Download Enhancement of Murine Lymphokine-activated Killer Cell Activity by

(CANCER RESEARCH 50, 3013-3018, May 15, 1990]
Enhancement of Murine Lymphokine-activated
Retinole Acid1
Killer Cell Activity by
Tsung-hsing Lin and T. Ming Chu2
Department of Diagnostic Immunology Research and Biochemistry, Rostrell Park Memorial Institute, Buffalo, New York 14263
ABSTRACT
Since retinólas have been suggested to be capable of potentiating
immunity, the present study was undertaken to determine the effect, if
any, on lymphokine-activated killer (LAK) cell activity by retinoic acid,
an active metabolite of vitamin A and a differentiation enhancer. Retinoic
acid alone was shown to induce no cytotoxicity generated from nylon
wool-treated nonadherent murine (BALB/c) splenocytes against natural
killer-resistant, LAK-sensitive syngeneic target tumor cells. When com
bined with human recombinant interleukin-2 (IL-2), retinoic acid aug
mented LAK cell activity in both a dose- and time-dependent manner.
The augmentation was detected at IO"10M retinoic acid and reached the
maximum at IO M, a >200% increase in lytic activity. Kinetic study
revealed that retinoic acid augmented significantly LAK cell activity
when incubated in IL-2-containing culture as short as for 6 h before
cytotoxicity was measured. The removal of retinoic acid from culture
resulted in the loss of the augmentation. Retinoic acid was found to
augment LAK cell activity in a wide range of IL-2 concentrations (75012,000 11 ml ), even at 6,000 11 /nil where the maximal induction of LAK
cell activity had been reached. No phenotype or proliferation of LAK
cells was altered by the addition of retinoic acid to IL-2-containing
culture. However, cellular serine protease activity, measured as N-abenzyloxycarbonyl-i.-lysine thiobenzyl-esterase, in LAK cells was in
creased by retinoic acid also in a dose- and time-dependent manner. The
increase in LAK cellular serine protease activity was significantly corre
lated with that of augmented LAK cell activity. Overall these results
demonstrated that IL-2-induced LAK cell activity was enhanced by
retinoic acid and that the augmentation may be mediated by means of
enhanced expression of cellular serine protease activity. This study also
suggests that, in addition to its use in chemoprevention of cancer, retinoic
acid is of potential in adoptive immunotherapy.
INTRODUCTION
Retinoids, namely, natural and synthetic analogues of vita
min A, have been shown to exhibit a variety of diverse effects
on the growth and differentiation of both normal and trans
formed cells (1). Evidence available has demonstrated that
retinoids inhibit the development of preneoplastic transforma
tion and decrease the formation of spontaneously developed,
chemically induced, and virally induced animal tumors (2-5).
The chemopreventive effect of retinoids on carcinogenesis in
animal models has been suggested to result from its direct
growth inhibitory and/or differentiation-promoting
action on
tumor cells and is attributed in part to the capacity of retinoids
to potentiate the immunity in the hosts (6, 7). Mice treated
with retinoids have been shown to possess an enhanced hostvmuf-graft reactivity and an increased resistance to tumor
transplantation (8, 9). An enhancement in both in vitro and in
vivo killer T-lymphocyte function and cell-mediated cytotoxicity
has been shown as a result of retinoid treatment (9-11 ). In spite
Received 12/18/89.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
'Supported in part by a research grant. ROI CA-37646. awarded by the
National Cancer Institute, and IM-416 from the American Cancer Society.
' To whom requests for reprints should be addressed, at Diagnostic Immunol
ogy Research and Biochemistry Department. Roswell Park Memorial Institute.
Elm and Carlton Streets. Buffalo, NY 14263.
of the extensive investigation on LAK3 in recent years, no
information regarding retinoids and IL-2-induced LAK cell
activity is yet available. In the present study, we have investi
gated the effect of retinoic acid, an active metabolite of vitamin
A and a differentiation-enhancing agent, on the generation of
murine LAK cell activity.
