0022-1767/83/1313-1234$02.00/0
JOURNALOF lYMlJNOLOGV
Copyright 0,1983 by The American Associatm of Immunologists
THE
Val. 131, No. 3, September 1983
Primed in U.S.A.
INDUCTION OF ANTIGEN-SPECIFIC ANTIBODY RESPONSE IN HUMAN PERIPHERAL
BLOOD LYMPHOCYTES IN VITRO BY A DOG KIDNEY CELL VACCINE AGAINST RABIES
VIRUS (DKCV)
FONS G. C. M. UYTDEHAAG, ALBERT D. M. E. OSTERHAUS,HENK G. LOGGEN,ROLANDH. J. BAKKER,
JACK A.A. M. VAN ASTEN, JOHAN G. KREEFTENBERG,PIETER VAN DER MAREL, AND BERT VAN STEENIS
From the National lnstitute of Public Health. P.O. Box 1, 3720 BA Bilthoven, The Netherlands
In the present report anin vitro method for obtaining a mocyanin of helix pomatia (12), and tetanus toxoid (13-15).
secondary human antibody response to a dog kidney cell Furthermore, systems have been described that have enabled
vaccine against rabies virus (DKCV) is described. Cultures
the measurement of specific antibody in cultures of human B
of peripheral blood mononuclear cells from normal rabiescells stimulated with influenza virus(16, 17) and varicella zoster
immune and nonimmune donors were stimulated in vitro virus (18).
byDKCV.Theproduction
of virus-specificantibodyin
In this paper, we report the in vitro induction of a secondary
supernatant fluids was monitored by ELISA. Antibody was
antigen-specific antibody response of human peripheral blood
produced by lymphocytes from rabies-immune individulymphocytes (PBL) by a dog kidney cell vaccine against rabies
als, whereas those of nonimmune subjects consistently
virus (DKCV). This vaccine has provento be a potent inducer of
failed to produce anti-rabies antibodies after
in vitro stim- neutralizing antibody in man(19). In addition, it has been shown
ulation with DKCV. The generation of the anti-rabies virus
to protect animals against an otherwise lethal challenge
with live
antibody response of lymphocytes stimulated with DKCV
rabies
virus
(19).
The
kinetics
of
this
response,
“the
quality”
of
wasshown to be an antigen-dependenty as well as an
antibody produced, as well as the cellular requirement for this
antigen-specificprocess.Optimalantigen-specificresponses were observed at relatively low concentrations response, have been investigated.
of antigen (10” to lo’* pg/culture). At increasing concenMATERIALS AND METHODS
trationsofantigeninculture
(>1 pg/culture),theantiImmunizations.Normal healthy volunteers received
rabiesvirusresponsewassuppressed.Antibodyproa primaryintramuscular
immunizationof one dose ofDKCVfollowedbya
1- and 6-mo booster
duced upon stimulation was capable of neutralizing rabies
Of by immunization on days 7 and 21 as indicated in Resuks.
virus. The response to rabies virus requires
T cell help irnmunizatlon
Before immunization, anti-rabies antibodies were not present in the sera of
because lymphocytes depleted of SE rosetting cells did the subjects as has been reported (19). After completion of the vaccination
not respond to an antigenic stimulus. Studies in which the
schedule, neutralizing anti-rabies antibodies were present in sera of all individuals. In vitrostudieswereperformedattimesafterimmunizationas
same individuals were followed over time showed a seindicated in Results.
quential development of circulating B cell subsets. The
Lymphocyte isolation. Mononuclear cells were isolated from heparinized
system may provide a model for the study of human
B Mood by density gradient centrifugation on Ficoll-lsopaque
(p = 1.077 g/cm3)
cell differentiationin vivoand in vitroand maybe valuable at 1000 X G for20 min and then were washed twice with
RPMl1640 (GIBCO,
Grand Island, NY) supplemented with penicillin, streptomycin, and Fungizone
for testing the potency of rabies vaccines
in vitro.
(100 IU, 100 pg, and 25 pg/ml, respectively).
Monocyte depletion.Mononuclear cells were depleted of monocytes by
adherence to plastic Falcon flasks at 37’ C in RPMl 1640 medium supple
Polyclonal B cell activators (PBA)’ have been a useful probe
mented with20% fetal calf serum (FCS). After60 min, Supernatants containing
to study activation and regulation ofB lymphocyte responses in lymphocytes were harvested and washed twice
in RPMl1640 medium. After
man. Antigen-specific antibody responses, however, that result the adherence procedure. the cell suspension usually contained between 5
and 10% monocytes determined by morphology and nonspecific esterase
fromexposureofhuman
B cells in vitro to PBA, such as staining.
pokeweed mitogen, are part of a polyclonal immunoglobulin (lg) T cell isolation and depletion. Mononuclear cells were separated into
T cell
response. The latter typeof response may greatly differ from B and non-T cell populations by methods previously described (5). Briefly, T
cells were isolated by incubation
of lymphocytes with a 50-fold number of SE
cell responses triggered by antigen itself(1-3).
