PDF hosted at the Radboud Repository of the Radboud University
Nijmegen
The following full text is a publisher's version.
For additional information about this publication click this link.
http://hdl.handle.net/2066/14791
Please be advised that this information was generated on 2016-10-01 and may be subject to
change.
Clin. exp. Immunol. (1986) 66, 111-117
Altered antigen-presentation in the induction of the
in-vitro antigen-specific T helper cell function in patients
with chronic granulomatous disease
C O B I J. H E I J N E N , J. W . M . V A N D E R M E E R * & B. J. M . Z E G E R S Department o f
Immunology, University Hospital for Children and Youth ‘Het Wilhelmina Kinderziekenhuis’,
Utrecht, The Netherlands and * Department o f Infectious Disease, University Hospital, Leiden,
The Netherlands
( Accepted fo r publication 21 M ay 1986)
SUM M ARY
Phagocytic cells o f patients with chronic g r a n u lo m a t o u s disease ( C G D ) are severely
impaired in the killing process, on the basis o f defective oxygen metabolism. In this study
we investigated the antigen-presenting function o f monocytes (i.e. adh erent cells) o f C G D
patients. A dherent cells o f C G D patients were investigated for their capacity to present
ovalbum in (OA) in such a way that T helper cclls are activated and OA-specific IgMplaquc forming cells (P F C ) are generated. The results showed that the dose o f OA
required for optimal P F C responses was 20-25 times higher than the dose o f antigen
which is necessary for induction o f P F C s from peripheral blood m o n o n u c le a r cells o f
norm al donors. We could show that the C G D m onocytes were responsible for the
observed shift which indicates peculiar antigen-presenting capacities. The finding could be
mimicked in normal d o n o r s by treating adheren t cells with phenylbutazone, a d rug known
to interfere with oxygen-dependent degradation o f micro-organisms. M oreover, ou r
results suggested that C G D monocytes and phenylbutazone-treated monocytes o f normal
d o n o r s a b s o rb OA but do not re-express the processed antigen on the mem brane. We
conclude that o u r system used to study antigen presentation allowed the discovery o f
altered antigen presentation o f C G D monocytes m ost probably on the basis o f defective
antigen processing. However, antigen presentation is not entirely absent, since relatively
high doses o f O A induced norm al P F C responses in C G D m o n o n u c le a r cells.
Keywords
chronic g r a n u lo m a t o u s disease
antigen presentation
IN T R O D U C T IO N
An essential step in the activation o f T helper cells by complex antigens is the presentation o f the
antigen by accessory cells in conjunction with Class II molecules o f the M H C complex ( Rosental &
Shevach, 1973; Benaccrraf, 1981). T helper cells do not recognize antigens in their native form but
they only respond to ‘processed’ antigens presented on the surface o f the accessory cells ( Benacerraf,
1981; Ziegler & U n a n u e , 1981).
Processing o f antigen by accessory cclls involves a sequence o f events comprising recognition
and ingestion o f the antigen, and d eg rad atio n and re-expression o f the antigen on the m em b ran e
Correspondence: Dr Cobi J. Heijnen, Dept, o f Im munology, University Hospital for Children and Youth,
‘Het Wilhelmina Kinderziekenhuis', Nieuwe G r a c h t 137, 3512 LK Utrecht, The Netherlands.
III
1 12
Cobi J. Heijnen ,
./. W7. A /, ¿y///
cler Meer & B. J. M. Zegers
(Ziegler & U n a n u e , 1981; Ziegler & U n a n u e , 1982). However, in more recent years the necessity o f
processing o f antigen lor effective presentation to T helper cells has been disputed.
In previous studies we have reported on the regulatory role o f T cells and monocytes (i.e.
adherent cells) in the in-vitro activation o f h u m a n peripheral blood B lymphocytes with antigens
like ovalbum in (OA) and sheep red blood cells (Ballieux et a/., 1979; Heijnen et a /., 1981). In this
system antigen-presenting monocytes and antigen-specific T helper and T suppressor lymphocytes
interact in the activation o f antigen-specilic resting B cclls into the stage o f a small IgM-secreting
plaque forming cell (P FC ) (Ballieux & Heijnen, 1983; Van Tol et a /., 1984; 't H a r t et al. , 1985). In the
present study it was investigated whether peripheral blood monocytes from patients with chronic
g r a n u lo m a to u s disease ( C G D ) can present OA in such a way that T helper cells are activated and
OA-specific Ig M -P F C s are generated. In this way m onocytes defective in oxygen metabolism
(Gallin & Fauci, 1983) are used as a model to study antigen presentation.