We have established a spontaneously developed, syngeneic
murine mammary adenocarcinoma cell line, designated JC,
which is resistant to NK cell but susceptible to LAK cell lysis
(12, 13). Because it is weakly immunogenic and highly tumorigenic, the major biological characteristics mimicking that of
human disease, JC is a more suitable model for experimental
immunotherapy than other chemically or virally induced mam
mary tumor models. Using JC as the target, we have demon
strated that retinoic acid was capable of exerting a positive
effect in both a dose- and time-dependent manner on the
generation of LAK cell activity from syngeneic BALB/c mouse
splenocytes and that the enhanced LAK cell activity was cor
related significantly with the increase in cellular serine protease
activity of LAK cells. These results also suggest that, in addition
to its possible use in chemoprevention of cancer, retinoic acid,
and retinoids in general, may be of potential in adoptive im
munotherapy.
MATERIALS
AND METHODS
Recombinant IL-2. Human recombinant IL-2, produced from Escherichia coli, was kindly supplied by Cetus Corporation (Emeryville, CA)
with a specific activity of 18 x 10' IU/mg. Endotoxin level was <0.01
ng/ml. The purity was >98% as examined by sodium dodecyl sulfatepolyacrylamide gel electrophoresis.
Mice. Eight-12-week-old BALB/c female mice were used. The
BALB/c mice were supplied by our institute's West Seneca Laboratory
(West Seneca, NY). Histocompatibility testing by skin grafting was
performed routinely to ascertain the maintenance of the line.
Target Cell. The JC tumor line was established from a spontaneously
developed mammary adenocarcinoma in an aged virgin female BALB/
c mouse (12). JC is an NK-resistant cell line but susceptible to LAK
cell lysis (13). JC was maintained in RPMI 1640 with 10% fetal calf
serum (GIBCO, Grand Island, NY).
Generation of LAK Cells. The generation of LAK cells had been
described previously (14). Briefly, the spleens from BALB/c mice were
harvested asceptically and minced in a stainless steel mesh to produce
a single cell suspension. The erythrocytes were lysed osmotically with
warm Tris-ammonium chloride buffer, 0.17 M, pH 7.2. The remaining
splenocytes were passed through a column of nylon wool (Cellular
Products, Buffalo, NY), and the nonadherent cells were incubated at
37°Cin complete medium containing IL-2 under a moist atmosphere
with 5% CÛ2in culture flasks or 6-well plates (Falcon, Lincoln Park,
NJ) at a cell concentration of 1 x 106/ml. The complete medium
contained RPMI 1640 with 10% fetal calf serum, 0.1 HIMnonessential
amino acids, 0.1 HIMsodium pyruvate, 5 x 10~5M 2-mercaptoethanol,
0.3 mg L-glutamine/ml, 100 nf>streptomycin/ml, and 100 units peni
cillin/ml.
Generation of LAK Cells in the Presence of Retinoic Acid. The nylon
wool-treated nonadherent splenocytes were prepared as described
3The abbreviations used are: LAK, lymphokine-activated killer; IL-2. interleu
kin-2: NK. natural killer: DMSO. dimethyl sulfoxide; BLT, A'-n-benzyloxycarbonyl-L-lysine thiobenzyl: PBS. phosphate-buffered saline.
3013
Downloaded from cancerres.aacrjournals.org on April 14, 2017. © 1990 American Association for Cancer Research.
AUGMENTATION
OF MURINE LAK ACTIVITY BY RETINOIC ACID
above. Retinole acid (tf-M-trans, Sigma, St. Louis, MO) was dissolved
in DMSO (Sigma) to make a stock solution of 10~2 M. The stock
solution was kept at —
80°Cuntil use. The retinole acid solution was
used in combination with IL-2 at a final concentration ranging from
10~6-10~'°
M in complete medium. After incubation for various periods
of time, the activated splenocytes were harvested, washed, and used as
the effector cells in the "Cr release assay.
"Cr Release Assay. LAK cell activity was measured by using a
standard 4-h "Cr release microcytotoxicity assay (14). The target cells,
JC (2 x IO6),were harvested and washed twice and then incubated with
100 MCiof "Cr (Na2"CrO4, 200-900 Ci/g; ICN, Costa Mesa, CA) for
1 h at 37°Cin a moist atmosphere with 5% CO2. The labeled cells were
used as the target for the "Cr release assay. Each cytotoxic assay was
carried out in triplicate, and the results were expressed as lytic units,
which were calculated from linear regression plotting from various
effector to target cell ratios as described previously ( 15). One lytic unit
was defined as the number of effector cells required to cause 30%
specific "Cr release from IO4 target cells. Statistical analysis was
performed with Student's / test.
