–B27. A, Flow cytometric analysis of B*1402, B*1403, and
B*2705 expression on C1R transfectants stained with W6/32 and HC10.
Nontransfected C1R cells were used as a control. Results of a representative experiment are shown. B, Comparison of the surface expression of
heterodimers (open bars) and free heavy chains (solid bars) in C1R,
B*1402, B*1403, and B*2705. Bars show the mean and SD fluorescence
of ⱖ6 experiments. Values above the bars are the mean.
Golgi enzymes to complex glycans. Thus, the export rate
of HLA class I molecules from the ER was assessed by
measuring the acquisition of Endo H resistance on the
heavy chain–␤2m–peptide complexes, which were immu-
noprecipitated with W6/32, in pulse–chase experiments.
C1R transfectants were labeled for 15 minutes and
chased for up to 4 hours. Following immunoprecipitation with W6/32, one half of the sample was treated with
Endo H, the other half was left untreated, and samples
were analyzed by SDS-PAGE (Figure 4A). For B*2705,
conversion to Endo H–resistant forms was calculated
from the ratio of Endo H–sensitive heavy chains to Endo
H–resistant heavy chains as a function of time. For
B*1402 and B*1403 the same procedure was used, but
Endo–H resistant heavy chains precipitated with W6/32
and HC10 were both taken into account, since the latter
presumably arose from dissociation of exported heterodimers (Figure 2A). Conversion to 50% Endo H–
resistant forms occurred for B*1402, B*1403, and B*2705
in ⬃113 ⫾ 2, 113 ⫾ 6, and 69 ⫾ 5 minutes (mean ⫾ SD),
respectively (Figure 4B). This indicated that both B14
subtypes had similar export rates, which were slower
than the rate for B*2705. Similar results were obtained
with B*1402 and B*1403 .221 transfectants (data not
To rule out the possibility that this difference
resulted from different assembly rates, C1R transfectants were metabolically labeled for 1 minute to allow
for completion of the heavy chain synthesis, chased for
various times, and immunoprecipitated with W6/32.
Properly folded class I complexes were observed after
1 minute of chase time in all cases (Figure 4C).
Differing interaction with tapasin and differing
stability of peptide repertoires in HLA–B14 subtypes
and B*2705. C1R transfectant cells were labeled for 2
hours, lysed, immunoprecipitated with the antitapasin
mAb PaSta-1, and analyzed by SDS-PAGE (Figure 5A).
Drastic differences in the amount of heavy chain that
coprecipitated with tapasin (expressed as the ratio of
heavy chain to tapasin) were observed. B*2705 had the
highest ratio of heavy chain to tapasin (mean ⫾ SD
1.3 ⫾ 0.1), followed by B*1402 (0.6 ⫾ 0.2) and then by
B*1403 (0.16 ⫾ 0.03). The lower interaction of B*1403
relative to that of B*1402 and B*2705 was also observed
on .221 transfectants (data not shown). To further
characterize these differences, metabolically labeled
C1R cells were chased for up to 4 hours and immunoprecipitated with PaSta-1. The B*1402, B*1403, and
B*2705 heavy chains were associated with tapasin all
along the chase (Figure 5B), but the ratios of heavy
chain to tapasin were highest for B*2705, intermediate
for B*1402, and lowest for B*1403 (Figure 5C). Moreover, the half-life of the tapasin-bound heavy chain was
shorter for B*1403 (mean ⫾ SD 75 ⫾ 4 minutes) than
for B*1402 (142 ⫾ 17 minutes) or B*2705 (176 ⫾ 13
minutes) (Figure 5D). These results suggested that the
Figure 4. Maturation rates of HLA–B27 and HLA–B14 subtypes. A,
SDS-PAGE analysis of Endo H–resistant and Endo H–sensitive heavy
chains in B*1402, B*1403, and B*2705. C1R transfectant cells were
metabolically labeled for 15 minutes and chased for the indicated
times. Properly folded class I heterodimers were recovered from the
lysates with W6/32. Half of the sample was digested with Endo H prior
to analysis. B, Time required to obtain 50% Endo H–resistant forms of
heavy chains in B*1402, B*1403, and B*2705. Radioactive signals of
Endo H–resistant and Endo H–sensitive heavy chains were quantified
by densitometry. Bars show the mean and SD of 3 experiments. Values
above the bars are the mean. C, SDS-PAGE analysis of properly folded class I complexes recovered from the lysates with W6/32.
