Download identification of a macrophage antigen

This information is current as
of August 3, 2017.
Pillars Article: Identification of a Macrophage
Antigen-Processing Event Required for IRegion-Restricted Antigen Presentation to T
Lymphocytes. J. Immunol. 1981. 127: 1869−
1875.
Kirk Ziegler and Emil R. Unanue
J Immunol 2007; 179:5-11; ;
http://www.jimmunol.org/content/179/1/5.citation
Permissions
Email Alerts
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2007 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on August 3, 2017
Subscription
0022-1767/81/1275-1869$02.00/0
THE JOURNAL OF hMUNOLOGY
Copyright @ 1981 by The American Assoclation of ImmunOlOgiStS
VOI.127, NO 5. November 1981
Pnnted In U.S A
IDENTIFICATION OF A MACROPHAGE ANTIGEN-PROCESSING EVENT REQUIRED FOR
REGION-RESTRICTED ANTIGEN PRESENTATION TO T LYMPHOCYTES'
KIRK ZIEGLER
AND
I-
EMlL R. UNANUE
From the Departmentof Pathology, Harvard Medical School, Boston,MA 02 1 7 5
phosphate-buffered saline (PES) containing 1 mCiof carrier-free Na'251
(New England Nuclear, Boston, MA) and 50 pg of chloramine-T (Eastman
Kodak, Rochester, NY) were incubated for 1 0 min at 4°C. Sodium metabiMacrophages are intimately involved in the antigen-specific sulfite (Sigma Chemical Company, St. Louis, MO) (20 pI of a 5 mg/ml
activation of L y l +2-3- T lymphocytes (reviewed in 1-3). The solution) was added to terminate the reaction, and then the labeled bacteria
initial event in T cell activation is thought to be the recognition were washed 5 times in PBS to remove free '251, Bacteria thus treated
of the immunogen and I region gene products present on the showed 1 to 2 counts per minute (cpm) per bacterium. More than 99% of
the radioactivity was precipitable in1090trichloroacetic acid(TCA). Listeria
macrophage cell surface (3, 4). In the present study, we monocytogenes labeled with lZ5l
('251-Listeria)were stored at 4°C in PBS
attempt to correlate the uptake, ingestion, and catabolism of for use over several weeks and were washed immediately before use.
In 1 experiment, Listeria monocytogenes were attached to culture dishes
antigen by the macrophages to their ability to present immuusing poly-L-lysine. Tissue culture wells (1 6 mm diameter) were treated (1
nogeneic molecules to T cells. To this effect, we have used the hr, 20°C) with an aqueous solution of poly-L-lysine (1 mg/ml) and then
binding of specific T cells to macrophages as a functional washed. Heat-killed Listeria inPES (0.5 ml of IO' bacteria/ml) was added,
assay of macrophage-associated antigens. It is known that the the plate was centrifuged (800 X G, 5 min), and then incubated for 1 hr at
interactions between T cells and macrophages bearing antigen 20°C. After washing with PES. a confluentlawn of bacteria remained firmly
attached to the dish.
may be analyzed directly by quantitating the antigen-specific
Uptake and ingestion of Listeria by macrophages. Iodinated Listeria
physical interactions taking place between both cells (5-1 3). monocytogenes was added to 3 X l o 5 to 1O6 peritoneal exudate cells
This quantitation of macrophage-associated antigenicity within (PEC)' planted in 16-mm diameter tissue culture wells (Falcon No. 3008).
The plates were centrifuged (800 x G, 5 min) and then incubated as
a short temporal framework is a unique way to study the required in the differentexperiments. The macrophages were then washed
mechanisms of macrophage handling of antigen in the context thoroughly, and Triton X-I00 to a 1 .O% concentration was added. Radioof I region gene effects. Previous studies have approached this active counts were made on the solubilized macrophage extract.
The ingestion of Listeria was followed microscopically.Heat-killed Listerla
(0.3 ml of 5 X 106/ml) were added to macrophage monolayers formed on
15-mm diameter coverslips resting ontissue culture wells. The plates were
Received for publication April 29, 1981.
centrifuged (800x G.5 min) and then washed to remove unbound bacteria.
Accepted for publication July 28, 1981.
The macrophages were then incubated for various periods of time (0to 120
The Costs of publication of this article were defrayed in part by the payment
of page charges. Thisarticle must therefore be hereby marked advertisement in min) at 37°C before fixation with 190 paraformaldehyde. The Listeria monaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.
I This work was supported by National Institutes of Health GrantsCA 14723,
Abbreviations usedin this paper: HEPES,N-2-hydroxyethylp1perazine-N"2AI 14732, and A I 17259.
ethanesulfonic acid; PEC. perltoneal exudate cells.
1869
Downloaded from http://www.jimmunol.org/ by guest on August 3, 2017
problem by using radioactively labeled antigen molecules to
The mechanism of macrophage-antigen handling was
follow
their fate in phagocytic cells, and long-term functional
studied using a system that involves the quantitation of
assays for the immunogenicity of macrophage-associated anthe antigen-specific binding of Listeria monocytogenesimmune T cells to macrophages. Specific T cells did not tigen (1 4-1 6). The phagosome-lysosome pathway of antigen
bind to native antigen. Because
the specific binding of T uptake and degradation and the potent immunogenicity of
cells to macrophages could be measured during a short macrophage-associated antigen have been well illustrated.
(5- to 18min) interaction, it was possible to follow the However, because of the dynamic nature of macrophage-antitemporal development of a T cell-binding substrate with gen handling events and the length of time required to funcincreasing time ofantigen-macrophage
interaction.In
tionally assess antigenicity, precise cause-and-effect relationcontrast to the rapid (5-min) uptake of Listeria by macships between antigen-handling events and immune recognirophages, the developmentofTcell-bindingabilityretion could not be determined with certainty.
quireda 3 0 to 60min period ofantigen-macrophage
Wehave employed as antigen the intracellular pathogenic
interaction. During this processing period, Listeria orgabacteria Listeria monocytogenes. The activation of Listerianisms bound to the macrophage surface were ingested
immune T cells requires macrophages in a process regulated
and partially catabolized. Unlike antigen
uptake, antigen
by the I region of the H-2 (1 1 , 12, 17). Moreover, Listeria can
processingwasatemperature-dependentandenergyrequiring
event.