Serine Protease Activity Assay. BLT-esterase was assayed essentially
according to the method of Pasternack and Eisen (16). Briefly, LAK
cells were washed 3 times in PBS, counted, and then lysed by incubating
(1-2) x 10' cells/ml in PBS containing 0.1% Triton X-100 (Sigma) for
30 min in ice with frequent vortexing. Esterase activity in the lysates
was measured by adding 300 n\ of a reaction mixture of 2 x IO"4 M
BLT (Sigma)-2.2 x IO"4M dithiobis(nitrobenzoic acid) (Sigma) in PBS,
pH 7.4, to 40 p\ of appropriate dilutions of cell extracts in Triton X100-PBS in each well of flat-bottomed 96-well plates (Dynatech Lab,
Alexandria, VA). After 30 min at 37°C,the absorbance at 405 nm was
measured in an enzyme-linked immunosorbant assay reader (Model EL
309; Bio-Tek, Winooski, VT) using 40 ^1 of Triton X-100 in PBS plus
300 n\ of the reaction mixture as a blank. One unit of BLT-esterase
activity was defined as that causing the formation of absorbance of l.O/
10" cells.
Immunofluorescence Examination. Rat monoclonal anti-Thy-1.2 an
tibody and rat monoclonal anti-Lyt-2 antibody were purchased from
Becton-Dickinson (Mountain View, CA). Rabbit heterosera against
asialo GMI wraspurchased from Wako Chemical (Dallas, TX). Hybrid
omas producing rat monoclonal antibody GK 1.5 against L3T4 and
7D4 against IL-2 receptor were obtained from American Type Culture
Collection (Rockville, MD). Analysis of cell surface markers was carried
out by indirect immunofluorescence (15, 17). Single cell suspensions ( 1
x IO6) were incubated for 45 min at 4°Cwith 100 M' of primary
antibody reagent diluted 50 times in PBS containing 0.5% bovine serum
albumin and 0.1 % NaN3. After washing, the cells were incubated further
with 100 n\ of affinity-purified fluorescein isothiocyanate-conjugated
goat anti-rat IgG (heavy and light chain) against anti-Thy-1.2, Lyt-2,
L3T4, and IL-2 receptor (diluted 50 times; Jackson Immuno-Research,
West Grove. PA) and goat anti-rabbit IgG fluorescein isothiocyanateconjugated against anti-asialo ( >-,, (diluted 50 times: Sigma) in PBS,
containing 0.5% bovine serum albumin and 0.1% NaN.,. Fluorescence
analysis of stained cells was performed with a FACSTAR PLUS flow
cytometer (Becton-Dickinson).
(mean ±SEM) and 504 ±32 lytic units, respectively, in
comparison with IL-2 alone, 224 ±24 lytic units (P < 0.01).
To assess whether or not retinoic acid alone caused any effect
on cytotoxicity of the splenocytes, nylon wool-treated nonadherent splenocytes were cultured in complete medium with
various concentrations of retinoic acid, again ranging from 1 x
10~'°to 1 x 10~6M for 3 days. No cytotoxicity against the NKresistant, LAK-sensitive JC tumor cells was detected (data not
shown). Also noted was that DMSO, the carrier of retinoic acid
in solution, alone or when combined with IL-2 exhibited no
induction of or effect on LAK cell activity at 0.01 % in complete
medium, the highest concentration used in the present study.
In the rest of the experiments reported in this study, appropriate
dilution of DMSO in complete medium was always included as
a control.
Retinoic acid at the concentrations of 1 x 10~7and 1 x 10~9
M were selected for further experiments. Retinoic acid was
added to complete medium containing IL-2 (3000 ID/ml) at
the beginning of culture, and LAK cell cytotoxicity was meas
ured daily for 5 consecutive days. LAK cell activity generated
from the culture containing IL-2 plus retinoic acid always was
found to be greater than that generated from IL-2 alone. As
shown in Fig. 2, the differences between LAK cell activities
generated from the culture in IL-2 plus retinoic acid at 1 x
10~7and 1 x 10-" M and in IL-2 alone for 1 day (33 ±8, 18 ±
6 versus 12 ±5 lytic units), 2 days (117 ±15, 75 ±11 versus
58 ±7 lytic units), 3 days (457 ±28, 308 ±20 versus 199 ±26
lytic units), 4 days (1112 ±112, 819 ±50 versus 599 ±33 lytic
units), and 5 days (1203 ±91, 918 ±55 versus 712 ±27 lytic
units) were all statistically significant (P < 0.05). Fig. 2 also
reveals that augmentation of LAK activity by retinoic acid was
in both a time- and dose-dependent manner.