that had been treated with neuraminidase. After rosette formation,
cells were
Severalinvestigatorshaverecentlyestablishedtechniques
centrifuged on a density gradient of Ficoll-lsopaque
(p = 1.077 g/cm3) for 30
that permit analysis of either the primary or secondary in vitro min at 1200 x G. The resulting T cell sediment was freed of contaminating
antibody response by human 6 cells after antigen stimulation. SE by lysis with cold buffered NH4CI. The cells from the interface of the
that were enriched for B cells were cycled again. Non-T
cell prepaAntigens appliedin these studies included particulated antigens, gradient
rations usedin the experiments contained52% SE rosetting cells.
i.e., sheep red cells(SE) (4-7), haptenated polyacrylamide beads Antigens. Rabies vaccine used for stimulation
in vitro was produced by the
(B), and soluble antigens, such as ovalbumin (4, 5), haptenated propagation of rabies virus strain Wistar PM/WI38-15033M in primary dog
cells (DKCV) and was inactivated with p-propiolactone as described
ovalbumin (9), keyhole limpet hemocyanin (KLH) (10, l l ) , he- kidney
elsewhere (20). One dose of lyophilized vaccine containing 300 r g protein
was dissolvedin 1 ml phosphatebuffered saline (PBS) and was further dlluted
cells was
in culturemedium.Rabiesvaccineproducedinhumandiploid
27, 1982.
Received for publication May
obtainedfrom MBrieux (France). Measles virus (Edmonston strain) was grown
Accepted for publication May
11,1983.
to methods
on Vero cells in microcarrier cultures and was purified according
The costs of publication of this artide were defrayed in part the
by payment of
used for the purification of rabies virus (20). Polio virus was produced and
pagecharges.Thisarticlemusttherefore
be herebymarkedadvertisementin
purified accordingto methods described by van Wezel et (20).
a/.
accordance with18 U.S.C. Section 1734 solely to indicate thisfact.
Culture conditions. Lymphocytes were suspendedin RPMl 1640 medium
’ Abbreviations used in this paper:DKCV, dog kaney cel vaccine; RFFIT, rapid containing
25 mM HEPES,‘ 2 mM L-glutamine, penicillin, streptomycin, and
Ruorescent focus hhibttiontest; SE. sheeperythrocytes; KLH, keyholelimpet
hemocyanin;HEPES.N-2-hydroxyethylpiperazineN’“hthanesulfonic acid; ELISA, Fungizone, 100 IU, 100 pg/ml, and 25 pg/ml, respectively, lo” M 2-mercap
toethanol, 1 mM Na pyruvate, and 10% horse serum. Cells were culturedin
enzyme-linked immunosorbent assay.
1234
1235
IN VlTRO RABIES-SPECIFIC
PRODUCTION
ANTIBODY
the presance of various concentrations of DKCV. Cultures were kept in 24- concentrationsrangingfrom
to 10" pglculture ofDKCV.
well, flat-bottomed plates (Costar Data Packaging, Cambndge. MA) at acell The further increase of antigen concentration in culture(>1 pg/
densityof2 x 1@/ml in a final volume of 1 ml at 37" C in a humidfied
1).
atmosphere of 5% Conin air. After 6 days the cells were harvested, washed culture) resulted in the abrogation of antibody formation (Fig.
twice, and recultured for5 days in culture medium containing
5% FCS (Amxxrr Thishasbeen aconsistentfindingwithcellsfrom
25 rabiesPharmaceutical Company, Kankakee,IL) without antigen. Supernatant fluids vaccinatedindividualswhowerestudied.Lymphocytesfrom
were then harvested and assayed for a n t i "
antibody, for antibody to
nonimmunized subjects failedto respond to rabies vaccine at all
@io virus, and for total lg.
Enzyme-linked immunoswbmt assays (€USA). Antibodies to rabies virus concentrations that were tested.
soridphase immunoassay withthe
in culture supernatants were detected by
To determine whether the antibody generated by DKCV
in
use of a m o d i t i o n of the method of Zanders et a/. (21). Flat-bottomed
microtiter plates (Titertek Type111, Flow LaboratoriesIN., Mclean, VA) were vitro resulted from de novo protein synthesis, antigen-stimulated
coated with 100pl of 12 pgglml DKCV in PBS, pH 7.4, for 1 hr at 37' C in a cells were cultured in the presence of100 pg/culture cyclohexihumidified atmosphere. Plates were washedtwice with PBS. incubated with mide. The addition of cycloheximide
5 days before the harvesting
PBS containing1% bovine serum albumin (BSA)for 30 min at 37" C
to block of culture supernatants reduced antibody production to less than
1%
BSA.
remaining binding sites, and washed again with PBS containing
10% of values found in supernatants of cells cultured without
This procedure consistently resulted
in low backgroundvalues. Afterwashing,
60ofculturesupernatant
and dilutionsthereof in PBS 0.5% BSAwere
this drug (Fig. 1).Furthermore, no antibody production couldbe
applied to the wells and were incubated for 1 hr at 37" C. Miluted culture detected in cultures of cells with medium alone(0 pg/culture of
supernatant,aswell
as three serialfivefolddilutions,wererun
for each
sample. Each plate also contained four serial fivefold dilutions of a referenceDKCV) (Fig.1).