Although
macrophages
treated with be traced in the macrophage and its uptake, ingestion, and
paraformaldehyde before antigen processing
did not de- catabolism monitored reasonably well. We now define a macvelop T cell-binding activity, macrophages treated with rophage processing event required for the T cell recognition of
paraformaldehydeafter a W m i n antigen-processing pe- Listeria antigen.
riod retained T cell-binding ability. The kinetics
of antigen
MATERIALS AND METHODS
catabolism correlated with antigen processing, and
inhiMice. A/St mice were purchased from West Seneca Laboratories, Bufbition of antigen catabolism was associated with a corfalo, NY. Male or female mice at 8 to 12 wk of age were employed.
respondinginhibitionofantigenprocessingforTcell
Listeria monocytogenes. The preparation of Listeria monocytogenes was
binding. Anti-la antibodies had no effect on Listeria up- previously
described (1 7). Live bacteria were used to immunize mice, and
take or catabolism. These results supply direct evidence heat-killed organisms served as antigen for use in vitro.
for a macrophage-antigen processing event relevant to T Heat-killed Listeria monocytogenes were labeled with '251 by the chloramine-T method (1 8). Briefly, 10' washed bacteria suspended in 130 pl of
cell recognitionof antigen.
1870
KIRK ZIEGLER AND EMlL
% specific binding =
x- Y
X 100
X
Anti-Listeria T cells were measured by 3 assays described in full detail
in previous studies (1 1, 12). These were: 1) the production of thymocyte
mitogenic protein (17);2) the development of tumoricidal activity by macrophages; and 3) T cell proliferation. All 3 assays require an interaction
between immune T cells and macrophages modulated by the I region ofthe
H-2 (1 7, 19). In the assay for thymocyte mitogenic protein, T cells and
macrophages were co-cultured for 24 hr, and then culture medium was
tested for mitogenic activity on A/St thymocytes. In the cytocidal assays,
after 24 hr of cultureof macrophages and T cells, the T cells were removed,
and "Cr-labeled P-815 mastocytoma cells were added to the macrophages
for a period of 18 hr. Percent specific cytolysis was determined from the
release of 5'Cr by the tumor cells (11, 19).T cell proliferation wasassayed
after 4 days of culture of T cells with macrophages.
Antibody reagents. The monoclonal antibody anti-LAh represents the
culture supernatant from a hybridoma cell line (clone 10-2.16) originating
inthelaboratory of Dr. L. A. Herzenberg (20). A.TH-anti-A.TL(anti-la')
antiserum was purchased from Cedarlane Laboratories (Hornby, Ontario,
Canada). The strength and specificity of these reagents were determined
by standard serologic means employing complement-mediated lysis of
appropriate spleen target cells.
UNANUE
[VOL. 127
immunofluorescence. The assessment of macrophage cell surface la
molecules was performed by indirect immunofluorescence with the monoclonal anti-LAh (21 ).
RESULTS
The binding assay. The antigen specificity and macrophage
dependence of antigen recognition by Listeria-immune T cells
is illustrated by the experiment of Table 1. Listeria monocytogenes-immune T cells were added to tissue culture dishes
containing the various binding substrates. Binding was monitored by testing the nonadherent T cells in the macrophage
cytotoxicity assay. Although complete depletion of Listeriaspecific T cell activity was observed with macrophages incubated with Listeria, no binding was observed with macrophages
incubated with the antigenically unrelated bacteria, Safmoneila
typhi. Furthermore, Listeria-immune T cellsdid not bind to
Listeria monocytogenes attached to the dish in the absence of
macrophages. The inability of immobilized Listeria monocytogenes to support binding was also observed when the mitogenic protein and proliferation assays were used to assess T
cell activity (data not shown). Clearly, both antigen and macrophages are required to generate a substrate to which T cells
will bind.
Kinetics of antigen handling. Several aspects of the handling
of Listeria by macrophages were studied. Macrophages from
Listeria monocytogenes-immune mice containing a high proportion of la-positive cells were used (21). Such macrophages
have been shown to bind Listeria-immune T cells much better
than macrophages from peptone-induced PEC (unpublished
observations). The uptake, ingestion, and catabolism of Listeria
by macrophages as function of time are shown in Figure 1.
Substantial uptake of Listeria occurred within 5 to 10 min by
using a centrifugation step (800 X G, 5 m i d to initiate the
interactions of bacteria with macrophage monolayers (Fig. 1a).
Routinely, 15 to 30% of the added '251-Listeria became macrophage associated during this time period. Rapid (5-minute)
uptake occurred equally well in the presence of both sodium
azide (10" M) and 2-deoxyglucose (10" MI, and also at 4OC
(data not shown).
After the rapid binding (5 min) of Listeria to the macrophage
cell surface, ingestion was evidenced by the progressive decrease in the number of macrophage surface-associated bacteria, reaching a 50% reduction by about 10 min (Fig. 16).
In Figure I C , macrophages were exposed (30 min 37OC) to
'251-Listeria to allow for uptake and ingestion, then washed
thoroughly; the radioactivity in the macrophage and the superTABLE I
AnOgen recognition by Listerla-immune T cells: specifioty and macrophage
dependence"
Activity of Nonadherent
Binding Substrate
T Cells: MacrophageMediated Cytolysis (%
i SEM)
?4 Spectfic
Binding
~
Macrophages
Macrophages and Satmoneffa
Macrophages andListeria
39.5 f 3.9
43.9 f 1.9
-5.4 k 1.6
Culture dish
Salmonella
Listeria
42.7 k 0.9
47.3 i 0.6
50.3 k 5.7
"7 1
100
-1 1
-18
Macrophage monolayers (1.5 X lo6 PEC per well) were incubated (1 hr.
37°C) with heat-killed Salmonella typhi or Listeria monocytogenes(0.3 pl of IO'
bacteria per ml). Some culture dishes contained bacteria attached to the culture
surface. Listeria-immune T cells (5 X lo6 T cells per well) were added to the
dishes containing the indicated binding substrate: the plates were centrifuged
(50 x G. 5 min. 20°C) andthenincubatedat
37OC for 1 hr. The T cells
for Specific
ionadher& to tile binding substrates were recovered and tested
activity in the macrophagecytotoxicity assay. The activityof lo5cells per well is
zk SEMof
duplicatebinding
givenasthemeanpercentspecificcytolysis
reactions.