C1R transfectants were metabolically labeled for 1 minute and chased
for the indicated times. Folded heterodimers were observed at 1 minute
in all 3 cases. The weaker signals (relative to the next time point)
observed at 1 minute for B*1402 and at 5 minutes for B*1403 reflect
only inaccuracies in the total amount of radioactivity loaded onto the
gel, as observed by the lower general background across the whole gel
in these particular lanes (not shown). See Figure 2 for definitions.
interaction of B*1403 with tapasin was weaker, shorter,
or both than the interaction of B*1402 or B*2705 with
Since the role of tapasin is to optimize the class I
MHC–bound peptide repertoire (16), we analyzed the
Figure 5. Tapasin (Tpn) binding of HLA–B14 subtypes. A, SDS-PAGE
analysis of C1R transfectants expressing B*1402, B*1403, or B*2705.
Transfectants were labeled for 2 hours. The lysates were subjected
to immunoprecipitation with the antitapasin monoclonal antibody PaSta-1. Results are representative of ⱖ3 experiments. B, SDSPAGE analysis of heavy chains coimmunoprecipitated with PaSta-1.
Cells were labeled for 2 hours and chased for the indicated times. A
representative experiment is shown. In A and B, upper and lower
asterisks denote nonspecific polypeptides observed with normal mouse
serum and in nontransfected C1R cells, respectively (not shown). C,
Ratio of heavy chains to tapasin in B*1402, B*1403, and B*2705
at the indicated time points. Radioactive signals of heavy chains
and tapasin were coprecipitated with PaSta-1 and quantified. Values
are the mean ⫾ SD of 3 experiments with the B14 subtypes
and 5 experiments with B*2705. D, Half-life (T ⁄ ) of the tapasin-bound
heavy chain (HC-Tpn) in B*1402, B*1403, and B*2705, calculated from the pulse–chase data. Bars show the mean and SD of
3 experiments with the B14 subtypes and 4 experiments with B*2705.
Values above the bars are the mean. See Figure 2 for other definitions.
global thermostability of B*1402, B*1403, and B*2705
as a measure of the stability of their peptide cargoes.
C1R transfectants were pulse-labeled for 15 minutes,
chased for 0, 2, or 4 hours, immunoprecipitated with
W6/32 after incubation of the lysates for 1 hour at
various temperatures, and analyzed by SDS-PAGE
(Figure 6A). The amount of heterodimer precipitated
at each temperature at any given time was expressed
as a percentage of the amount precipitated at 4°C and,
for each time point, was plotted as a function of the temperature (Figure 6B). At 0 hours, significant differences
were observed among the 3 allotypes, with B*1403 showing
the lowest thermostability. Progressive improvement
was observed for this allotype, and the thermostability
profile of B*1403 equaled that of B*1402 at 2 hours of
chase. After 4 hours, further improvement occurred in
both B14 subtypes, which was particularly evident at 42°C,
and in B*2705. These results indicated that B*1403 was
globally less stable than B*1402 shortly after assembly.
Although the thermostability of both B14 subtypes
increased with time, they did not reach that of B*2705.
Differences in heavy chain expression among
transfectants do not account for the differential features
of HLA–B14 and HLA–B27. To rule out the possibility
that the observed differences in folding, maturation, and
thermostability resulted from different heavy chain expression levels among the transfectants used, the heavychain proteins were quantified by Western blot analysis
(Figure 6C). Mean ⫾ SD heavy chain expression in
B*1402 and B*1403 relative to that of B*2705 was 1.4 ⫾
0.3 and 1.3 ⫾ 0.1, respectively. The relative heavy-chain
protein levels and the relative values for the different
parameters we analyzed were compared among the 3
allotypes (Figure 6C). Except for the maturation rate
(50% heavy chain resistant), there was no correlation
between heavy chain expression and other parameters.
The only differences between B*1402 and B*1403 were
those related to the quality control of the peptide cargo,
that is, the parameters measuring tapasin interaction (the
ratio of heavy chain to tapasin and the half-life of
tapasin-bound heavy chain) and thermostability of the
MHC–peptide complexes. For both of these features,
B*1403 and B*2705 showed maximal disparity.