M A T E R IA L S A N D M E T H O D S
Mononuclear cell isolation. Peripheral blood m o n o n u c le a r cells were isolated by density gradient
centrifugation as described by Heijnen et al. (1979).
Monocyte depletion and isolation. M onocytes were isolated from the m o n o n u c le a r ccll
suspension (1 x 10 (’ cells/ml) by allowing them to adhere to plastic Falcon flasks at 37 C in R P M I 1640 medium (Gibco) supplemented with 20°,, fetal calf serum (FCS). After 60 min o f incubation at
37 C the su p e rn a ta n t containing the lymphocytes was decanted and the ad h eren t cells were
harvested by gently scraping with a ru b b e r policeman and then collected in Earle's balanced salt
solution without calcium and magnesium (Gibco) supplemented with 0 002 m E D T A .
Cell cultures. This m ethod has been described extensively (Heijnen et al.. 1979). In brief, 5 x 106
peripheral blood lymphocytes (PBL) were cultured in the presence o f the antigen ovalbum in (OA;
G r a d e VI. Sigma, St Louis) a n d 10"o adherent cells. Cultures were performed in r o u n d b o t t o m tubes
(Falcon) in 10 ml medium which consisted o f RPM I-1640. supplemented with 2 m.vi L-gutamine,
antibiotics and 10"o h u m a n AB serum. Before use the serum was heat-inactivated and extensively
absorbed with sheep red blood cells (SE) to prevent pseu d o p laq u e form ation (Ballieux et al ., 1979).
Pulsing o f adherent cells with OA. M onolayers o f adheren t cells were incubated with 1000 //g
OA ml R PM I-1640 supplemented with 10"/.. heat-inactivated SE -absorbed AB serum for 2 h. After
the incubation period the cells were harvested and washed three times with R PM I-1640
supplemented with 5 % pre-treated AB serum.
Enzyme treatment o f the adherent cells. A dherent cclls ( l O x 1 0(> cells/ml RPM I-1640) were
incubated with pronase (8 mg ml. Sigma, St Louis) and D N A s e (10 //g/ml) for 30 min at 37 C. After
the incubation 1 ml FCS was added to the cells which were spun dow n (4 C) and washed twice with
RPM I-1640 supplemented with 10% FCS.
Phenylbutazone treatment. M o nolayers o f adherent cells were incubated with ph enylbutazone
(100 fig/ ml) for 30 min at 37 C. After this procedure the cells were washed twice.
Plaque forming ccll assay. P F C were determined as described in detail earlier (Ballieux et al ..
1979; Heijnen et al ., 1979). In brief. O A -coated SE (Heijnen et al ., 1979; G o d in g , 1969) were
centrifuged for 5 min at 1000 g on the b o tto m o f Falcon microtitre plates, coated by incubation with
poly-L-lysine (Sigma. St Louis, M r ^ 100,000) in a concentration o f 100 //g/ml distilled water. OAstimulated cells in various dilutions were incubated for 60-90 min in the presence o f SE-absorbed
guinea pig com plem ent (C). C ontrols comprised the addition o f cultured cells to the m onolayers
without C and vice versa. Only Ig M -P F C s are detected in this system and antigen-spccificity has
been investigated extensively as described earlier (Ballieux et a i , 1979; Heijnen et al ., 1979).
RESULTS
PFC-response of PBL from normal donors and o f CGD patients. PBL from norm al d o n o rs and
from four C G D patients were cultured with various doses o f the antigen ovalbum in (OA) for 6 days
Antigen presentation in chronic granulomatous disease
1 13
at 37 C. After the incubation period, the cells were harvested and tested for their capacity to secrete
IgM a n t i - 0 A antibodies in vitro by measuring the n u m b e r o f anti-O A P F C (Heijnen et al ., 1979).
It was shown that in PBL from norm al d o n o r s form ation o f P F C can be generated when the
dose o f antigen in the eulture is in between 1 and 3 pg O A /m l (Fig. 1). When norm al PBL are
stimulated with 100 pg O A or more, the PFC -response decreases to baseline levels. We have shown
previously that the decline in the PFC -response o f lymphocytes cultured in the presence o f the high
concentration o f antigen is due to the activation o f T suppressor effector cells (U y td e H a a g , Heijnen
& Ballieux, 1978). However, when PBL o f C G D patients are cultured with various doses o f antigen,
a shift o f the PFC -response tow ards the right side o f the curve was observed in all patients, with an
optimal PFC -response at 100 //g O A /m l (Fig. 1).