Downloaded from http://www.jimmunol.org/ by guest on August 3, 2017
ocytogenes organisms on the surface of the macrophages remained reactive with antidistena antibody and could bevisualized by indirect immunofluorescence. Inverted coverslips were incubated (30min, 4°C) on 20 pl of
a 1:IO dilution of rabbit anti-Listeria serum, washed, and then treated with
fluorescein-conjugated F(ab% goatanti-rabbit globulin (20 pl of 50 pg/ml).
Catabolism of '%Lisferia b y macrophages. Macrophage monolayers
were cultured with '251-Listeriato allow uptake (as described above) and
then washed thoroughly. After incubation for various periods of time, the
supernatant was removed, and Triton X-1 00 (1 % v/v in H20)was added to
the macrophages. The solubilized macrophages were recovered from the
dish with the aid of a rubber policeman. Both the macrophage andsupernatant fractionswere incubated with anequal volume of 20% TCA. and the
precipitate was separated by centrifugation (1000 x G. 30 min, 4°C).
Radioactivity was measured in the precipitate and in the soluble supernatant.
7 c e h andmacrophages. The preparation of cell populationswas exactly
as described previously (1 1, 17, 19).Briefly. T cells were purified from the
peritoneal exudates of mice infected i.p. with 1 to 5 x lo4 live Listeria
organisms. After 1 wk, the mice received an i.p. injection of 10% proteose
peptone and then were sacrificed 3 days later. T cell enrichment, accomplished by the removal of cells adherent to tissue culture dishes and nylon
wool, routinely resulted in a more than 95% Thyl.2-positive lymphocyte
population containing lessthan 0.5% macrophages.
Normal and Listeria monocytogenes-immune mice (as described above
for T cells) were used as macrophage sources. In both cases, PEC were
harvested 3 days after an i.p. injection of proteose peptone(1 0% w/v). The
PEC were incubated (2 hr at 37°C) in tissue culture vessels to allow
macrophage adherence, and then the nonadherent cells were removed by
washing.
Media. Peritoneal lavage was performed wtth Hanks' balanced salt solution containing 2% fetal calf serum, 10 mM HEPES' buffer, and heparin
( I O U/ml). For cell culture, RPMl 1640 containing 5% fetal calf serum, 10
mM HEPES buffer, 2 mM L-glutamine, penicillin (50U/ml), and streptomycin
(50 pg/ml) was used. These reagents were purchased from Grand Island
Biological Company (Grand Island, NY).
Quantitation of J ce/l-macrophage binding. This was done as described
in our previous study (1 1). PEC (1.5x 1 O6 per well) trom Listeria-immune
mice were incubated (2 hr. 37°C) in 16-mm diameter tissue culture wells
(Falcon No. 3008, Multiwell tissue culture plate) to allow adherence of
macrophages, and then the nonadherent cells were removed by washing.
The resulting confluent monolayer of macrophages was used as the binding
substrate. Heat-killed Listeria monocytogenes (0.3ml of lo5to 10' bacteria/ml) was added, and the plates were centrifuged (800 X G, 5 min) and
then incubated as indicated inthe various experiments. Macrophages were
washed extensively to remove unboundbacteria. Control macrophages
were incubated in parallel with medium alone. Listeria-immune T cells were
added to the macrophage monolayers (0.5 ml containing 0.3 to 1 X lo6
cells/well), the dishes centrifuged (50 x G, 5 min, 20°C), and then
incubated for 5 to 60 min at 37°C. After incubation, the plates were gently
agitated, and the cells nonadherent to the macrophages were collected.
These were washed, counted (adlusted to equivalent cell densities), and
then their specific T cell function was measured on a new set of macrophage
monolayers. The decrease in the specific functional activity of T cells
recovered from the macrophage
monolayers treated with Listeria relative to
the activity of T cells from control macrophages was taken as a measure of
specific T cell-macrophage binding. Percent specific binding was calculated
as follows: Let X equal the activity of T cells nonadherent to macrophage
monolayers without antigen. Let Y equal the activity or T cells nonadherent
to Listeria-treated macrophagemonolayers. Then:
R.
19811
HANDLING
1871
BY MACROPHAGES
ANTIGEN
of Listeria-macrophage interaction after simple antigen uptake
is required for the generation of a T cell binding substrate.
Increasing the amount of Listeria 10-fold did not overcome this
requirement. This operationally defined time period of Listeriamacrophage interaction will be termed processing.
To more carefully analyze the temporal requirements for
processing, several experiments were performed by varying
the time of Listeria-macrophage interaction. In the experiment
shown in Figure 3, both the time period macrophages were
exposed to Listeria before addition of T cells (processing time)
MINUTES
HOURS
and the times of T cell-macrophage interaction (binding time)
Figure 1 . a. The uptake of '251-labeledListeria monocytogenes by macrowere varied. Little or no specific binding was observed when
phages is shown as a function of time. "%Listeria (0.5 ml per well, l o 5cpm per
the combined binding time and processing time was less than
well) was added to tissue culture wells containing macrophage monolayers
formed with 1O6 PEC. The plates were centrifuged (800X G, 5 min), incubated
30 min. It appeared that the presence of T cells for various
for the indicated periods of time at 37% and then washed to remove unbound
lengths of time had no substantial influence on the temporal
bacteria. The macrophage-associated radioactivity recovered after treatment
with 1% Triton X-1 00 is shown (closed symbols). Open symbol represents '*% progression in thedevelopmentof
T cell-binding substrate,
Listeria associated with culture well in the absence of macrophages. b. The
since the magnitude of specific binding was directly related to
READOUT SYSTEM
'*%
natant was followed over time in culture at 37°C. The catabolism of '251-labeledListeria by macrophages was evidenced by
the decrease in TCA-precipitable radioactivity associated with
0 40
0 40 80
0 40 EO
the macrophage, concomitant with the release of TCA-soluble
X SPECIFICBINDJNG
lZ5l
released into the supernatant. Less than 10% of total lZ5l Figure 2. Macrophage monolayers (MAC) were exposed (5 min, 800 x G,
remained macrophage associated as TCA soluble material for 20°C) to heat-killed Listeria monocytogenes (LM) (0.3 ml of 1O6 or 1O7 bacteria
ml), washed, and then incubated for 0 or 60 min at 37°C before addition of
was secreted in a TCA- Tpercells
several hours; less than 5% of total lZ5l
(5 x 1O5 per well in 0.5 ml). T cell-macrophage interaction initiated by a
precipitable form. In 10 experiments, after 1-hr culture at 37"C, brief centrifugation (50 X G, 5 min) was conducted for a total of 15 min at 37°C.