The relevance of our study to the pathogenesis of
AS is dependent on 2 issues concerning B*1403. The
first is its epidemiology. This is necessarily more limited
than for the much more prevalent B*2705, which is
found in populations throughout the world. However,
given the low prevalence of both B*1403 and AS in
Figure 6. Thermostability of B*1402, B*1403, and B*2705. A, SDS-PAGE analysis of B*2705, B*1402, and B*1403 at the indicated
temperatures and time points. C1R transfectants were labeled for 15 minutes and chased for the indicated times. Equal aliquots of
the lysates were kept at 4°C or heated at the indicated temperatures for 1 hour prior to immunoprecipitation with W6/32, separated
by SDS-PAGE, and analyzed by fluorography. Asterisks denote a nonspecific polypeptide observed with normal mouse serum (not
shown). Arrows indicate heavy chains. B, Percentage of W6/32-reactive HLA–peptide complexes recovered after heating after 0, 2,
or 4 hours of chase, plotted as the intensity value of the class I heavy chain at any given temperature (HCX) relative to that at 4°C
(HC4). Values are the mean ⫾ SD of 4 experiments with B*2705 and B*1402 and 5 experiments with B*1403. C, Relative heavy chain
expression, HC10-to-W6/32 ratio, half-life (T ⁄ ) of the Endo H–sensitive heavy chain (HCs), heavy chain–to–calnexin (CNX) ratio,
maturation rate (time to conversion to 50% Endo H–resistant forms), heavy chain–to-tapasin ratio, half-life of the tapasin-bound
heavy chain (HC-Tpn), thermostability at 42°C, and surface HC10-to-W6/32 ratio in B*1402, B*1403, and B*2705. Values were
compared with the relative heavy-chain protein expression in the corresponding transfectants, as determined by Western blotting. A
Western blot showing representative results of 3 independent experiments is shown. See Figure 2 for other definitions.
sub-Saharan Africa, it seems very unlikely that the
frequency of this allotype among unrelated AS patients
from 2 African populations is independent of a pathogenetic role of this allotype. The co-occurrence of B*1403
in the only B*2709-positive Sardinian patient with AS
detected so far (29) must be assessed in this context.
The second issue concerns the putative mechanism by which B*1403 may be associated with AS. There
are 3 possibilities. The first is that the predisposing factor
is not B*1403, but a closely linked gene. The second
possibility is that B*1403 predisposes to AS through a different mechanism than does B*2705, and the third possibility is that B*1403 predisposes to AS by the same
mechanism as does B*2705. We cannot rule out any of
these alternatives. However, we consider the first one
unlikely for the following reason. AS is an HLA class I–
associated disease in which the notion of a direct involvement of HLA–B27 is generally accepted based on
its association with disease across different ethnic groups
and haplotypes, and on the occurrence of spondylarthritis in HLA–B27–transgenic rats (3). Therefore, it is
reasonable that if AS is associated with another class I
allotype, the latter should also be directly involved in
pathogenesis, without excluding additional susceptibility
genes that also influence HLA–B27–associated disease.
We also consider the second possibility unlikely for the
following reason. If the same disease is associated with 2
different HLA class I allotypes, and it is assumed that in
both cases they are directly involved in the disease
mechanism, the simplest explanation is that their pathogenetic behavior is related to a common feature, rather
than to unrelated ones. Thus, we interpreted the results
of the present study under the assumption that B*1403 is
associated with AS through a similar mechanism as
The folding, assembly, export, and thermostability of B*2705 were characterized in previous studies
(16,39,41–43) and were used in this study only for
comparison with HLA–B14. Consistent with the results
of those previous studies, B*2705 showed slow folding
and a tendency to misfold and accumulate in the ER.
B*2705 also has a relatively slow export rate compared
with other HLA–B27 subtypes or with HLA–B7 (39).
Although not to the extent of other HLA class I molecules
(44), both the nature and significant stability of the
B*2705-bound peptide repertoire are tapasin-dependent
(16,45). Thus, at the cell surface of tapasin-proficient
cells, such as C1R, only low amounts of ␤2m-free heavy
chain are detected (40).