Kineticsofthe
in vitroanti-rabiesresponse.Timecourse
serum composed of poded serum of a hyperimmunedonor. The plates were
washedwithPBScontaining
0.05% Tweenbyusingasemi-autom@c
studies were performedto define thein vitro kineticsof the antiwashingdevice and wereincubatedat room temperaturefor2hrwith
rabies antibody production. Cells were cultured for various peantihuman Ig conjugated with peroxidase; goat anti-p chain (Cappel Laborariods
after which the supematants were tested for the presence
tories, Downingtorm, PA), sheep anti-human y-chain, and polyspedfic sheep
anti-human lg were used. The production of the latter two conjugates has
of specific antibody in the ELISA. Specific antibody production
been described elsewhere (22). The conjugates were diluted in PBS contain- wasnotdetectedthroughday
3 of culture.Thenantibody
ing 0.5% BSA. After incubation with conjugated antisera the plates were
washed with PBS containing
0.05% Tween, and 100 pl of the substrate were production increasedto a maximum by day 10 or11 (Fig. 2). To
added to each well. The substrate consistedof 1.67 ml tetrarnethyl-benzidine avoid residual antigen present in the culture supernatant affecting
in dimethylsulfoxide(6 mg/ml) in 100 mlO.1M sodium acetate buffer,pH 5.5. thedetectionofspecificantibody,thefollowingexperiments
and 10 of
3
0
% solution of H h (Pehydrole, Merck. Darmstadt). After 10
wereperformed.Afteranincubationperiodof
4 daysinthe
min, the reaction was stopped by
the a d d i i of 100 &well 1 N HzS04(23).
free of antigen and were
Abmrbewe at 450 nm was measured by using a Multiskan
Tiiertek automatic presence of antigen, cells were washed
plate reading machine connected with a PDP 1170 computer system. To
cultured again in fresh media for an additional period of 6 to 7
calculate the concsntration of IgMand IgG in supernatant fluids, the optical days.Thesupernatantswerethenassessedforantibodies
densitii of samples and reference serum were compared.For this purpose
we developed a line comparison method. To find out w h i i combination of against rabies virus. The results of these experiments showed
configurationswould givethe best c o n e l a t i o n , a computer program was
used again that maximal and constant production of specific antibody
been pre wasdetectedafteraperiodofcultureof
(vanAsten et a / . , unpublisheddata).Configurationsthathad
10 days(datanot
grammed were: linear,bgarithm, redprocal, probit square, and square root.
On the basis of data obtainedwith 72 samples, the combination of bgarithm shown). These results argue against an assay artifact related to
of concentration vs linear extinction was selected. Next, the actual concen- antigen in culture and confirm the data obtained with cyclohexitration in nanogramsof lg in a sample was calculated by the determination mide studies that indicate that in vitro protein synthesis is r e
of the slope ratio of the slopes of the best-fitting line of both the reference
quired for this response.
and sample. By using this method, slight differences in background values
Antigen-specific nature of the response to rabies vaccine. The
did not influence the final results. Large differences in background values,
however, were noticeaMe by poor correlations that automatically indicated
the necessityto repeat the test.
Standardization of the reference serum. The reference serum was standardized according to the method described by Stevens and Saxon (24),
which provides an estimate ofthe antibody content in nanograms
per milliliter.
The amount of rabies-specmc lg in the referencewasestimatedby
the
comparison of the amount of lg reference serum bound to microtiter wells
coated with DKCV with the amount of an lg standard bound to microtiter
wells coated with anti-lg antibody. The plates were coated with 150 pl/well,
5 pg/ml pdyspecific sheep anti-human lg antibody, or 12 pg/ml DKCV in
carbonate buffer, pH 9.6, by incubation at 20° C for 24 hr. After washing,
wells coated with anti-Ig antibody were filled with 100
@Iof dilutions ofan lg
standardpreparation, whereas wellscoatedwithvirusreceived100of
serial fivefold dilutions ofthe reference serum. After 2 hr at 37" C, the plates
werewashed, and 100 pl ofpurifiedperoxidaseconjugatedpdyspecmC
sheep anti-human lg (22) were applied to each well. After 2 hr at 37" C,
plates were devebpedas described above. The amount of rabies-specificlg
in the reference was estimated comparison
by
of the optical densities obtained
with virus-coatedwells with the curve obtainedwith anti-lgcoatedwells. The
determinationof rabies-spedfic IgM and IgGin the reference was estimated
in ani d e n t i manner. Goatanti-human p-chain-specific (Cappel Laboratories)
and sheep antihuman y-chain-speafic(22)antiserawere
used to coat
microtiter wells. Purified human lgGand IgM were used as standard prepar a t i s . To develop the plates, a polyspecifk sheep
anti-human lg conjugated
with peroxidase wasused (22).
Virus-neutra/izinQantibodydetermination. Theneutralizing capacity of
culture supernatantson rabies virus (CVS strain) was evaluated by the rapid
fluorescent focus inhibition test(RFFIT) (25).
o
10-4
10-3
10-2
10-1
1
10 30
DKCV concentration (,ug/culture)
RESULTS
Figure 1. Antigen dose depandency of anti-rabies response. PBL were cultured
Antigenic requirements. Anti-rabies antibody responses were in the presence of different concentrations of DKCV. Specific antibodies in the
culture supernatant fluid were determined on day 11. Three representativeexperiseen at concentrations of DKCV in lymphocyte cultures as low ments from different donors are shown. Values are expressed asmean ng $/ml f
as 10" pg protein/culture. Peak responses were observed at SE of triplicate determinations.
I
1236
/N VITRO PRODUCTION
RABIES-SPECIFIC
ANTIBODY
-
tion with measles virusor polio virus didnot sighificantly change
the level of antibodies in supernatant fluids of rabies vaccinestimulated cultures (TableI).