Percent specific binding was monitored by the readout systems as indicated and
34.8 f 2.9% (mean f standard error of the mean [SEMI) of
percent specific binding calculated as described in Materials and Methods. The
the total radioactivity taken up by the macrophages was re- mean f SEM for duplicate binding reactions is shown. Control values for the
leased in the mediumas small m.w. material. If the rate of activity of T cells exposed to macrophage monolayers in the absence of Listeria
catabolism was measured beginning immediately after the were 12,401 f 410 Acpm for mitogen protein and 7.853 f cpm for proliferation.
binding of '251-Listeriato the macrophage cell surface (a 5-min
exposure instead of a 30-min exposure), catabolism was deloo Bindtng Time
layed by 15 min, presumably reflecting the time required for
5min
A 15 min
ingestion and lysosomal interaction. Thus, the catabolism of
30 min
Listeria by macrophages was an extremely rapid event, digestion being detectable within a few minutes and proceeding over
several hours in culture. Catabolism did not occur at 4OC or
after treatment of macrophages with paraformaldehyde (1 YO,5
min). About 50% inhibition of catabolism was observed when
measured in the presence of both sodium azide (1O-' M) and
2-deoxyglucose (10" M) (data not shown).
Kinetics of antigen processing. Having analyzed the kinetics
of uptake, ingestion, and catabolism of Listeria by macrophages, we investigated the time of Listeria-macrophage interaction requiked to generate a substrate for T cell binding. In
Figure 2, macrophage monolayers were exposed to Listeria for
5 min to allow uptake and then washed. One set was immediately tested for their ability to bind specific T cells, whereas the
PUOCESShVG TIME fminl
other was incubated for 60 min at 37°C before addition of T
Figure 3. Macrophage monolayers were incubated at 37OC with Listeria (0.3
cells. Thus, the antigen-uptake phase (5 min) and the T cellml of lo7 bacteria per ml) for various periods of time (Processing Time)as
macrophage binding reaction (1 5 min)were held constant. indicated
and then washed before addition of T cells. T cells were added to the
Macrophagestested for T cell binding immediately after antigen macrophages (5 x lo5T cells per well in 0.5 ml) and incubated at 37°C for the
times
indicated
(Binding Times). The Listeria-macrophage interaction and T celluptake showed little or no specific binding, whereas those
macrophage interaction were initiated by centrifugation steps as described in
incubated for 60 min at37OC before the T cell-macrophage Materials and Methods. The mitogenic protein assay wasused to monitor binding:
reaction showed substantial binding of T cells. Clearly, a period T cell activity for the control reaction (No Listeria) was 6002 f 353 Acpm.
F
Downloaded from http://www.jimmunol.org/ by guest on August 3, 2017
ingestion of Listeria monocytogenes by macrophages was monitored visually by
indirect immunofluorescence with an anti-Listeria antibody. Bacteria on the
macrophage cell surface but not those that are ingested are reactive with the
antibody. Each point represents the number of surface bacteria per macrophage
(mean f SEM). Two different experiments are indicated by the different symbols.
Time zero represents macrophages exposed to Listeria for a 5-min centrifugation
period (as in Fig. la), washed, and fixed immediately. c. The catabolism of
Listeria macrophages was followed with time. Macrophage monolayers (1'
0 PEC
per well) were exposed (30 min) to '*'I-Listeria to allow uptake and ingestion as
in Figures 1a and 1 b and then incubated for various periods of time at 3 7 T .
Trichloroacetic acid- (1 0%) soluble (open symbols) and -precipitable (closed
symbols) radioactivity was followed in the macroqhage (triangles) and in the
supernatant (circles).
1872
KIRK ZIEGLER AND EMlL R. UNANUE
T
I
T/M€ fmiid
Figure 4. The T cell-macrophage binding assay was conducted byemploying
variable times of Listeria-macrophage interaction. The time indicated represents
the period oftime beginning with the addition of Listeria
to the macrophages and
ending with the removal of T cells from the macrophage monolayers. The pooled
data from 9 experiments is shown. Each point represents the mean f SEM of
more than 3 separate determinations. Two experiments were carried out with a
5-min T cell-macrophage binding time, 4 with a 15-min binding time, and 3 with
a 30-min binding time. Listerra monocytogenes was used at an optimal density of
10' per ml. Binding was monitored bythe mitogenic protein (3 experiments) and
proliferation assays (6experiments).
127
TABLE II
Effect ofmetabolic and proteinsynthesis inhibitors on antigen presentation'
Binding Substrate
Treatment
Mitogenic Protein
Activity of Macrophage Nonedherent T Cells. hcpm
i SEM
% Specific
Blnding
Expt. 1
Macrophages
Macrophages and
Listeria
None
7.629 f 487
4,359 215
Macrophages
Macrophages and
Listeria
50 pg per ml cycloheximide
8.670 102
4.489 -+ 263
48
Macrophages
Macrophages and
Listeria
10 pg per ml cycloheximide
7,086 f 590
4,969 f 471
38
Macrophages
Macrophages and
Listeria
2 pg per ml puromycin
7,501 f 101
4,521 It 432
39
Macrophages
Macrophages and
Listeria
l o - * M sodium azide
7,930 -C 373
4.025 +. 447
49
Macrophages
Macrophages and
Listeria
10" M 2-deoxyglucose
7,490 f 310
5.574 f 654
30
1 O-' M sodium azide
10" M 2-deoxyglucose
8.351 f 352
7,247 f 77
9
37°C
14,139 f 360
2.654 f 94
81
4°C
13.31 2 f 476
6
Macrophages
Macrophages and
Listeria
*
42
*
Expt. 2
Macrophages
Macrophages and
Listeria
Macrophagesand
Listeria
Macrophage monolayers were prepared asbinding substrates by using PEC
from Listeria-immune mice (1.5 x 10'PEC per well). In Experiment 1 , monolayers
were incubated (30 min. 37'C) with the inhibitors at the indicated concentrations
in a total volume of 0.5 ml. Heat-killed Listeria monocytogenes (50pl per well of
10' bacteria per ml) were added as indicated; the plates were centrifuged (800
x G.5 min) and then incubated for30 min at 37°C in the continued presence of
the inhibitor. After Listeria-macrophage interaction,the
macrophages were
washed to remove the inhibitors and unbound bacteria. Listeria-immune T cells
were added (5 x l o 5 per well in 0.5 ml). the plate centrifuged (50 X G.5 min).