B*1402 and B*1403 showed similar folding and
export rates, as assessed by the HC10:W6/32 ratio (Figure
1B) and the percentage of Endo H–resistant heavy chain
precipitated with W6/32 (Figure 4B), respectively, but
they differed from B*2705 in that the amount of heavy
chain that failed to assemble was smaller for both B14
subtypes and in that their assembly was faster (Figures
1A and B). Misfolded B14 heavy chains remained in the
ER as long as misfolded B27 heavy chains, but there was
no correlation between either the amount or the half-life
of misfolded heavy chains and the association (B*2705
and B*1403) or lack of association (B*1402) with AS.
Like B*2705, both B*1402 and B*1403 formed covalent
heavy-chain homodimers, although their electrophoretic
heterogeneity was lower than that of HLA–B27. This
difference might be related to the fact that HLA–B14
lacks the Cys325 residue present in B*2705. The cellular
localization of the B14 homodimers was not analyzed.
They might be in the ER, at the cell surface, or both. We
favor the latter possibility based on the presence of Cys67
in B14, since this residue participates in the formation of
homodimers in both cell compartments (22,42,46).
In contrast to their faster folding rates, the export
rates of the B14 subtypes were slower than that of B*2705.
This is consistent with the possibility that B14 heterodimers need more time than B*2705 to obtain the level
of peptide optimization required to exit the ER, perhaps as a consequence of their less efficient interaction
with tapasin. The finding of Endo H–resistant free heavy
chain, which was precipitable with HC10, whose rate of
appearance correlated with a time-dependent decrease
in B14 heterodimers, suggested that even after exiting
the ER, the B14-bound peptide cargo was less optimized
in terms of its stability than that of B*2705. The most
likely interpretation of this observation is that B14–peptide
complexes dissociate from the cell surface at late maturation stages, probably upon endosomal recycling (22).
Indeed, a high proportion of ␤2m-free B14 heavy chain
was detected at the cell surface compared with B*2705.
In all of these features, B*1402 and B*1403 showed similar
behavior. Thus, neither the export rate, the dissociation
extent of dissociation of the exported, nor the surface
expression of ␤2m-free heavy chains account for the
differential association of B*1402 and B*1403 with AS.
The interaction with tapasin and the global thermostability shortly after assembly distinguished B*1402
from B*1403. Coprecipitation experiments showed less
interaction of both B14 subtypes with tapasin than of
B*2705 with tapasin. Yet, this interaction was ⬃4-fold
more efficient for B*1402 than for B*1403, as assessed
by the ratio of heavy chain to tapasin. This may be due
to the L156R change in B*1403, since it has been
suggested that this position affects interaction with TAP
(47), which is mediated by tapasin (12,48). The lower
thermostability of B*1403 relative to B*1402 is consistent with the less optimized peptide repertoire of
B*1403 and with its less efficient interaction with tapasin.
B*1402 and B*1403 ligands show a high frequency of acidic P1 residues. In addition, B*1403 has an
increased preference for peptides with acidic P3 residues
(32). Acidic P1 and P3 residues are strongly disfavored
for TAP binding (49), a situation very much unlike that
of B*2705 ligands, whose preferred N-terminal motifs,
basic P1 and aliphatic/aromatic P3 residues (50), are
much better adjusted to TAP preferences. Thus, both
the lower interaction with tapasin and the suboptimal
coupling of peptide specificity with TAP preferences
suggest that tapasin may be less efficient in optimizing
the B*1402 and, especially, B*1403 peptide repertoires
than the B*2705 peptide repertoire.
The thermostability ranking of newly synthesized
HLA–peptide complexes (with B*2705 showing the
highest degree of thermostability, followed by B*1402,
followed by B*1403) mirrored the relative efficiency of
their interactions with tapasin, indicating that immediately after assembly, the B*1402- and, to an even greater
degree, B*1403-bound peptide repertoires are less optimized than that of B*2705. A progressive optimization
of the B*1402- and B*1403-bound peptide cargoes was
observed with time, which also occurred, but to a lesser
extent, in B*2705. The time-dependent optimization of
B14 occurred equally in the presence of brefeldin A
(Merino E, et al: unpublished observations), suggesting
that the optimization of B*1402 and B*1403 with time is
tapasin-dependent and takes place in the ER. This is
consistent with the slow export rate of the B14 subtypes.