PFC-response o f lymphocytes o f normal donors and o f CGD patients after stimulation with
antigen-pulsed adherent cells. T o study the process o f antigen-presentation by accessory cells o f
C G D patients m ore precisely, ad herent cells o f both C G D patients and normal d o n o rs were pulsed
with ovalbum in (1000 pg O A /m l) for 2 h at 37 C. After the incubation period, the non-adhered
antigen was washed away and the pulsed cells were cultured for 6 days with PBL that were depleted
o f adherent cells under s ta n d a rd conditions. The results depicted in the upper part o f Fig. 2 show
/
(£>
o
0
o
Ll
Q_
CM
O
i
J
/
//
0
0-3
ù
1000
1 0 0
OA (¿ x g /m l)
Fig. 1. P F C response o f C G D patients (□, // = 4) and norm al control d o n o r s (□, // = 4), generated in the presence
o f various doses o f the antigen OA. Lymphocytes (5 x 10'’) (T/B ratio: a b o u t 7/1) were cultured with O A and
10"() ad here nt cclls. The T/B ratio in C G D patients an d n o rm al d o n o r s did not dilTer significantly. OA-specific
I g M - P F C s were measured after 6 days o f culture.
xIO2 P F C / I 0 6 ly
Adherent cells
5
~ T
10
T
~
15
“ I
OA pulsed
OA pulsed +
V
///i
pronase treatment
Fig. 2. P F C response o f PBL (5 x I06) from C G D patients (□,/? = 5) and o f normal control d o n o r s (□, n = 5) in
the presence o f 10% OA-pulsed adherent cells. * Assay system: PBL depleted o f ad h e re n t cells and reconstituted
with the indicated ad here nt cell suspensions. Background values (PBL cultured with 10% non-pulsed adherent
cells) are less than 200 PFC.
1 14
Cobi J. Heijnen, J. W. M . van der M e e r & B. J. M . Zegers
that C G D lymphocytes cultured with antigen-pulsed adherent cclls m o u n t a PFC -response which
indicates that C G D adheren t cclls are able to present antigen. However, the P F C response obtained
is much lower than th at o f cultures o f antigen-pulsed adherent cclls and lymphocytes o f control
donors.
PFC-response o f normal donors and o f CGD patients after stimulation with antigen-pulsed,
pronase-treated adherent cells. Next, it was investigated w hether the PFC -response observed in
cultures o f antigen-pulsed monocytes and lymphocytes o f C G D patients is actually the result o f
antigen-processing and re-expression o f the antigen on the m e m b ra n e o f the ad h eren t cclls.
T o that end. antigen-pulsed adherent cells were ‘stripped' by pro nase-treatm ent to remove cellb o u n d antigen that had not yet been ingested. After several washing procedures the ad h eren t cclls
were added to the lymphocyte suspension that was depleted o f adherent cells and cultured for 6
days.
These experiments show that when the adheren t cells o f C G D patients are clcarcd from cellbo und antigen, the antigen-pulsed adh eren t ccll population is incapable o f inducing a P F C response (lower part o f Fig. 2). This finding suggests that surface-bound O A is responsible for the
low P F C response obtained when antigen-pulsed adherent cells, not treated with pronase, were used
for induction o f PFCs. Unlike C G D adherent cells, antigen-pulsed pronase-treated ad herent cells
from norm al d o n o rs can activate the lymphocytes, leading to the differentiation o f resting B cells
into IgM-secreting PF C s (Fig. 2).
PFC-response o f PBL fro m normal donors in the presence o f adherent cells treated with
phenylbutazone. In the next scries o f experiments, we tried to mimic the delect o f m onocytes o f C G D
patients, by pre-incubating normal ad h eren t cells with p hen ylbu tazo ne (100 //g/ml). This agent is
kno w n to inhibit intracellular killing o f phagocytes, prob ably by interference with the hexose
m o n o p h o s p h a te shunt (Solbey, 1975). When norm al adherent cclls arc pre-treated with p h en y lb u ­
tazone for 30 min and cultured with PBL and various doses o f OA for 6 days, a shift tow ards the
right of the PFC-response curve is observed, resembling the PF C -antigen dose relationship o f C G D
patients (Figs 1 and 3).