and then incubated at 37°C for 20 min. The T cells nonadherent to the macrophage monolayers were recovered, washed, counted, and then tested for activity
in the mitogenic protein assay. The activity of 1 O5 cells per well is given as Acpm
(mean f SEM) of duplicate binding reactions. Percent specific binding was
calculated by using the activity of T cells nonadherent to macrophages that
received no treatment with Listeria. The protein synthetic capacity of macrophages measured in parallel byusing 'H-leucine incorporation was inhibited over
the time period of Listeria-macrophage and macrophage-T cell interaction. For
example, with 10 pg per ml cycloheximide, 80% inhibition of 3H-leucine incorporation was observed. In Experiment 2, the macrophage monolayers were
incubated with Listeria monocytogenes to allow uptake by macrophages and
then incubated for 60 min at either 37OC or 4°C. T cells were added. and the
binding reaction was carried out for 15 min at37°C.
lAk reagent (neat) showed 30 to 60% inhibition. Inhibition of T
cell-macrophage binding with these reagents was observed
even when macrophages were exposed to the antibodies before addition of antigen. For example, whenmacrophages were
exposed (30 min, 37°C) to the anti-IAk reagent, either before
or after a 30-min period of Listeria-macrophage interaction, 30
f 7% or 35 f 5% inhibition (mean & SEM,3 experiments) of
specific binding, respectively, was obtained. It was of interest,
therefore, to address the possible role of macrophage la molecules in the handling of antigen by macrophages. In the
experiment shown in Table 111, macrophages exposed to anti-la
antibodies were tested for their ability to bind and catabolize
'251-labeledListeria. No effect on the uptake or catabolism or
antigen was observed under the conditions known to inhibit
specific T cell-macrophage binding.
Downloaded from http://www.jimmunol.org/ by guest on August 3, 2017
the total time of handling of the bacterium, i.e., thebinding time
plus processing time.
The pooled results of 9 experiments in which specific binding
was measured as a function of the total time of Listeria-macrophage interaction is shown in Figure 4. The time indicated
represents the period beginning with the addition of Listeria to
the macrophage and ending with the removal of T cells from
the macrophage monolayers. The T cell-binding ability developed after a lag period of about 20 min to maximal levels by 60
min. One-half maximal binding was observed at 45 min.
In a control experiment, macrophages were incubated with
Listeria for 60 min and were then given Listeria for a 2nd time
immediately before addition of T cells. Specific binding was not
inhibited (82% specific binding after 1 hr of Listeria vs 86%
binding after the 2 exposures). We considered that the Listeria
bound to the macrophage cell surface did not interfere with T
cell binding.
Processing requires active macrophage metabolism. If antigen handling was carried out in the presence of sodium azide
(10
" M) and 2-deoxyglucose (10" M), the generation of a T
cell-binding substrate was markedly inhibited (80% inhibition)
(Table 11). Likewise, incubation of macrophages with bound
Listeria at 4°C markedly reduced the binding of T cells (Table
11). Inhibitors of protein synthesis did not affect the binding of
T cells to macrophages (Table 11).
The need for metabolically active macrophages during the
antigen-processing period was also apparent when testing the
binding ability of paraformaldehyde-treated macrophages (Fig.
5). T cells did not bind to macrophages treated (5 min, 20°C)
with 1% paraformaldehyde immediately after antigen uptake.
However, substantial (approximately 60% of control) T cell
binding was observed to macrophages allowed to process
Listeria for 50 min at 37°C before fixation.
Role of la in antigen handling. As indicated in previous
studies, the exposure (30 to 60 min, 37°C) of macrophages to
antibodies directed against I-region gene products resulted in
inhibition of T cell-macrophage binding (1 1 ) . With an A.THanti-A.TL serum (anti-lak)at a 1:10 dilution, 70 to90%inhibition
of specific binding was routinely observed; a monoclonal anti-
[VOL.
ANTIGENHANDLING
19811
1a73
BY MACROPHAGES
READOUT
SYSTEM
f
Treatment of Binding
MITOGENIC PROTEIN PROLlF€RAT/ON
CYTOTOXICITY
TCElL
NUMBER
Macrophages
?
X)
0
20
40
0
20
40
60
80
0
20
40
60
% SPECIFIC BINDING
TABLE 111
lnabflity of anti-fa antibody to after the uptake or catabolfsm of antigen by
macrophages"
Treatment of Macrophages
None
Anti-IA* (undiluted), 30 min
Anti-lak ( 1 :lo). 30 min
% Antigen uptake" (5 Min)
30
28
33
% Antigen Catabolism'
15 Mm
50 Mln
8.3 32.721.8
7.4 33.321.6
19.8
6.5
84 Min
31.4
Macrophage monolayers were prepared from the PEC (lo6 per well) of
Lfsteria monocytogenes-immune mice. Macrophages in this experiment were
76% IA-positive as judged by indirect Immunofluorescence. Macrophages were
incubated (30 min, 37°C) with either a monoclonal ant"*
antibody preparation
or an A.TH anti-A.TL serum (anti-la!').
'251-Listefia(lo5cpm per well) was added;
the plates were centrifuged (800 x G . 5 min), and then the '251-Listerianot bound
to macrophages removed by washing. The macrophages bound with 1251-Listeria
were then incubated for various periods of time at 37OC. and the TCA-soluble
(1 0%) radioactivity released into the supernatant was assessed. Antibodies were
present throughout the 3 7 T incubation period.
I
I
!
,
I
Percent antigen uptake represents the percentage of added '251-Listefiathat
0
20
40
60
i o ' 100 ' 1%
became macrophage associated during a 5-min exposure to '251-Listeria.