The suboptimal stability of the B14-bound peptide repertoires compared with B*2705 explains the observed
intracellular dissociation and the high surface expression
of free B*1402 and B*1403 heavy chains. That both
subtypes showed similar post-ER dissociation suggests
that the optimization of their peptide cargoes is similar
when they exit the ER.
There are at least 2 potentially pathogenetic consequences of a suboptimal B*1403-bound peptide repertoire. Upon dissociation in late cell compartments,
B*1403 molecules may 1) be subjected to peptide interchange, enabling peptides with lower dissociation rates
(for instance, foreign or self ligands processed outside
the ER) to be presented at the cell surface, or 2) reassociate with covalent heavy-chain homodimers that
would be expressed at the cell surface. The first mechanism may allow the presentation of bacterial peptides
generated at late cell compartments, making B*1403 a
suitable molecule for presenting these peptides and eliciting T cell responses against arthritogenic pathogens. The
second mechanism would favor the formation of homodimers that, as proposed for HLA–B27 (6,7), may be
recognized by leukocyte receptors and modulate T cell
However, neither of these possibilities explains
the differential association of B*1402 and B*1403 with
AS, since for those features in which the 2 subtypes
differ, B*1402 is more similar to B*2705 than is B*1403.
For the same reason, a pathogenetic mechanism based
on ␤2m release following dissociation of the HLA–
peptide complexes (9) would also seem to be inconsistent with the relative thermostability of the 3 allotypes.
Although our study was carried out mainly with C1R
cells, many of the experiments were also performed on
.221 transfectants and yield similar results. Thus, it is
unlikely that the behavior of B*1402 or B*1403 was
biased by the cell line used.
Our results could reflect that B*1403 and B*2705
predispose to AS by different mechanisms or that the
association of B*1403 with disease is due to a closely
linked gene. Without ruling out these possibilities, we
discussed above why we do not favor them. Thus, if the
mechanism underlying the association of B*1403 with
AS is the same as that for HLA–B27, which seems to us
the most reasonable alternative, our results are inconsistent with a pathogenetic role of misfolding or free
surface-expressed heavy chains. This led us to reassess
the significance of the low level of peptide and T cell
epitope sharing between B*1403 and B*2705 (32). In
view of the lack of correlation of the folding and stability
of B*1402, B*1403, and B*2705 with an association with
AS, the pathogenetic relevance of a shared ligand of
B*1403 and B*2705 might currently be the only obvious
alternative that is not in opposition to the comparative
biology of the 3 allotypes. Regardless of the precise
mechanism, our study shows the relevance of peptide
binding as the central feature determining the cell
biology of B*1402, B*1403, and B*2705.
We thank Dr. Michael B. Brenner (Brigham and
Women’s Hospital, Harvard Medical School, Boston, MA) and
Dr. Peter Cresswell (Howard Hughes Medical Institute, Yale
University School of Medicine, New Haven, CT) for providing
the anticalnexin and antitapasin monoclonal antibodies, respectively. We also thank Dr. Carlos López-Larrea (Universidad de Oviedo, Oviedo, Spain) and Dr. John Reveille
(University of Texas Health Science Center, Houston, TX) for
kindly sharing their unpublished data.
Dr. López de Castro had full access to all of the data in the
study and takes responsibility for the integrity of the data and the
accuracy of the data analysis.
Study design. Galocha, López de Castro.
Acquisition of data. Merino, Galocha, Vázquez.
Analysis and interpretation of data. Merino, Galocha, Vázquez, López
de Castro.
Manuscript preparation. López de Castro.
1. Brewerton DA, Hart FD, Nicholls A, Caffrey M, James DC,
Sturrock RD. Ankylosing spondylitis and HL-A 27. Lancet 1973;
2. Gonzalez-Roces S, Alvarez MV, Gonzalez S, Dieye A, Makni H,
Woodfield DG, et al. HLA-B27 polymorphism and worldwide
susceptibility to ankylosing spondylitis. Tissue Antigens 1997;49:
3. Hammer RE, Maika SD, Richardson JA, Tang JP, Taurog JD.
Spontaneous inflammatory disease in transgenic rats expressing
HLA-B27 and human ␤2m: an animal model of HLA-B27associated human disorders. Cell 1990;63:1099–112.