Augmentation by Retinoic Acid of LAK Cell Activity Induced
with Various Concentrations of IL-2. Induction of LAK cell
activity has been shown to be IL-2 dose dependent and to reach
the maximum at high concentration of IL-2 (14, 15, 18). In
this series of experiments, we determined whether or not reti
noic acid could augment further the maximum LAK cell activity
that had been induced by IL-2. Nylon wool-treated nonadherent
mouse splenocytes were cultured in complete medium contain
ing 1 x IO"7 M of retinoic acid plus various concentrations of
IL-2 for 3 days, and the LAK cell activity was measured. As
shown in Fig. 3, the generation of IL-2-induced LAK cell
activity was IL-2 dose dependent and reached its maximum
activity at around IL-2 6,000 and 12,000 lU/ml. At 6,000 IU/
500 •
RESULTS
Effect on IL-2-Induced Murine LAK Cell Activity by Various
Concentrations of Retinoic Acid. The splenocyte preparations
from BALB/c mice were treated with nylon wool and the
nonadherent cells, which were devoid of macrophages and Blymphocytes, were cultured in complete medium containing
either IL-2 (3000 lU/ml) alone or IL-2 plus various concentra
tions of retinoic acid for 3 days. As shown Fig. 1, the addition
of retinoic acid to IL-2-containing culture resulted in an aug
mentation of LAK cell activity, and the augmentation was
retinoic acid dose dependent. The more retinoic acid added to
the IL-2-containing culture, the greater was the LAK cell activ
ity; especially retinoic acid at 1 x 10~7and 1 x 10~6 M plus IL2 which augmented LAK cell activity to 482 ±46 lytic units
300
200
O
IO'10 IO'9
IO'8
IO'7
IO'6
Retinole acid ( M )
Fig. I. The effect of various concentrations of retinoic acid, i.e.. dose depend
ent, on IL-2-induced LAK cell activity. LAK cells were generated from culturing
nylon wool-treated nonadherent BALB/c mouse splenocytes (1 x 10*/ml) with
IL-2 (3000 lU/ml) for 3 days. Retinoic acid at the various concentrations as
indicated was added to IL-2-containing complete medium at the beginning of
culture. The LAK cell activity was determined and expressed in lytic units. Points.
means of triplicate assays: bars. SEM.
3014
Downloaded from cancerres.aacrjournals.org on April 14, 2017. © 1990 American Association for Cancer Research.
AUGMENTATION
1500
OF MURINE LAR ACTIVITY BY RETINOIC ACID
•: .
B IL-2 +R*tif>oic«id,io"
D IL-2 +Retinoleacid, 10"
1
2345
Day
Fig. 2. The time-dependent manner of augmentation on LAK cell activity by
retinoic acid. Nylon wool-treated nonadherent BALB/c mouse splenocytes (1 x
106/ml) were prepared and cultured with IL-2 (3000 lU/ml) and IL-2 plus retinoic
acid at 2 concentrations (1 x 10~7 and 1 X IO"9 M). LAK cell activity was
determined daily for 5 consecutive days and expressed in lytic units. Columns.
means of triplicate assays; bars, SEM.
plete medium at the beginning of culture, then removed 2 days
later, the LAK cell activity was measured. The remaining LAK
cells in both retinoic acid-treated and the control (IL-2 alone)
groups were washed. The cells from the control group (A) were
added with fresh complete medium containing IL-2; while those
from the retinoic acid-treated group were divided into 2 groups;
one group (B) was continuously cultured in IL-2 alone and the
other (C) was continuously cultured in IL-2 plus retinoic acid
(Fig. 5). LAK cell activities were obtained daily from these 3
groups for the next 3 days. As shown in Fig. 5, the group (C)
cultured in complete medium supplemented with IL-2 plus
retinoic acid exhibited the highest induction rate of LAK cell
activity among the 3 groups. Although quantitatively the ini
tially retinoic acid-treated/removed group (B) always generated
a greater LAK cell activity than that of the control (A), the
induction rates (slopes) of LAK cell activity were parallel be
tween A and B groups, indicating that the removal of retinoic
acid from complete medium containing IL-2 resulted in the loss
of augmentation in LAK cell activity.