TIMEfmml
Percent catabolism represents the percentage of macrophage-associated
radioactivity released into the supernatant as TCA-soluble material after the
figure 6 . The percentage (90)antigen uptake, ingestion, catabolism, and
indicated time of incubation at 37°C.
recognition is shown as a function of time. Antigen uptake is shown as a
percentage of'251-Listeria taken up. Antigen ingestion is expressed as the
percentage decrease in the number of Listeria associated with the macrophage
cell surface. Antigen catabolism represents the percentage of total '251-Listeria
DISCUSSION
taken up by macrophages which is secreted as TCA-soluble material. Antigen
recognition represents the percent specific binding of T cells.
The major thrust of this study was to analyze the kinetics of
antigen uptake, ingestion, and catabolism by macrophages
and to relate such events to the ability of macrophages to serve
as a binding substrate for antigen-specific T lymphocytes. The
system that we chose for analysis of antigen-handling events
clearly involves the recognition of Listeria antigen by T cells in
the context of I-region products of the macrophage. Because
the specific binding of T cells to macrophages could be measured during a short (5- to 15-min) incubation period, it was
possible to follow the temporal development of a T cell-binding
substrate with increasing time of antigen-macrophage interactions (summarized in Fig. 6). We found that T cell binding to
macrophages developed after a 30- to 60-min period of antigen
handling. These kinetics were in marked contrast to the rapid
uptake and ingestion of Listeria by the macrophages, indicating
that antigen uptake alone-a surface event-was insufficient
to generate a macrophage-associated immunogen for T cell
binding. These results formed the basis for the operational
definition of macrophage post-antigen-uptake events required
for antigen recognition by T cells as "antigen processing."
Antigen processing events may alsobe dissociated from
antigen uptake on the basis of temperature and energy dependence. Although simple antigen uptake by macrophages
can occur at 4OC or in the presence of inhibitors of oxidative
and glycolytic metabolism, azide, and 2-deoxyglucose, little or
no T cell-binding ability develops under these conditions (Table
II, Fig. 6). These results are in keeping with previous studies
(22, 23) which suggested that the accumulation of immunogenically relevant antigen by macrophages proceeds by membrane binding and then subsequent metabolic-dependent sequestration of bound antigen. Our findings that the ingestion of
bacteria by the macrophage precedes the development of T
cell-binding ability is in keeping with this pathway of antigen
handling.
The requirement for active macrophage events during antigen processing was also apparent when the binding ability of
paraformaldehyde-fixed macrophageswas studied (Fig. 5).We
conclude that once antigen processing by macrophages is
Downloaded from http://www.jimmunol.org/ by guest on August 3, 2017
Figure 5. Macrophage monolayers were exposed to Listeria monocytogenes (LM) as described in Figure 4 to allow uptake and then treated as indicated in the
schematic. The 1st group was treated with PES instead of 1 % paraformaldehyde (PCH20) and cultured in parallel as a positive control. All culture periods were at
37%. and treatment with PCH20 and PBS was carried out at 20°C. Extensive washing is indicated by the vertical curved line. The T cell-macrophage binding
reaction was performed as described before. With each condition, binding macrophages without exposure to Listeria served as controls, no alteration in these control
activities being seen with paraformaldehyde treatment. Specific binding was monitored by the indicated readout systems. The pooled results of 4 separate
experiments are shown as the mean f SEM percent specific binding.
1874
KIRK ZIEGLER AND EMIL R. UNANUE
127
surface la molecules. It should be noted that studies ofthe
interaction of macrophage-bound molecules in B cell-T cell
interactions have provided evidence for macrophage-associated molecules, but which retained their native structure and
sensitivity to trypsinization (15, 34). Thus,the macrophage
apparently serves through 2 distinct pathways of antigen handling to present antigen to both major sets of lymphocytes (2,
34).
The survival values of antigen degradation in the handling of
pathogens may be viewed as a way to increase the number of
different structural moieties that can serve as antigens. This
advantage may be particularly relevant with regard to complex
intracellular pathogens such as Listeria monocytogenes. Thus,
bacterial components normally sequestered in the interior of
organisms could conceivably serve as antigens, and the multiplicity of such antigenic determinants would make it less likely
that a nonresponder status with respect to I-region gene function would be generated.
Acknowledgments. The excellent technical assistance of Ms.
Andrea P. Justice is gratefully acknowledged. We also thank
Ms. Barbara K. Gricus for her editorial and secretarial assistance during the preparation of this manuscript.
REFERENCES
1. Moller. G., ed. 1978. Role of macrophages in the immune response. Immunol. Rev. 40:l.
2. Unanue, E. R. 1981. The regulatoryrole of macrophages in antigenic
stimulation. Part Two: Symbiotic relationship between lymphocytes and
macrophages. Adv. Immunol. 31 : l .
3. Rosenthal, A. S . 1980. Regulation of the immune response-role of the
macrophage. N. Engl. J. Med. 303:1153.
4. Shevach, E. M.. and A. S . Rosenthal. 1973. Function of macrophages in
antigen recognition byguinea pig T lymphocytes. II. Role of the macrophage
in the regulation of genetic control of the immune response. J . Exp. Med.
138:1213.
5. Lipsky. P. E..and A. S. Rosenthal. 1975. Macrophage-lymphocyte interaction. II. Antigen-mediated physical interactions between immune guinea pig
lymph node lymphocytes and syngeneic macrophages. J. Exp. Med. 141:
138.
6. Ben-Sasson. S . Z., M. F. Lipscomb. T. F. Tucker, and J. W. Uhr. 1977.
Specific binding of T lymphocytes to macrophages. II. Role of macrophageassociated antigen. J. Immunol. 11 0:1493.
7. Werdelin. O.,and E. M. Shevach. 1979. Role of nominal antigen and la
antigen in the binding of antigen-specific T lymphocytes to macrophages. J.
Immunol. 123:2779.
8. Werdelin. O., 0.Eraendstrup. and E. M. Shevach. 1979. Specific absorption
of T lymphocytes committed to soluble protein antigens by incubation on
antigen-pulsed macrophage monolayers. J. Immunol. 123:1755.
9. Lipscomb. M. F., S. Z. Ben-Sasson. T. F. Tucker, and J. W. Uhr. 1979.
Specific binding of T lymphocytes to macrophages. IV. Dependence on
cations, temperature, and cytochalasin E-sensitive mechanisms. Eur. J.
Immunol. 9.1 19.
10. Lyons, C. R., T. F. Tucker, and J. W. Uhr. 1979. Specific binding of T
lymphocytes to macrophages. V. The role of la antigens on macrophages in
the binding. J. Immunol. 122:1598.