4. Benjamin R, Parham P. Guilt by association: HLA-B27 and
ankylosing spondylitis. Immunol Today 1990;11:137–42.
5. Colbert RA. HLA-B27 misfolding: a solution to the spondyloarthropathy conundrum? Mol Med Today 2000;6:224–30.
6. Kollnberger S, Bird L, Sun MY, Retiere C, Braud VM, McMichael
A, et al. Cell-surface expression and immune receptor recognition
of HLA–B27 homodimers. Arthritis Rheum 2002;46:2972–82.
7. Allen RL, Trowsdale J. Recognition of classical and heavy chain
forms of HLA-B27 by leukocyte receptors. Curr Mol Med 2004;
8. Edwards JC, Bowness P, Archer JR. Jekyll and Hyde: the transformation of HLA-B27. Immunol Today 2000;21:256–60.
9. Uchanska-Ziegler B, Ziegler A. Ankylosing spondylitis: a ␤2mdeposition disease? Trends Immunol 2003;24:73–6.
10. Marcilla M, Lopez de Castro JA. Peptides: the cornerstone of
HLA-B27 biology and pathogenetic role in spondyloarthritis.
Tissue Antigens 2008;71:495–506.
Vassilakos A, Cohen-Doyle MF, Peterson PA, Jackson MR,
Williams DB. The molecular chaperone calnexin facilitates folding
and assembly of class I histocompatibility molecules. EMBO J
Sadasivan B, Lehner PJ, Ortmann B, Spies T, Cresswell P. Roles
for calreticulin and a novel glycoprotein, tapasin, in the interaction
of MHC class I molecules with TAP. Immunity 1996;5:103–14.
Wearsch PA, Cresswell P. Selective loading of high-affinity peptides onto major histocompatibility complex class I molecules by
the tapasin-ERp57 heterodimer. Nat Immunol 2007;8:873–81.
Ortmann B, Androlewicz MJ, Cresswell P. MHC class I/␤2microglobulin complexes associate with TAP transporters before
peptide binding. Nature 1994;368:864–7.
Dick TP, Bangia N, Peaper DR, Cresswell P. Disulfide bond
isomerization and the assembly of MHC class I-peptide complexes. Immunity 2002;16:87–98.
Williams AP, Peh CA, Purcell AW, McCluskey J, Elliott T.
Optimization of the MHC class I peptide cargo is dependent on
tapasin. Immunity 2002;16:509–20.
Saveanu L, Carroll O, Lindo V, Del Val M, Lopez D, Lepelletier
Y, et al. Concerted peptide trimming by human ERAP1 and
ERAP2 aminopeptidase complexes in the endoplasmic reticulum.
Nat Immunol 2005;6:689–97.
Wiertz EJ, Jones TR, Sun L, Bogyo M, Geuze HJ, Ploegh HL. The
human cytomegalovirus US11 gene product dislocates MHC class
I heavy chains from the endoplasmic reticulum to the cytosol. Cell
Hughes EA, Hammond C, Cresswell P. Misfolded major histocompatibility complex class I heavy chains are translocated into
the cytoplasm and degraded by the proteasome. Proc Natl Acad
Sci U S A 1997;94:1896–901.
Barber LD, Patel TP, Percival L, Gumperz JE, Lanier LL, Phillips
JH, et al. Unusual uniformity of the N-linked oligosaccharides of
HLA-A, -B, and -C glycoproteins. J Immunol 1996;156:3275–84.
Helenius A, Aebi M. Intracellular functions of N-linked glycans.
Science 2001;291:2364–9.
Bird LA, Peh CA, Kollnberger S, Elliott T, McMichael AJ,
Bowness P. Lymphoblastoid cells express HLA-B27 homodimers
both intracellularly and at the cell surface following endosomal
recycling. Eur J Immunol 2003;33:748–59.
Lopez-Larrea C, Mijiyawa M, Gonzalez S, Fernandez-Morera JL,
Blanco-Gelaz MA, Martinez-Borra J, et al. Association of ankylosing spondylitis with HLA–B*1403 in a West African population.
Arthritis Rheum 2002;46:2968–71.