AlthoughDKCV used to stimulate the cultures is relatively
E
pure and antibodies to dog kidney cells have never been found
fl50
in individuals immunized with DKCV (unpublished observation),
>r
73
the generationof antibody against dog kidney cel antigens was
B
consideredafterstimulationoflymphocyteswithDKCV.The
.C
supernatant fluids from cultures of different individuals containing
0
rabies-specific antibody were tested in ELISA against a coat of
g 100
noninfected dog kidney cell extracts. None of these supernatant
.-L
>
fluids showed a significant reaction
in ELSA (Table I).
ul
The nonpolyclonalnature of DKCV. Thepossibilitythatan
W
6
antigen may behave like a polyclonal activator cannot be excluded
50(26). This possibility was tested by culturing PBL with
.various concentrations of DKCV or in the absence of antigen.
C
0
As can be seen in Figure 3, the characteristic bell-shaped doseresponse curve was obtained for rabies-specific antibody when
PBL were exposedto various concentrations of DKCV. Totallg
e31
3
5
7
9
11
13
15
17
production, however, did not vary significantly among stimulated
and non-stimulated cultures indicating that, at least in this sysdays in culture
tem, DKCV does not provide a polyclonal stimulatory signal to
Figure 2. The kinetics of anti-rabies antibody formation in cultures of PBL. CellsPBL.
were cultured for various periods in the presence of 0.1 pglcutture DKCV. One
In vivo kinetics of rabies-specific antibody production. In PBL
representative experimentis shown. Values are expressed as mean ng
Ig/ml f SE
of triplicate determinations.
cultures from six individuals who were followed over time,
an in
vitro
antigen-inducible
antigen-specific
IgM
antibody
response
TABLE I
wasobserved 1 wk afterprimaryimmunization. Whentaken
Smificitv of anti-rabies resmnse
before immunization with DKCV, PBL of these subjects did not
Specific
Antibod9
Supematant
respond to antigenic stimulation in vitro (Fig. 4). Maximal in vitro
Fluid AdELSA w a t
Culture
sorbed
Stimulus
antigen-induced, rabies-specific IgM responses were observed
DKCV DKCE' Polio virus
witH
in PBL cultures of all individuals2 to 5 wk after thefirst booster
0.1 11f 0.017 0.094 f 0.023
None
immunization
(Fig. 4). Whenexamined 1 wk after the primary
1.699 f 0.013 0.070 f 0.021
DKCV
immunization, PBL of some subjects produced rabies-specific
0.065 f 0.026 1.494 f 0.013
Polio virusd
f 0.049 1.477 f 0.154
OKCV 0.131
IgM antibody when culturedin the presence of very low, other0.330 2 0.012
DKCV
DKCV
wise nonstimulatory concentrations of antigen; moreover, this
1.620 f 0.022
DKCV
Measles virus
200
-
-
-
L
P
L
pd$
DKCV
Polio virus
1.660 f 0.034
1 volhighlyconcentratedantigen:
"supernatant fluidswereadsorbedwith
33 pg/ml. for 1 hr at 37°C.
DKCV, 300 pg/ml; measles virus,150 pg/ml; polio virus,
After centrifugation supernatant fluids were tested for specific antibody.
a Data expressed as mean OD 450 nm f SE. One representative experiment
out of three is shown.
Dog kidney cells were cultured as described
in Reference 20. The noninfected
cells were washed free of serum and were subjected to three cycles of freezing
and thawing. The clarified supernatant fluid was diluted with PBS, pH 7.4,
to a
protein content of12 pg/ml. One hundred microliters of this material was used as
coat in ELISA DKCE.dog kidney cell extract.
The antibody response induced by polio virus
in vitro will be described in detail
elsewhere (UytdeHaag et a/., manuscript in preparation). Background values in
ELISAfordeterminationof anti-pdii virusantibodywereusuallybelow0.100
absorbence.
specificity of the response was assessed by culturing cells of
individuals who were double immune
to rabies virus andto polio
virus with DKCV or polio virus and by testing supernatant fluids
for antibodies against rabies virus and polio virus. The outcome
of such experiments revealed evidence that antibody generated
by DKCV in vitro was specific for the inducing antigen, because
no significant production of antibodies against polio virus was
observed(Table I). On theotherhand,supernatantfluidsof
cultures stimulated with polio virus did not bind to DKCV used
in the ELISA.In the latter cases, specific antibodiesto polio virus
could be detected by using the homologous antigen
in the ELISA.
The specificityof the response couldbe further established by
the absorption of the culture supernatant fluids with the immu0
1o-L
l f i 10-2 10-1
1
10
100
nizing antigen or an unrelated antigen. The addition of excess
DKCV concentration ( ~ ~ g l c u l t u r e )
rabies vaccineto culture supernatant fluids of the rabies vaccine-Figure 3. The nonpolyclonal nature of antibody responses
of PBL of five individstimulated cells before determinationof the antibody content in uals induced by DKCVin vitro. Data are expressedas geometric means f SE. 0.
the ELISA resulted in a diminished antibody response. Absorp- specific anti-rabies virus IgG;0. total IgG.