Effect of Retinoic Acid on Proliferation of LAK Cells. Nylon
wool-treated nonadherent splenocytes (1 x 106/rnl) were cul
tured in complete medium alone, complete medium containing
400
750
1500
3000
6000
300
12000
IL-2 ( Ill/ml )
Fig. 3. The enhancement of LAK cell activity induced with various concentra
tions of IL-2 by retinoic acid. LAK cell activity was generated from nylon wooltreated nonadherent BALB/c mouse splenocytes (1 x 10'/ml) by incubation with
various concentrations of IL-2 as indicated plus retinoic acid (1 x 10~7 M) for 3
200
100
days. The LAK cell activity was determined and expressed in lytic units. Columns,
means of triplicate assays; bars, SEM.
ml, the LAK cell activity generated by IL-2 alone, 461 ±47
lytic units, was further augmented to 882 ±86 lytic units by
the addition of retinoic acid to IL-2. An enhancement also was
detected at 12,000 lU/ml of IL-2 by retinoic acid (492 ±47
versus 910 ±68 lytic units).
Also shown in Fig. 3 is that retinoic acid enhanced the LAK
cell activities induced from a whole spectrum of IL-2 concen
trations examined. The differences between LAK cell activities
induced by IL-2 alone and that augmented by retinoic acid were
all statistically significant (P < 0.05). The augmentation on
LAK cell activities by retinoic acid averaged 200% and ranged
from 185% for 12,000 ID/ml to 229% for 1,500 lU/ml.
Kinetic Analysis of Augmentation of LAK Cell Activity by
Retinoic Acid. Fig. 4 illustrates the results obtained from the
addition of retinoic acid (1 x 10~7 M) to complete medium
containing IL-2 (3000 lU/ml) at various time intervals (3, 6,
12, 24, 48, and 72 h) before the completion of a 3-day culture;
i.e., 3, 6,... 72 h before the 5lCr cytotoxicity assay, retinoic
acid was added to the IL-2-containing medium and cocultured
for 3, 6,.,. 72 h. As shown, culture of retinoic acid with IL-2
as short as only for 6 h, i.e., by adding retinoic acid 6 h prior
to the 5lCr release assay for LAK cell activity, resulted in a
significant increase in LAK cell activity (257 ±24 versus 188
±22 lytic units, P < 0.05). The greatest enhancement of LAK
cell activity was detected when retinoic acid was added to the
medium simultaneously along with IL-2 at the beginning of
culture.
In another series of experiments, in which retinoic acid (1 X
10~7 M) was added to the IL-2 (3000 IU/ml)-containing com
0
12 24
48
72
Retinoic acid in IL-2 culture ( hr )
Fig. 4. The effect of the length of incubation time with retinoic acid on LAK
cell activity. LAK cell activity was generated from culturing nylon wool-treated
nonadherent BALB/c mouse splenocytes (1 X 10'/ml) with IL-2 (3000 ID/ml)
for 3 days. Retinoic acid (1 x 10~7M) was added to the IL-2-containing culture
at the time indicated prior to the end of 3 days of culture. The LAK cell activity
was determined and expressed in lytic units. Control indicates the addition of
DMSO (0.001%) alone, the solvent used in preparation of retinoic acid solution.
Points, means of quadruplicate assays; bars, SEM.
BOO-
600 -
200 -
Doy
Fig. 5. The effect of the removal of retinoic acid on LAK cell activity. Nylon
wool-treated nonadherent BALB/c mouse splenocytes (I x lO'/ml) were cultured
in complete medium containing IL-2 (3000 lU/ml) and IL-2 plus retinoic acid (1
x 10~7M) for 2 days. The LAK activity was measured. The remaining LAK cells
from both IL-2 alone and IL-2 plus retinoic acid cultures were washed with
complete medium 3 times. The retinoic acid-treated LAK cells were then divided
into two groups. One group was cultured in IL-2 alone (B), and the other group
was cultured in IL-2 plus retinoic acid (C). The LAK cell activity from the original
IL-2 culture alone which continued to be cultured in IL-2 (A) and the newly
divided 2 groups (B. IL-2; C, IL-2 plus retinoic acid) from initial retinoic acidtreated/removed group was measured daily for the next 3 days and expressed in
lytic units. Points, means of triplicate assays; bars, SEM; except on day 5 when
only the average of duplicate assays was available.