Ellis JM, Mack SJ, Leke RF, Quakyi I, Johnson AH, Hurley CK.
Diversity is demonstrated in class I HLA-A and HLA-B alleles in
Cameroon, Africa: description of HLA-A*03012, *2612, *3006 and
HLA-B*1403, *4016, *4703. Tissue Antigens 2000;56:291–302.
Cao K, Hollenbach J, Shi X, Shi W, Chopek M, Fernandez-Vina
MA. Analysis of the frequencies of HLA-A, B, and C alleles and
haplotypes in the five major ethnic groups of the United States
reveals high levels of diversity in these loci and contrasting
distribution patterns in these populations. Hum Immunol 2001;62:
Tu B, Mack SJ, Lazaro A, Lancaster A, Thomson G, Cao K, et al.
HLA-A, -B, -C, -DRB1 allele and haplotype frequencies in an
African American population. Tissue Antigens 2007;69:73–85.
Mijiyawa M, Oniankitan O, Khan MA. Spondyloarthropathies in
sub-Saharan Africa. Curr Opin Rheumatol 2000;12:281–6.
Diaz-Pena R, Blanco-Gelaz MA, Njobvu P, Lopez-Vazquez A,
Suarez-Alvarez B, Lopez-Larrea C. Influence of HLA–B*5703
and HLA–B*1403 on the susceptibility to spondyloarthropathies
in the Zambian population. J Rheumatol 2008. In press.
Cauli A, Vacca A, Mameli A, Passiu G, Fiorillo MT, Sorrentino R,
et al. A Sardinian patient with ankylosing spondylitis and
HLA–B*2709 co-occurring with HLA–B*1403. Arthritis Rheum
30. Paladini F, Taccari E, Fiorillo MT, Cauli A, Passiu G, Mathieu A,
et al. Distribution of HLA–B27 subtypes in Sardinia and continental Italy and their association with spondylarthropathies.
Arthritis Rheum 2005;52:3319–21.
31. Cascino I, Paladini F, Belfiore F, Cauli A, Angelini C, Fiorillo MT,
et al. Identification of previously unrecognized predisposing factors for ankylosing spondylitis from analysis of HLA–B27 extended
haplotypes in Sardinia. Arthritis Rheum 2007;56:2640–51.
32. Merino E, Montserrat V, Paradela A, Lopez de Castro JA. Two
HLA-B14 subtypes (B*1402 and B*1403) differentially associated
with ankylosing spondylitis differ substantially in peptide specificity, but have limited peptide and T-cell epitope sharing with
HLA-B27. J Biol Chem 2005;280:35868–80.
33. Zemmour J, Little AM, Schendel DJ, Parham P. The HLA-A,B
“negative” mutant cell line C1R expresses a novel HLA-B35 allele,
which also has a point mutation in the translation initiation codon.
J Immunol 1992;148:1941–8.
34. Calvo V, Rojo S, Lopez D, Galocha B, Lopez de Castro JA.
Structure and diversity of HLA-B27-specific T cell epitopes:
analysis with site-directed mutants mimicking HLA-B27 subtype
polymorphism. J Immunol 1990;144:4038–45.
35. Shimizu Y, DeMars R. Production of human cells expressing
individual transferred HLA-A,-B,-C genes using an HLA-A,-B,-C
null human cell line. J Immunol 1989;142:3320–8.
36. Barnstable CJ, Bodmer WF, Brown G, Galfre G, Milstein C,
Williams AF, et al. Production of monoclonal antibodies to group
A erythrocytes, HLA and other human cell surface antigens: new
tools for genetic analysis. Cell 1978;14:9–20.
37. Stam NJ, Spits H, Ploegh HL. Monoclonal antibodies raised
against denatured HLA-B locus heavy chains permit biochemical
characterization of certain HLA-C locus products. J Immunol
38. Hochstenbach F, David V, Watkins S, Brenner MB. Endoplasmic
reticulum resident protein of 90 kilodaltons associates with the Tand B-cell antigen receptors and major histocompatibility complex
antigens during their assembly. Proc Natl Acad Sci U S A 1992;
39. Galocha B, Lopez de Castro JA. Folding of HLA–B27 subtypes is
determined by the global effect of polymorphic residues and shows
incomplete correspondence to ankylosing spondylitis. Arthritis
Rheum 2008;58:401–12.