11. Zlegler. K.. and E. R. Unanue. 1979. The specificbinding of Listeria
rnonocytogenes-immune T lymphocytes to macrophages. I. Quantitation and
role of H-2 gene products. J. Exp. Med. 150:l 143.
12. Ziegler, K.. and E. R. Unanue. 1980. Quantitation of genetically restricted T
cell-macrophage binding. In Macrophage Regulation of Immunity. Edited by
E. R . Unanue and A. S. Rosenthal. Academic Press, New York. Pp. 245262.
13. Geha. R. S..M. E. Jonsen, B. H. Ault, E. Yunis, and M. D. Broff. 1981.
Macrophage-T cell interaction in man: binding of antigen-specific human
proliferating andhelper T cells to antigen-pulsed macrophages. J. Immunol.
126:781.
14. Unanue, E. R. 1972. The regulatory roleof macrophages in antigenic
stimulation. Adv. Immunol. 1 5 9 5 .
15. Unanue, E. R.. and J.-C. Cerottini. 1970. The immunogenicity of antigen
bound to the plasma membrane of macrophages. J. Exp. Med. 131 :711.
16. Kolsch, E., and N. A. Mitchison. 1968.The subcellular distribution ofantigen
on macrophages. J . Exp. Med. 128:1059.
17. Farr, A. G.,J. M. Kiely. and E. I?. Unanue. 1979. Macrophage-T cell
interactions involving Listeria rnonocytogenes-role of the H-2 gene complex. J. Immunol. 122:2395.
18. Greenwood, F. C.,W. M. Hunter, and J. S. Glover. 1963. The preparation
of ‘3’1-labeled human growth hormone of highly specific radioactivity. Eio-
Downloaded from http://www.jimmunol.org/ by guest on August 3, 2017
complete, then the macrophage cell surface serves to display
antigen that T cells recognize. Thus, a mechanism of T cellantigen recognition involving an active T cell-induced display
of antigen previously sequestered by the macrophage seems
unlikely. The abilities of aldehyde-fixed tumor cells to serve as
targets for T cell-mediated cytotoxic attack (24) and aldehydefixed TNP-modified macrophages to stimulate TNP-specific T
cell proliferation (25) have been noted previously.
The possible role of antigen catabolism in the macrophagehandling events relevant to T cell-antigen recognition is suggested by several pieces of evidence. First, T cells did not bind
to intact antigen (Table I). Second, the kinetics of antigen
catabolism correlated most closely with the kinetics of development of a T cell-binding substrate (Fig. 5). Third, both the
appearance of T cell-binding ability and antigen catabolism
were inhibited by low temperature, metabolic energy inhibition,
and aldehyde fixation. Also, in results to be published, ammonium chloride, a well-studied inhibitor of protein degradation in
cultured cells, resulted in a substantial inhibition of antigen
catatjolism without effect on antigen uptake or ingestion. This
inhibition of antigen catabolism was associated with a corresponding inhibition of the development of T cell binding. Since
inhibition of protein degradation by ammonia is thought to
result from an increase in lysosomal pH, thereby inhibiting the
action of acid hydrolases (26,27), and/or by inhibiting phagosome-lysosome fusion (28), these results implicate the lysosome in an intracellular pathway of antigen handling relevant
to T cell-antigen recognition.
Finally, we examined the possibility that I-region gene products might be involved in handling of antigen by macrophages
either by serving as receptors for the initial uptake of antigen
or as molecules functioning in antigen catabolism. The inability
of anti-la antibodies to alter the uptake or catabolism of antigen
by macrophage populations highly enriched in la-positive cells
(Table 111) argues against such I-region gene involvment in initial
antigen-handling events.Thus, I-region gene products may
function either as receptors for fragments of antigen derived
from antigen catabolism and/or as structures recognized directly by T lymphocytes.
The pathway of immunologically relevant antigen handling
may be envisioned as follows: Antigen interacts with the macrophage cell surface directly via trypsin-sensitive receptors
(29) or by way of Fc/C3 receptors and is interiorized within
phagosomes. Antigen-containing phagosomes then fuse with
lysosomes, and partial degradation occurs. Some small fragments of antigen created by action of lysosomal proteinases
are then released and transferred to the macrophage cell
surface by a process akin to exocytosis (30). At some point
after catabolism, the antigen fragments may cooperate with
and/or interact with la molecules and thus form a multi-molecular antigenic structure recognized by T cells (3, 4, 31, 32). In
contrast to the receptor initially involved in antigen binding, the
putative membrane form of this processed antigen like la is
relatively trypsin insensitive (23; Ziegler and Unanue, unpublished observations). It is not known whether the intracellular
handling of antigen or the putative association of antigen fragments with la molecules is the rate-limiting step in antigen
processing. In this regard, it would be of interest to determine
whether antigens of minimalsize, such as the octapeptide
studied by Thomas et a/. (331, would show a requirement for
an antigen-processing period and ammonia effects as described here for a complex bacterial antigen. It is possible that
the intracellular pathway of antigen handling may be bypassed
by direct interaction of antigen fragments with macrophage cell
[VOL.
1981J
HANDLING
ANTIGEN
chem. J. 89:114.
19. Farr, A. 0 . .W. J. Wechter, J. M. Kiely. and E. R. Unanue. 1979. Induction
of cytocidal macrophages following in vitro interactions between LiStefi.9immune T cells and macrophages-role of H-2. J. Immunol. 1222405.
20. 0 1 , V. T., P. Jones, J. W. Goding. L. A. Herzenberg. and L. A. Herzenberg.
1978. Properties of monoclonal antibodies to mouse lg allotypes, H-2. and
la antigens. In Lymphocyte Hybridomas. Edited by F. Melchers. M. Potter.
and N. Warner. Springer-Verlag. NewYork. Pp. 11 5-237.
21. Beller. D. I.,J. M.Kiely. and E. R . Unanue. 1980. Regulation of macrophage
populations. I. Preferential induction of la-rich peritoneal exudates by immuno\ogicalstimuli. J. Immunol. 124:1426.
22. Ellner, J. J.. and A. S. Rosenthal. 1975. Qualitativeand immunologicaspects
of the handling of 2.4-dinitrophenyl guinea pig albumin by macrophages.J.
Immunol. 114:1563.