3015
Downloaded from cancerres.aacrjournals.org on April 14, 2017. © 1990 American Association for Cancer Research.
AUGMENTATION
OF MURINE LAK ACTIVITY BY RETINO1C ACID
retinoic acid (1 x 10~7 M), IL-2 (3000 IU/ml), and IL-2 plus
retinoic acid. Then the cell density was measured daily for 5
consecutive days. As indicated in Table 1 retinoic acid did not
alter the proliferation of LAK cells in comparison with the
control and IL-2.
Effect of Retinoic Acid on Phenotype of LAK Cells. Cell
surface markers were assessed on LAK cells generated from
culture for 3 days in IL-2 (3000 IU/ml) alone and IL-2 plus
retinoic acid (10~7 and 10~9 M). As summarized in Table 2, in
comparison with the freshly prepared and uncultured nylon
wool-treated nonadherent splenocytes, the culture with IL-2
alone and with IL-2 plus retinoic acid for 3 days resulted in a
decrease in the percentage of cells expressing Lyt-2 and no
change in cells expressing Thy- 1.2 and L3T4 but an increase in
the cells carrying asialo GM, and IL-2 receptor. However, no
significant difference was detected between the phenotyping
obtained from the cells cultured in IL-2 alone and those in IL2 plus retinoic acid in expression of these cell surface markers.
Enhancement of LAK Cellular Serine Protease Activity by
Retinoic Acid. Identical to the experiments described and pre
sented in Figs. 1, 3, and 4, and instead of LAK cell activity in
lytic units, serine protease activity measured as BLT-esterase
activity was obtained from IL-2-induced and retinoic acidaugmented LAK cells. Fig. 6 illustrates that LAK cellular BLTesterase activity was enhanced by retinoic acid in a dose-de
pendent manner. Similar to the augmentation of LAK cell
activity, a detectable increase in LAK cellular BLT-esterase
activity was found at 1 x 10~'°M (10.95 ±0.36 BLT-esterase
activity units, P < 0.05) in comparison with the control (9.57
±0.64 BLT-esterase activity units) and reached the maximum
(15.72± 1.44 BLT-esterase activity units, P < 0.01) at 1 x 10~7
M. DMSO (0.01%) at the highest concentration used in this set
of experiments showed no effect on the enzyme activity.
Retinoic acid at l x 10~7 M was shown to enhance cellular
BLT-esterase activity in LAK cells cultured from a wide range
of IL-2, 750-12,000 IU/ml (Fig. 7). As shown, highly elevated
LAK cellular BLT-esterase activity was induced by IL-2 at
6,000 or 12,000 IU/ml and retinoic acid was still able to
enhance the highly elevated enzyme activity even to a signifi
cantly greater level (6,000 IU/ml, P < 0.05; 12,000 IU/ml,
Kinetic study of the effect on LAK cellular BLT-esterase
activity by retinoic acid is shown in Fig. 8. An apparent increase
in enzyme activity (9.67 versus 9.22 BLT-esterase activity units)
was detected after IL-2-induced LAK cells were incubated with
retinoic acid (1 x 10~7 M) for only 6 h prior to the assay of
enzyme activity. The enhancement in BLT-esterase activity was
also in a time-dependent manner. The longer the LAK cells
were cultured with retinoic acid, the greater was the cellular
BLT-esterase activity expressed.
DISCUSSION
In the present report, we have investigated the effect of
retinoic acid on murine LAK cell activity, using a human-like
murine syngeneic. spontaneously developed, weakly immunogenic, highly tumorigenic, NK-resistant, and LAK-sensitive
mammary adenocarcinoma as the target. Retinoic acid alone
was shown to be incapable of inducing any cytotoxicity from
nylon wool-treated nonadherent BALB/c mouse splenocytes.
However, when used in combination with IL-2, retinoic acid
was shown to augment IL-2-induced LAK cell activity in both
a dose- and time-dependent manner.