PFC-response o f PBL o f normal donors in the presence o f phenylbutazone-treated, antigen-pulsed
and pronase-treated adherent cells. N o rm a l adheren t cells, pre-treated either with m edium o r with
phenylbutazone, were pulsed with antigen for 2 h. After the pulse-period, one part o f the cells was
subsequently treated with pronase to remove cell-bound antigen. After several washing procedures
the adherent cell preparations were added separately to lymphocyte suspensions which were
carefully depleted o f adherent cells, and cultured for 6 days at 37 C. Figure 4 dem o nstrates that
77
yy
/
J
5
>
,
z
CD
o
o
IL
/
/
0
CL
C\J
o
v/
/
y
Z
'A
0
0-3
30
100
1000
0 A (¿¿.g/ml)
Fig. 3. P F C response o f PBL from normal d o n o r s with either phenylb utazone treated adherent cells (□, n = 4) or
untreated cells (□, // = 4) in the prcscnce o f various doses o f OA.
Anligen presentation in chronic granulom atous disease
115
x IO2 P F C /IO 6 Ly
Adherent cells
5
T
10
15
1
OA pulsed
Phenylbutazone +
OA pulsed
Phenylbutazone +
OA pulsed + pronase
treatment
OA pulsed +
pronase treatment
Fig. 4. P F C response o f P B L from norm al d o n o r s in the presence o f ad here nt cells treated as indicated
indeed ph enylbutazone treatm ent mimics the results obtained with the adherent cells o f C G D
patients (Fig. 2), i.e. P F C are generated in cultures with phenylbutazone-treated adherent cells but
just as in C G D patients the results are lower than with untreated adherent cells. When
phenylbutazone-treated, antigen-pulsed adherent cells are cleared o f cell-bound antigen, the
ad herent cells are no longer capable o f presenting the antigen in such a way that it leads to P F C
formation.
D ISC U SSIO N
The question w hether the antigen has to be processed by an accessory ccll and re-expressed on the
m e m b r a n e before it can activate T helper cells is a subject o f intense investigation ( U n a n u e et al .,
1984). In contrast to T helper cells. T suppressor (Ts) cells recognizc the antigen in its soluble form,
since T s cells can be stimulated efficiently in the absence o f m onocytes with low doses o f antigen
(0-003 f.ig OA/ml; U y t d e H a a g et al ., 1982). However, when the n u m b e r o f adherent cells in the
culture is increased, m ore antigen is required (30-100 /.ig O A /m l at 10% adheren t cells) to activate T
suppressor cells ( U y t d e H a a g el al., 1982). The bell-shaped P F C -respon se curve is therefore a
reflection o f the net result o f the balance between T helper and T suppressor cell activities induced by
the various doses o f antigen (Ballieux et al ., 1979).
W hen m o n o n u c le a r cclls from C G D patients are cultured with various doses o f OA, the P F C response curve has an o p tim u m at 100 //g O A /m l, indicating that unlike normal d o n o rs relatively
high doses o f antigen are required to let Th cell activity prevail over T s activity. In separate
experiments it was shown that C G D m o n o n u c le a r cells produced neither OA-specihc T helper cell
factors nor OA-specific T suppressor ccll factors (Ballieux et al ., 1979; Heijnen et al ., 1981) at a
concentration o f 3 ¡ig O A /m l (data not shown). The latter finding indicates that with this dose o f OA
no free antigen is available to activate T s cells which suggests that OA is internalized in the C G D
monocytes without induction o f T helper function. When subsequently adherent cells are pulsed
with antigen, extensively washed and treated with pronase to remove non-ingested cell-bound
antigen, the PFC -response is abolished (Fig. 2). We therefore conclude that the ad h eren t cells o f
C G D patients apparently do not process the antigen in a way it is rc-exprcssed on the m e m b ra n e for
the activation o f T helper cells. F ro m the d a ta presented it appeared that adherent cells o f C G D
patients that have been pulsed with O A but not stripped for non-ingested O A can present antigen to
T helper cells, although the P F C - responses are much lower than those o f normal d o n o rs (Fig. 2).
These results imply that there are T cells present in the peripheral circulation that can recognize
antigen that has not been processed by monocytes. We do not yet know w hether the antigen is
116
Cobi J. Heijnen, J. W. M . van dev M eev c£ B. J. M . Zegevs
partially processed at the level o f the m e m b ra n e o f the adheren t cells or w hether enzymes present in
the AB serum pool can partially degrade the antigen. It might also be possible that a specialized
subset o f antigen-presenting cells, e.g. dendritic cclls which are kno w n to be capable o f stimulating T
helper cclls without ingesting and processing the antigen (Van Voorhis et al., 1982). is responsible
for the presentation o f the antigen to the T helper cclls.