40. Vazquez MN, Lopez de Castro JA. Similar cell surface expression
of ␤2-microglobulin–free heavy chains by HLA–B27 subtypes differentially associated with ankylosing spondylitis. Arthritis Rheum
41. Mear JP, Schreiber KL, Munz C, Zhu X, Stevanovic S, Rammensee HG, et al. Misfolding of HLA-B27 as a result of its B
pocket suggests a novel mechanism for its role in susceptibility to
spondyloarthropathies. J Immunol 1999;163:6665–70.
42. Dangoria NS, DeLay ML, Kingsbury DJ, Mear JP, UchanskaZiegler B, Ziegler A, et al. HLA-B27 misfolding is associated with
aberrant intermolecular disulfide bond formation (dimerization)
in the endoplasmic reticulum. J Biol Chem 2002;277:23459–68.
43. Goodall JC, Ellis L, Hill Gaston JS. Spondylarthritis-associated
and non–spondylarthritis-associated B27 subtypes differ in their
dependence upon tapasin for surface expression and their incorporation into the peptide loading complex. Arthritis Rheum
44. Peh CA, Burrows SR, Barnden M, Khanna R, Cresswell P, Moss
DJ, et al. HLA-B27-restricted antigen presentation in the absence
of tapasin reveals polymorphism in mechanisms of HLA class I
peptide loading. Immunity 1998;8:531–42.
45. Purcell AW, Gorman JJ, Garcia-Peydro M, Paradela A, Burrows
SR, Talbo GH, et al. Quantitative and qualitative influences of
tapasin on the class I peptide repertoire. J Immunol 2001;166:
46. Antoniou AN, Ford S, Taurog JD, Butcher GW, Powis SJ.
Formation of HLA-B27 homodimers and their relationship to
assembly kinetics. J Biol Chem 2004;279:8895–902.
47. Neisig A, Wubbolts R, Zang X, Melief C, Neefjes J. Allele-specific
differences in the interaction of MHC class I molecules with
transporters associated with antigen processing. J Immunol 1996;
48. Ortmann B, Copeman J, Lehner PJ, Sadasivan B, Herberg JA,
Grandea AG, et al. A critical role for tapasin in the assembly and
function of multimeric MHC class I-TAP complexes. Science
49. Van Endert PM, Riganelli D, Greco G, Fleischhauer K, Sidney J,
Sette A, et al. The peptide-binding motif for the human transporter associated with antigen processing. J Exp Med 1995;182:
50. Lopez de Castro JA, Alvarez I, Marcilla M, Paradela A, Ramos M,
Sesma L, et al. HLA-B27: a registry of constitutive peptide ligands.
Tissue Antigens 2004;63:424–45.
DOI 10.1002/art.24133
Clinical Images: Back pain, rash, and multiple spinal lesions
The patient, a 48-year-old woman, presented with a 6-week history of fatigue, maculopapular rash on the face, neck, anterior aspect
of both legs, right arm, and dorsum of both hands (A), polyarthralgia, and lower back pain. The erythrocyte sedimentation rate
(ESR) was 97 mm/hour, and serum concentrations of alkaline phosphatase, transaminases, and angiotensin-converting enzyme were
elevated. Findings on plain radiography of the spine were normal, but magnetic resonance image (MRI) showed numerous round,
well-circumscribed lesions within the medullary space throughout the spine. The lesions were isointense to muscle on T1-weighted
imaging (B), slightly hypointense on T2 imaging, and displayed mild contrast enhancement. The findings suggested possible
hematopoietic malignancy or disseminated infection. Bone marrow biopsy showed multiple noncaseating granulomas with negative
staining for microorganisms, consistent with sarcoidosis. Prednisone (40 mg/day) was prescribed, along with infliximab. Four weeks
later, the back pain and the rash had both resolved, and the ESR had normalized. Followup MRI showed significant improvement
in the spinal lesions.
Anthony Krajcer, MD
Inmaculada del Rincón, MD, MS
University of Texas Health Science Center
at San Antonio
Без категории
Размер файла
630 Кб
disparate, b2705, protein, ankylosis, spondylitisassociated, hlab, 24045, pdf1403, stability, folding
Пожаловаться на содержимое документа