Kinetic study demonstrated that a greatly enhanced cytotox
icity was detected when retinoic acid was added to the IL-2containing medium at the early induction phase of LAK cell
activity. The addition of retinoic acid to complete medium
containing IL-2 at the beginning of culture resulted in the
maximal augmentation of LAK cell activity. Although an en
hancement was detectable from an incubation for only 6 h prior
to cytotoxicity assay, a continuous presence of retinoic acid in
culture is required for the maximal augmentation of IL-2induced LAK cell activity. When retinoic acid was removed
from the IL-2-containing culture, the induction rate of LAK
cell activity was decreased to a level identical to that of IL-2
alone, i.e., a loss of the augmentation.
It has been reported that, when fed with vitamin A acetateenriched diet, the splenocytes of Mycobacterium ¿>ov/.v-infected
mice increased in vitro IL-2 production (19). To assess whether
or not the augmentation of LAK cell activity by retinoic acid is
mediated by an increase in IL-2 production in culture, a fixed
and constant concentration of retinoic acid was added to the
culture containing various concentrations of IL-2. The results
indicated that retinoic acid (10~7 M) augmented LAK cell activ
ity by an average of 2-fold over a wide range of IL-2 concentra
tions even at high IL-2 concentrations in which the maximal
induction of LAK cell activity by IL-2 had already been reached.
These data along with the fact that retinoic acid alone induced
no cytotoxicity suggest that the enhanced effect of retinoic acid
on LAK cell activity is not mediated by the induction of an
increased in vitro IL-2 production.
Retinoic acid has been shown to increase IL-2 receptor on
activated human thymocytes but not on peripheral blood lym
phocytes (20). It has been suggested that retinoic acid may
modulate an IL-2-dependent immune response, in part, by upregulating the expression of IL-2 receptor (20). In our study,
we have examined the IL-2 receptor as expressed by LAK cells
cultured in medium containing IL-2 alone and IL-2 plus retinoic
acid. Our results revealed that retinoic acid exhibited little, if
any, effect on the expression of IL-2 receptor. Therefore, the
enhanced effect of LAK cell activity by retinoic acid appears
not to be mediated by an increase in the expression of IL-2
receptor.
The effect of retinoic acid on proliferation of the lymphocytes
also was examined in our study because some investigators have
reported that retinoic acid enhanced lymphocyte proliferation
in response to mitogens (21,22), although others have indicated
no such enhancement of proliferative response (10, 23). Our
results indicated that retinoic acid did not alter the proliferation
Table 1 The effect of retinoic acid on proliferation of l^AK cells
culture30.34
in
in"Complete
Cultured
±0.03*
±0.05
±0.06
±0.11
±0.07
medium
Retinoic acid, IO"7 M
0.53 ±0.09
0.28 ±0.05
0.30 ±0.07
0.45 ±0.08
0.23 ±0.02
IL-2, 3000 IU/ml
0.74 ±0.05
0.96 ±0.07
1.20 ±0.04
1.41 ±0.12
0.70 ±0.05
IL-2 + retinoic acid10.26
0.75 ±0.06Days
0.95 ±0.1240.40
1.17±0.1150.51
1.35 ±0.09
0.67 ±0.0620.27
°Nylon wool-treated nonadherent BALB/c mouse splenocytes (1 x 106/ml) were cultured in complete medium alone, complete medium containing retinoic acid
(1 x 10~7M). IL-2 (3000 IU/ml), and IL-2 plus retinoic acid. The cell density (xlO*/ml) was counted daily for 5 consecutive days by using a cytometer.
* The values represent means ±SEM from triplicate assays.
3016
Downloaded from cancerres.aacrjournals.org on April 14, 2017. © 1990 American Association for Cancer Research.
AUGMENTATION
OF MURINE LAK ACTIVITY BY RETINOIC ACID
results were obtained from the culture with IL-2 plus retinoic
acid, indicating that the combination of IL-2 with retinoic acid
did not increase the IL-2-induced LAK precursor cells, T-cells
and NK-like cells.