1237
IN VlTRO RABIES-SPECIFICANTIBODYPRODUCTION
TABLE II
Rabies virus-neutralizing antibodies producedin Vitro
~xperiment
\
\
1
1
2
3
4
5
6
Days after
Bmstef
18
18
183.5
18
250
46
Antiqabies lg Culture Supematant
Fluids
173 f 14"
1 174 -e 31
<0.1 93 f 17
161 f 8
127 f 29
106-e9
.r
Anti-rabies
Serum
1
11.4"
14.3
0.3
0.2
co.1
14.0
5.0
7.0
.o
Peripheral Mood was obtained from six different donors after booster immunizaticm at the times indicated. After stimulation of the cells with DKCV, culture
supernatant fluas were testedfor antigen binding the
in ELSA and for the presence
of neutralizing antibodies asdescribed in Maierias and Methods.
Values are expressedas ng lg/ml f SE of duplicate determinations in ELISA.
International units ( I V ) determined in RFFIT. Essentially the same results were
obtained in a mwse protection test.
a
I
i
111
PBL
3
c s-"
"&""d.
0
7
21
42
90
days after immunization
280
of eght
Figure4.The in vivokinetics of invitroanti-rabiesvirusresponse
individuals.Upper panel; 0,rabies virus-specific non-antigen-induced
IgM antibody;
0, rabies virus-specific DKCV-induced IgM antibody. Lower panel;0,rabiis virusspecific non-antigen-induced IgG antibody; 0 , rabies virus-spectfic DKCV-induced
IgG antibody. Data are expressedas mean responses of eight individuals in ng lg/
ml f SE.
Figure5 Tcell dependency of anti-rabiesantibodyproduction. HumanPBL
were separated in Tand non-T cells.1 x 10' non-T cells were culturedwith various
numbers of T cells and in
the presence of0.1 pg DKCV/culture. Two representative
experiments with two different donors are shown. Antibody responses
of separated
reconstituted cells and unseparatedPEL are expressed as mean ng lg/mlk SE of
triplite determinations.
however,lackeddetectableamountsofneutralizingantibody
activity. In the sera of all subjects, neutralizing antibodies were
found (Table 11).
Cellular requirement. To investigate whether there was a T
response was also seen
in the absence of antigen. This phenomin vitro,
celldependencyofthehumananti-rabiesresponse
enon of nonantigen-induced rabies virus-specificIgM production human PBL wereseparatedinto
T cellsandnon-Tcellsby
was markedly expressed in cultures of PBL of most individuals rosetting of T cells with neuraminidase-treated SE. The T cells
2 wk after
when tested 3 wk after primary immunization, i.e.,
andnon-T cells obtainedwereculturedwithrabiesvaccine
it was no longer detectable13 wk after separately or were reconstituted at various T to B ratios. As
repeat immunization, and
primary immunization (Fig.4). When tested 40 wk after primary shown in Figure 5, neither T cells nor non-T cells alone responded
immunization, PBL of mostindividualsshowedadecreased
after stimulation with rabies vaccine, whereas combined Tcells
antigen-inducible specific IgM response, whereas PBL of other and non-Tcells responded almost as well as unseparated
PBL.
individualsbecameunresponsive.Thisphenomenonoccurred
Maximal anti-rabies antibody formation by 1 x IO6 non-T cells
irrespective of persistent levels of circulating neutralizing antiof 0.75 x 1O6 to 1.25 x 1O6 T cells.
was obtained by the addition
bodies. Cells from other subjects, however, couldbe stimulated Theseresultsindicatethattheanti-rabiesresponseisTcell
by DKCV to produce specific IGM antibodyin vitro up to 40 wk dependent. When more than 30% monocytes were present in
after the primary injection (data not shown).
cultures, specific antibody production decreased. To standardize
Low levels of rabies virus-specific IgG antibody were found in the system, monocytes were always removed by adherence to
cultures of PBL of some subjects as early as 1 wk after primary plastic, and the percentage of monocytes in cultures was adof all but one individual could
be triggered justed to 5 to 10% (data not shown).
immunization. The cells
by antigento produce rabies virus-specific IgG antibody between
2 and 40 wk after the first booster immunization (Fig. 4). The
DISCUSSION
magnitude of those responses was highly variable. PBL of one
Thus far studies on the
host immune responseto viruses have
subject could not be stimulated by rabies virus to produce IgG
antibody despite the ability of those cells
to secrete specific IgM been performed largely in experimental animals. In the present
spontaneously,aswellasinresponse
to antigen(datanot
paper, we have described a simple and reproducible
culture and
shown).
assay system for the specific stimulation of unselected human
Neutralizingantibodyactivity in culture supernatantfluids.
PBL with rabies virus and for the subsequent detection of rabies
Culture supernatant fluids containing antigen-binding rabies vi- virus-specific antibodies in culture supernatant fluids by ELISA.
rus-specific antibodies shown by ELISA were examined for the Some investigators have recently reported the successful actipresenceofneutralizingantibodies
to rabiesvirus. Inmost
vation of human B lymphocytes in vitro with influenza virus(16,
supernatant fluids that were tested, neutralizing antibodies
to 17) and varicella zoster virus(18) as well as the measurement of
rabies virus couldbe demonstrated. Some culture supernatants, virus-specific antibodiesin culture supernatant fluids.