23. Ellner. J. J.. P. E. Lipsky. and A. S. Rosenthal. 1977. Antigen handling by
guinea pig macrophages: further evidence for the sequestration of antigen
relevant for activation of primed T lymphocytes. J. hmunol. 118:2053.
24. Bubbers. J. E.. and C. S . Henney. 1975. Studies on the synthetic capacity
and antigenic expression of glutaraldehyde-fixed target cells.
J. Immunol.
114:??26.
25. Thomas, D. W. 1978. Hapten-specific T lymphocyte activation by glutaraldehyde-treated macrophages: an argument against antigen processing by
macrophages. J. Immunol. 121:1760.
26. Seglen, P.S.. 0. Grinde. and A. E. Solheim. 1979. Inhibitionof the lysosomal
1875
BY MACROPHAGES
pathway of protein degradation in isolated rat hepatocytesbyammonia.
methylamine, chloroquine. and leupeptin. Eur. J. Biochern. 9 5 2 1 5 .
27. Ohkuma. S . , and 8 . Poole. 1978. Fluorescence probe measurement of the
intralysosomalpH in living cells andthe perturbation of pH byvarious
agents. Proc. Natl. Acad. Sci. 753327.
28. Gordon, A. H.. P. D'Arcy Hart, and M. R. Young. 1980. Ammonla inhibits
phagosome-lysosome fusion in macrophages. Nature286:79.
29. Weinberg, D. S., and E. R . Unanue. 1981. Antigen-presenting function of
alveolar macrophages: uptake and presentation of
listeria rnonocytogenes.
J. lmmunol. 126.794.
30. Calderon, J., and E. R. Unanue. 1974. The release of antigen molecules
from macrophages: characterization ofthephenomena. J. Immunol. 112:
1804.
31. Erb, P., 6. Meier, and M. Feldmann. 1979. Is genetically relatedfactor (GRF)
a soluble immune response (Ir) gene
product? J. lmmunol. 122:1916.
32. Lonai, P.. J. Puri. and G. Hammerling. 1981. H-2-restricted antigen binding
by a hybridoma clone that produces antigen-specific helper factor. Proc.
Natl. Acad. Sci. 78~549.
33. Thomas, 0.W.. K. Hsieh, J. L. Schauster. M. S. Mudd. and G. D. Wilner.
1980. Nature of T lymphocyte recognition of macrophage-associated antigens. V. Contributionof individual peptide residues of humanfibrinopeptide
B to T lymphocyte responses. J. Exp. Med. I52:620.
34. Unanue. E. R. 1978. The regulation of lymphocyte functions by the macrophage. Immunol. Rev. 4O:lS.
VoI 127,No. 5 . November 198l
Prrnted In U S A
SUBSTRATE HYDROLYSIS BY IMMUNE COMPLEX-ACTIVATED NEUTROPHILS: EFFECT
OF PHYSICAL PRESENTATION OF COMPLEXES AND PROTEASE INHIBITORS'
KENT J. JOHNSON
AND
JAMESVARANI
From the Department of Pathology, The University of Michigan Medical School, Ann Arbor, MI 48 109
ineffectivenessof fluid phase protease inhibitorsblock
to
Theabilityof
rat leukocytestohydrolyzearadiolathe protease activityof contact-activated leukocytes
may
beled, surface-bound protein substrate in a solid phase
can take place in the
assay wasdetermined, andvarious factors that influence explain how immune complex injury
presence of high concentrations of serum inhibitors.
the processweremeasured.Unstimulatedleukocytes
hydrolyzed verylittle substrate. Whenthe cell suspension
was mixedwithzymosanparticles
orincubatedwith
preformed immune complexes,the amount of substrate
Much interest in recentyearshasbeengenerated
in the
hydrolysisincreaseddramatically.Notsurprisingly,imstudy of the pathogenesis of immune complex-mediated tissue
mune complexes at equivalence proved to be the most injury. Perhapsthesimplestmodel
of animmune-complex
effective ineliciting the response.Immunecomplexes
lesion is the Arthus reaction, which is a localized necrotizing
attached to the surface along with the protein substrate
vasculitis, with the antigen and antibody complexing within the
were able to effectively induce hydrolysis, though they
the tissue injury
vessel wall(1 ). Earlier studies have shown that
were not as effective as immune complexes in suspenseen
with
the
Arthus
reaction
is
complement
and neutrophil
sion. Three protease inhibitors, a,-antitrypsin, an-macrodependent,
with
abolition
of
either
abolishing
the
tissueinjury
globulin, and soybean trypsin inhibitor, which were able
toneutralizenearly
all oftheproteaseactivityin
rat (2).
Speculation has therefore centered in recent years on what
neutrophil lysates, were tested for their ability to inhibit
immune
complex-induced
protein
hydrolysis.
It was constituents of neutrophils are responsiblefor the tissueinjury.
more specifically,
found that when the inhibitors were surface bound alongStudiesdone with neutrophillysatesor,
are
withthesubstrateprotein,theywere
effective in pre- lysates of cytoplasmic lysosomal granules, reveal that they
capable of damaging basement membrane preparations
in vitro
ventingtheneutrophilsfromhydrolyzingtheprotein.
However, when the same inhibitors were present in the (3). In rabbit neutrophils the specific lysosomal constituents
fluid phase, they were much less effective. The relative responsible were found to be cathepsins 0 and E, in human
neutrophils, neutral proteases (4). Many other studies (5, 6)
Received for publication February 10, 1981.
Accepted for publication July 24, 1981.
The costs of publication of this article were defrayed in part by the payment
Of Dage charges. This article must therefore be hereby marked advertisementin
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
' This work was supported in part by National Institutes of Health Grants HL28442. HL-26809, HL-00889, and CA-29551.
haveshownthatlysosomalenzymesarereleased
into the
extracellular fluid during events such as phagocytosis of immune complexes.Therefore,aplausibleexplanation
ofthe
tissue injury seen in the Arthusreaction is the releaseof neutral
proteasesfrom the neutrophils during phagocytosisofthe
immune complexes in the vessel wall. It would therefore follow
Downloaded from http://www.jimmunol.org/ by guest on August 3, 2017
0022-1767/a1/1275-1a75%02.00/0
THE JOURNAL OF IMMUNOLOGV
Copynghi Q 1981 by The Amerlcan Associalion of Immunologists