The present result, i.e. a shift o f the O A -dosc response tow ards high concentrations in
m o n o n u c le a r cells o f C G D patients, is reminiscent o f earlier observations using cord blood
m o n o n u c le a r cells. In this respect van Toi et al. ( 1984) showed that neonatal and adult adh erent cells
differ in antigen-handling which is rcflcctcd in a shift o f the OA dose needed to induce an optimal
P F C response in cord blood cells to low (i.e. 0 03 //g) concentrations. In addition it was shown that
neonatal adherent cells sccrctc m ore P G E 2 than adh erent cclls from adult controls. The difference
between the adult and cord blood dose response curves could be abolished by treatm ent with
indomethacine, an inhibitor o f cyclooxyeenase. These results suggest a relation to exist between
antigen-handling and P G E 2 secretion a n d / o r sensitivity. We did not measure P G E 2 secretion by
C G D adherent cells but, unlike o u r results, one would expect that if P G E 2 secretion by C G D
adherent cclls is increased, a shift o f the antigen-dose/'PFC relationship tow ards low co ncentrations
would be obtained. It should be mentioned that the C G D patients had no overt infections when
investigated. F u rth e rm o r e several patients with infections were investigated for P F C responses in
the last years in o u r laboratory. However a shift in the antigcn-dose response relationship tow ards
high antigen concentrations was never observed.
In order to d e m o n stra te m ore directly that the a b b c r a n t PFC-responses are due to a defccl in the
antigen-presenting cells, we tried to mimic the defect o f m onocytes o f C G D patients by treating
adherent cells from norm al d o n o rs with phenylbutazone. The latter drug interferes with oxygendependent deg radatio n by inhibiting H :0 : p ro d u c tio n through an effect on the hexose m o n o p h o s ­
phate shunt pathw ay (Soleby, 1975). The P F C dose response using phenylbutazone-treated
adherent cells o f normal d o n o rs completely mimics those o f C G D patients. F u rth e rm o re , results
using antigen-pulsed adherent cclls pretreated with p henylbutazone and stripped with pronase
show unequivocally that the adherent cclls are no longer capable o f presenting the antigen in a way
that it can activate T helper cells. These results indicate also that oxygen-dependent biochemical
pathw ays play an im p o rta n t role in antigen-processing and presentation o f OA. The im p o rtan c e o f
lysosomal degradation is suggested by d a ta from Allen & U n a n u e (1984), who showed that agents
like chloroquine inhibit the subsequent activation o f T helper cells.
T ak in g these data together we can conclude that un d er the particular circumstances o f o u r invitro system adherent cells o f C G D patients can internalize the antigen, but are defective in the
presentation o f the antigen in a processed form to h u m a n peripheral blood T helper cells.
Consequently a dcfect in oxygen metabolism may interfere at some stage with antigen-processing o f
OA, possibly since oxidants needed to d en atu re sulfhydryl-containing proteins like OA before
processing are lacking (Weiss, Lam pert & Test. 1983).
m
y
Km*
This study was supported in part by the F o u n d a ti o n for Medical Research (F ung o, grant No. 13-4091). which is subsidized by the N etherlands O rganisation for the A d v an cem en t o f Pure Research
(Z.W.O.) The a u th o r s arc indebted to D r J. C. K luin-Nclcmans for the clinical assessment o f one o f
the patients studied and to Dr H. C. W. van Velzen-Blad for the co-operation shown in the course o f
this study. They also thank Mrs Hélène van Moorsel for her secretarial assistance.
REFERENCES
P.M. & U n a n u e , E.R. (1984) Differential
requirements for antigen-processing by m a c r o ­
phages for lysosome-specific T cell hybridomas. J.
Immunol. 132, 1077.
B a l l i e u x , R.E., H e i j n e n , C.J., U y t d e h a a g F. &
Z e g e r s , B.J.M. (1979) Regulation o f B cell activity
in man: role o f T cells. Immunol. Rev. 45, 3.
A llen,
R.E. & H e i j n e n , C.J. (1983) Im m u n o re g u latory T cell subp o p u latio n s in man: dissection by
monoclonal antibodies and Fc-receptors. Immunol.