1238
IN VlTRO RABIES-SPECIFIC
ANTIBODY
PRODUCTION
In these systems, however, as well as in most other human have been shown
to be only demonstrable
in circulation between
models described thus far, PBL were taken from individuals who
days 5 and 12 after booster immunization
(27). The time intervals
had been previously primed
for the antigens that have been used
for collecting blood samples in our study were 2 or 3 wk. As a
to trigger antibody responsesin vitro judged by serum antibody consequence, the phenomenon of spontaneous IgG production
(14, 15, 24, 26). As a consequence it is reasonable to assume by short-term memory cells may not have been detected in the
present study.
that reexposure to these antigens may occur at any moment
during a lifetime. Because the kinetics of the circulating B cell
The antibody response observed here requires T cell help, as
repertoireisgreatlyinfluenceduponcontactoftheimmune
wellasthepresenceofantigen
intheinductionphase.This
system with these recall antigens (24, 27-29), those systems observation is compatible with results obtained in similar systems
may not be themost ideal toolsto study B cell activation in man. in which influenza virus and varicella zoster virus were used as
inducingantigens (16-18). TherewasadiminutioninthereIn the present study a vaccine against rabies virus has been
used to trigger anin vitro immune responseof PBL from subjects sponse to rabies virus at high antigen concentrations in culture.
after in vivo primary sensitization with the same vaccine. RabiesAlthough the induction or activation of suppressor T cells could
be an explanation for this phenomenon, the tolerizationcells
of B
is usually always fatal unless immediate local wound treatment
is done and postexposure prophylaxis is administered. Conse by antigen cannot be excluded (1 1, 33-36). The T cell dependquently this virus, apart from being of biologic importance, mustency of the anti-rabies antibody response has also been docube considered to be a new antigen for most individuals. Indeed, mented in experimental animals by using nude mice and cyclorabies virus serum antibodies havenot been demonstrated thus phosphamide-treated animals and by adoptive transfer studies
far in individuals who havenot experienced vaccination or infec- (37-40). The role of T cells in the regulation of the human antirabies responsein vitro is currently under study.
tion and postexposure treatment.
This study clearly demonstrates that
in vitro synthesis of rabies A number of explanations may be suggested regarding the
virus-specific IgM antibody triggered by DKCV in vitro requires finding that in culture supernatant fluids of PBL of some individuals, apparently only nonneutralizing anti-rabies virus antibodies
primaryimmunization.Therabiesvirus-specificIgMantibody
were detected.
response induced by DKCV in vitro seems to follow the same
First, theELSA must be consideredto be more sensitive than
pattern of responsiveness as has been found for anti-KLH IgM
RFFIT. In 10- to 100-fold concentrated PBL culture supernatant
antibody response of PBL of immunized subjects induced by
KLH in vitro (11). In contrast, however, to the results reported fluids of those individuals, however, neutralizing antibodies could
not be found.
by Lane et al. (II), unresponsiveness to antigen after a 4-wk
Another explanation for the lack of neutralizing antibodies in
booster immunization was not observed in the course
of a rabies
virus-specific IgM antibody response. When followed over time, culture supernatant fluids of PBL of some individuals may be
rabies virus-specific IgM antibody was produced
in vitro by PBL that B cells responsible for the production of neutralizing antiof most individuals at a rather constant quantity until at least
12 bodies were not present in the circulation or had become unresponsive to antigen exposure in vitro. The presence of neutralwk after the first booster immunization.
Thetemporaryappearanceinthecirculationof
cells that izing antibodies in sera of the subjects in question need not be
secrete rabies virus-specific antibody without in vitro exposure in contradiction with this explanation.
Kinetics studies on the appearance of circulating memory B
to DKCV may reflect the presence ofin vivo primed rabies viruscellsforthedifferentproteinsofrabiesviruscombined
with
reactive cells that in vitro can be induced easily by nonspecific
factors generated in the culture system to differentiate to anti- precursor frequency analysis of the respective subsets are in
progress. Furthermore, the use of PBA, as well as subunits of
body-secreting cells (30). Another possibility may be that this
transient phenomenon represents the appearance in the circu- rabies virus(41-45), for in vitro stimulation of lymphoid cells from
lation of a subset of antigen-reactive B cells that spontaneously different organ sources may help to clarify this intriguing observation.
synthesize specific IgM antibody without the need for an addiThe stimulation of human PBL with this biologically relevant
tionalsignal in vitro. Fromthe kinetics ofappearance in the
antigen thatis also applicablein vivo is potentially a valuable
tool
circulation, as well as the differential requirements for an antigenic stimulusin vitro to differentiate to antibody-secreting cells, for the dissection of the function of human T, as well as B cells,
it cannot be deduced at present whether spontaneously
IgM- in the antibody response to rabies virus. In addition, this techfor the
secreting cells and in vitro DKCV-induced IgM-producing cells nique, using a simple and highly reproducible method
detection of specific antibody, may prove to be a Useful system
represent different maturational stages of the same B cell subset
or if they developed independently. With regard
to the declineof for potency testing of vaccinesin vitro.
the in vitro DKCV-triggered specific IgM response between 6
and 13 wk after primary immunization, the depletion of circulating
Acknowledgments. The authors are greatly indebted to Mrs.
long-term memory cells soon
after in vivo immunization has been A, M.A.vanRijn
fortypingthemanuscript.Thetechnical
assistance of Mrs. H. B. M. de Koning,A. H. Hazendonk, and E.
reported in rat and mouse (31, 32). The assumption that antirabies virus IgM-producing cells represent long-term memory B H. Weitering is gratefully acknowledged.
cells is supported by the observation that specific IgG production
becomesdetectableafterthefirstboosterimmunizationand
REFERENCES
continuously increases after repeat booster immunization 3 wk 1. Faud, A. S., and R. E.Ballieux, eds. 1979. AntibodyProductionin Man.
later. Saxonet a/. (28) and Stevenset a/.(27), however, reported
vitro Synthesis and Clinical Implications. Academic Press,New YO*.
data to suggest that long-term memory cells may be processed 2. MBller, G., ed. 1979. Activation of antibody synthesis in humanB IymphNyteS.
Immund. Rev. 45:
by lymph nodes to become short-term memory cells that are
3. Fa&, A. s., and R. E. Ballieux. eds. 1982 Human 6 LymphocyteFunction.
Activation and Irnmunocegulation. Raven Press, New York.
largelymphoblastoidB cells spontaneouslysecretingspecific
4. Dosch, H. M., and E. W. Gelfand. 1976. /n vitro induction of hemolytic plaqueIgG.
forming cells in man. J. Immunol. Methods 11:107.
In the present study, spontaneous rabies virus-specific IgG
5. C. J. Heijnen. F. UytdeHaag, H. J. Gmelig-Meyling, and R. E. Ballieux. 1979.
Localizationofhumanantigenspecifichelperandsuppressorfunctionin
productionhas not beenobserved.Short-termmemorycells
PRODUCTION
; ANTIBODY
IN VITRO RABIES-SPECIFIC
1239
25. Smith, J. S., P. A. Yager. and G. M. Baer. 1973. A rapid reproducible test for
distinct T-cell subpopulations. Cell. Immund.43282.
determining rabies neutralizing antibody. Bull.
WHO 48535.
6. Luzzati, A.L.,M. J. Taassig, T. Meo, and B. Pernis. 1976. Induction of an
S. Fauci. 1981. Antigen-induced in vitro
26. Vdkman, P. J., H.C.Lane,andA.
blood lymphocytes. J. Exp.
antibody response in cultures of human peripheral
antibody production in humans: a model for B cell activation and immunoMed. 144:573.
regulation. Proc. Natl. Acad.Sci. USA 78:2528.
7. Misiti, J., and T. A. Waldmann. 1981. In vitro generation of antigen-specific
27. Stevens, R. H.. E. Macy. C. Morrow, and A. Saxon. 1979. Characterization of
hemdytic plaque-formingcells from human peripheral
bloodmononmar cells.
a circulating subpopulation of spontaneous antitetanus toxoid antibody prcJ. Exp. Med. 154:1069.
ducing B cells followingin vitro booster immunization.J. Immund. 122:2498.
8. Delfraissy. J. F., D. Galanaud. J. Dormot, and C. Wallon. 1977. Primary
in vitro
28. Saxon,A., M. A. Tamaroff, C. Morrow, and R. H. Stevens. 1981. Impaired
antibodyresponsefromhumanperipheral
Mood lymphocytes. J. Immund.
generation of spontaneous and mitogen-reactive anti-tetanus
toxoid antibody118:630.
in vivoboosterimmunization.Cell.
producingBcellsfollowingrepetitive
J. Heijnen,K.H.Pot,and
R. E.Ballieux.1981.Antigen9.UytdeHaag,F.,C.
Immunol. 59:82.
speafic human Tc e l l factors. II. T cell suppressor factor: biologic properties.
J. Babbage,and R. L.Souhami.1982.
29.Callard. R. E.,G.W.McLaughan.
J. Immunol. 126:503.
Specific in vitro antibody responses by human blood lymphocytes apparent
10. Morimoto, C.,E. L. Reinhem, and S. F. Schlossman. 1981. Primary in vitro
nonresponsiveness of PBLis due to a lack of recirculatingmemory B cells.J.
anti-KLH antibody formation by peripheral
Mood lymphocytesin man: detection
Immunol. 129153.
with a radioimmunoassay.J. Immunol. 127:514.
S. Fauci.1983.Selectiveactivationofantigen-specific
30.
Peters,
M.,andA.
11. Lane, H. C., D. J. Vdkman. G. Whalen. and A. S. Fauci. 1981.In vitro antigenhuman B cells in recently immunized individuals by nonspecific factorsin the
J. Exp. Med. 154:1043.
induced, antigen-specific antibody production in man.
absence of antigen. J. Immund. 130:678.
12.Ballieux, R. E.,C. J. Heijnen, F. UytdeHaag, and B. J. M. Zeegers. 1979.
31. Rowley, D. A., J. L. Gowans, R. C. Atkins, W. L. Ford, and M. E. Smith. 1972.
Regulation of B cell activityin man: role of T cells. Immund. Rev. 45:3.
The specific selection of recirculating lymphocytes by antigen in normal and
13. Geha, R. S., F. Mudawar, and E. Schneeberger. 1977. The specificity of T cell
preimmunized rats.J. Exp. Med. 136:499.
helper factor in man. J. Exp. Med. 1451436.
14. Brenner, M. K., andA. J. Munro. 1981. Human anti-tetanus antibody response 32. Sprent, J., J. F. A. P. Miller, and G. F. Mitchell. 1971.Antigeninduced selective
recruitment of circulating lymphocytes.C e l l . Immunol. 2:171.
in vitro: autologous and allogeneic T cells provide help by different routes.