Rev. 74, 5.
B e n a c e r r a p , B. (1981) Role o f M H C gene products in
immune regulation. Science , 212, 1229.
G a l l i n , J.I. & F a u c i , A . S. (1983) C h ro n ic g r a n u l o ­
Ballieux,
Anligen presentation in chronic granulom atous disease
m a t o u s disease. In: Advances in Host Defence
Mechanisms Vol. 3 (ed by .1.1. Gallin & A. S. Fauci),
p. 1. Raven Press. New York.
C o d i n g , J.W. (1969) Tlic chrom ic chloride m eth o d o f
co u p lin g antigens to erythrocytes: definition o f
so m e important parameters. ./. immunol. Methods.
10 , 61.
L.A., H e i j n e n , C.J., Z i j l s t r a , J. & B a l ­
l i e u x , R.E. (1985) Analysis o f the antigen-induced
in vitro differentiation o f h u m a n peripheral blood B
cells: stimulation with ovalbum in induces the t r a n ­
sition from the resting state into the excitecf state.
Clin. exp. Immunol. 61, 648.
H e i j n e n , C.J., U y t d e h a a g , F., G m e l i g M e y l i n g ,
F.H.J. & B a l l i e u x , R.E. (1979) Localization o f
h u m a n antigen-specific helper and suppressor func­
tion in distinct T cells subpo pulations. Cell. Immu­
nol. 43, 282.
H e i j n e n , C.J., U y t d e h a a g , F., P o t . K . H . & B a l ­
l i e u x , R.E. (1981) Antigen-specific h u m a n T cell
factors. I. Tcell helper factor: biologic p r o p e r t i e s . ./.
Immunol. 126, 497.
R o s e n t a l , A.S. & S h e v a c h , E.M. (1973) F u n ctio n o f
m acro p h ag es in antigen recognition by guinea pig T
lymphocytes. I. R equirement for histocompatiblc
m acro p h ag es and lymphocytes. ./. exp. Med. 138,
1194.
S o l b e y , C.O. (1975) Influence o f therapeutic c o n c e n ­
trations o f phenylbutazone on granulocyte func­
tion. Acta Path. Microbiol. Scand. Sect. B 83, 100.
van T o l,
Z i j l s t r a , J., T h o m a s , C .M . G . .
Z e g e r s , B.J.M. & B a l l i e u x , R.E. (1984) Distinct
't H a r t ,
117
role o f neonatal and adult monocytes in the regula­
tion o f the in vitro antigen-induced plaque forming
cell response in man. J. Immunol. 134, 1902.
U n a n u e , E.R., B e l l e r , D.I., Lu, C.Y. & A l l e n , P.M.
(1984) Antigen presentation: c o m m e n ts on its regu­
lation and mechanism. J. Immunol. 132, 1.
U y t d e h a a g , F., H e i j n e n , C.J. & B a l l i e u x , R.E.
(1978) Induction o f antigen-specific h u m a n s u p ­
pressor T lymphocytes in vitro. Nature 271, 556.
U y t d e h a a g , F., H e i j n e n , C.J., P o t , K . H . & B a l ­
l i e u x , R.E. (1982) Antigen-specific h u m a n T cell
factors. V. Interference o f adherent cells in the
generation o f T cell suppressor factor by T cells. In:
Human B Cell Function. Activation and Immunoregulation (ed. by A. S . Fauci & R. E. Ballieux). p.
253. Raven Press, New York.
V a n V o o r h i s , W.S., H a i r , L.S., S t e i n m a n , R.M. &
K a p l a n , G . (1982) H u m a n dendritic cells: enrich­
ment and characterization from peripheral blood.
J. exp. Med. 155, I 172.
W e i s s , S.J., L a m p e r t , M.B. & T e s t , S.T. (1983) Longlived oxidants generated by h u m a n neutrophils:
characterization and bioactivity. Science 222, 626.
Z i e g l e r . K . & U n a n u e , E.R. (1981) Identification of
a m a c ro p h a g e antigen-processing event required
for I-region-restricted antigen presentation to T
lymphocytes. J. Immunol. 127, 1869.
Z i e g l e r , K . & U n a n u e , E.R. (1982) Decrease in
m a c ro p h a g e antigen catabolism by a m m o n i a and
chloroquine is associated with inhibition o f antigen-presentation to T cells. Proc. Natn. Acad. Sci.
USA. 79, 175.