Clin. Exp. Immunol. 46171.
33. UytdeHaag, F., C. J. Heijnen, and R. E. Ballieux. 1978. Induction of a n t i i n 15. Volkman. D. J., S. P.Allyn,andA.
specific humansuppressor T lymphocytesin vitro. Nature 271556.
S. Fauci. 1982. Antigeninduced in vitro
antibody production in humans: tetanus toxoid-specific antibody synthesis.J.
3 4 . Shore, A., H. M.Bosch, and E.Gelfand. 1978. Induction and antigen separation
Immunol. 129:107.
cells in man. Nature274598.
of antigendependent T helper and T suppressor
35. Heijnen. C. J., F. UytdeHaag, C. H. Pot, and R. E. Ballieux. 1979. Feedback
16. Callard. R. E. 1979. Specific in vitro antibody response to influenza virus by
inhibition by human primed T helper cells. Nature 280:589.
human Mood lymphocytes. Nature 282:734.
36.Stevens, R. H., E. Benveniste. and E.Pineda.1982.Theselectiveroleof
17. Yarchoan, R., 8. R. Murphy, W. Strober, H. S. Schneider, and D. L. Nelson.
1981.Specificanti-influenza virus antibodyproduction in vitro by human
membrane IgG in the antigeninduced inhibition of human in vitro antibody
peripheral Mood mononuclear cells. J. Irnrnund. 127:2588.
synthesis. J. Immunol. 128398.
18. Souhami, R. L., J. Babbage, and R. E. Callard. 1981. Specificin vitro antibody 37. Prabhakar, B. S., and N. Nathanson. 1981. Acute rabies death mediated by
response to varicella zoster. Clin. Exp. Immund. 46:98.
antibody. Nature 290590.
19. Steenis, G. van, A. L. van Wezel, P. van der Marel, and C. A. Hannik. 1981.
in the central
38. Smith, J. S. 1981. Mouse model for abortive rabies infection
Dog kidney c e l l rabies vaccine: some aspects of its control and efficacy in
nervous system. Infect. Immun.31297.
in Man.
man. In CellCultureRabiesVaccinesandTheirProtectiveEffect
39. Mifune, K., E. Takeuchi, P. A. Napiorkowski, A. Yamada, and
K. Sakamoto.
T. J. Wiktor, and H. KcProceedings of WHO Consultation. E. K. Kurvert.
1981. Essential rde of T cells in the postexposure prophylaxis of rabies in
prowski. eds. International GreenCross, Geneva. P. 72.
mice. Microbiol. Immund. 25:895.
20. van Wezel, A. L., G. van Steenis, C. A. Hannik, and H. Cohen. 1978. A new
40. Prabhakar, 8. S., H. R. Fischman.andN.Nathanson.1981.Recovery
from
approach to the production of concentrated and purified inactivated polio and
experimental rabies by adoptive transfer of immune
cells. J. Gen. Virol. 5625.
rabies tissue culture vaccine. Dev.Bid. Stand. 41:159.
41. Flamand, A., T. J. Wiktor, and H. Koprowski. 1980. Use of hybridoma moncR. E. Callard. 1981. A micromethod for the
21. Zanders, E. D., C. M. Smith, and
clonal antibodiiesin the detection of antigenic differences between rabies and
induction and assay of specific
in vitro antibody responses by human lymphorabies-related virus proteins.I. The nudeocapsid protein.J. Gen. Vird. 48:97.
cytes. J. Immund. Methods47:333.
42. Fiamand, A.,T. H. Wiktor, and H. Koprowski. 1980. Use of hybridoma mono22. Hagenaars, A. M., A. J. Kuipers, and J. Nagel. 1980. Preparation of enzyme
donal antibodies in the detectionof antigenic differences between rabies and
antibodyconjugates. In Developments in ClinicalBiochemistry. Immune
rabieWelated virus proteins.II. The glycoprotein.J. Gen. Virol. 48:105.
Enzymatic Assay techniques. R. Malvano, ed. Nijhoff. The Hague, The Neth43. Cox, J. H., B. Dietzschold, and L. G. Schneider. 1977. Rabies virus glycoprcerlands. 16.
16754.
tein. II Biological and serological characterization. Infect. Immun.
23. Bos, E. S., A.A.vander
Doden. N.van Rooij, and A.H.W.
M. Schuurs. 44. Wiktor, T. J., E.GyMgy,H. D. Schlumberger,F. Sokol, andH. Koprowski.
1981. 3,3'.5,5'-Tetramethylbenzidine as an Ames test negative chromogen
J. Immund. 110:269.
1973. Antigenic properties of rabies virus components.
for horseradish peroxidase in enzyme immunoassay.
J. Immunoassay 2:187.
S. Fame, and J. Sisman. 1974. Extraction
45. Atanasiu, P., H. Tsiang. P. Perrin,
24. Stevens, R. H.. and A. Saxon. 1978. lmmunoregulation in humans. Control of
d'un antigbe soluble (glycoprotein) par le Triton X-100 A partir d'un vaccin
antitetanus toxoid antibody production after booster immunization. J. Clin.
antirabique de culture tissulaire de premier explant. RBsultats d'immunisation
Invest. 62:1154.
et pouvoir protecteur. Ann.Microbid. 1258539.