Yet, one biochemical parameter, namely BLT-esterase, a
serine protease, activity in IL-2-induced LAK cells was found
to increase profoundly by retinoic acid. Significantly, the in
crease in LAK cellular BLT-esterase activity by retinoic acid
cellsAsialoCultured
% of positive
was correlated with that of the augmented LAK cell activity in
lytic units. This conclusion was based upon the comparison of
inComplete
receptor24564171922±
the results obtained from three different sets of experiments.
medium0IL-2.IL-2IL-2,3000
±82
129± ±27
1±2±4±3
First, both LAK cell activity (Fig. 1) and LAK cellular BLT/ml+ IU
±82
±424 ±24
M-1retinoic acid. 10"'
±84
±223 ±23
esterase
activity (Fig. 6) were augmented by retinoic acid in a
retinoic acid, 10"' MThy-'77
±3L3T425
±3534G37515658M±+±-*-IL-2i
±1.22"212Lyt-245
dose-dependent manner. An increase in both parameters was
°Freshly prepared nylon wool-treated nonadherent splenocytes cultured in
detected at 10~'°M of retinoic acid and reached the maximum
complete medium alone.
at 10~7 M, with a highly significant coefficient of correlation of
'The values represent means ±SEM from triplicate assays.
Table 2 The effect of retinole acid on phenotyping of UK cells
LAK cells were generated from nylon wool-treated nonadherent BALB/c
mouse splenocytes (I x 106/ml). Retinole acid, at the concentrations indicated,
was added to complete medium containing IL-2 (3000 ID/ml) and cultured for 3
days. Then, the cells were incubated with anti-Thy-1.2. Lyt-2. L3T4, asialo GMi.
and IL-2 receptor antibody at 4'C for 45 min. after which the cells were washed
and incubated with affinity-purified fluorescein isothiocyanate-conjugated goat
anti-rat IgG (heavy and light chain) against anti-Thy-1.2. Lyt-2, L3T4. and IL-2
receptor and anti-rabbit IgG fluorescein isothiocyanate against anti-asialo GMi.
Fluorescence analysis was performed with flow cytometry
0.997. Second, when retinoic acid was combined with various
concentrations of IL-2 in culture, IL-2 plus retinoic acid groups
always exhibited a greater LAK cell activity (Fig. 3) and a
greater LAK cellular BLT-esterase activity (Fig. 7) than those
of the IL-2 alone group. A highly correlated association (r =
0.961) was found between these two experiments. Last, but not
least, kinetic analysis revealed that both LAK cell activity (Fig.
4) and LAK cellular BLT-esterase activity (Fig. 8) were aug
mented by retinoic acid in a time-dependent manner. An aug
mentation was detectable from an incubation of only 6 h prior
to 5lCr release and enzyme activity assays, and the maximal
•E
16
12
B
O
IO'10 IO'5
IO'8 IO'7
IO'6
Retinoic acid ( M )
Fig. 6. The effect of retinoic acid on LAK cellular serine protease activity.
LAK cells were generated from eulturing nylon wool-treated nonadherent BALB/
c mouse splenocytes (1 x lO'/ml) with IL-2 (3000 IL'/ml) for 3 days. Retinoic
acid at various concentrations as indicated was added to IL-2-containing complete
medium at the beginning of culture. LAK cellular BLT-esterase activity was
measured at the end of the 3-day culture. Points, means of triplicate assays; bars.
SEM.
20 -
enhancement was found when retinoic acid was added at the
beginning of culture. Again, a significant correlation (r - 0.964)
between LAK cell activity and BLT-esterase activity was found
between the results obtained from these experiments.
Data available in recent years have provided some circum
stantial but compelling evidence in favor of the granule exocytosis pathway as an important mechanism underlying the killer
cell-mediated cytotoxicity (24, 25). One potent lytic protein
called perforin/cytolysin has been identified as a granule com
ponent and shown to damage the target cell membrane by
forming transmembrane lesions (26, 27). Other major constit
uents of cytolytic granules involving in the lytic mechanism of
cell-mediated cytotoxicity include a family of highly related
serine proteases (28, 29). The expression of serine protease
enzymes and its genes has been shown to be induced by IL-2 in
16
0
750
1500
ÃŽOOO
6000
12000
IL-2( lU/ml )
Fig. 7. The effect of retinoic acid on LAK cellular serine protease activity
cultured with various concentrations of IL-2. LAK cells were generated from
nylon wool-treated nonadherent BALB/c mouse splenocytes (1 x 10*/ml) with
various concentrations of IL-2 as indicated plus retinoic acid (1 X IO"7 M) for 3
days. LAK cellular BLT-esterase activity was determined. Columns, means of
triplicate assays: bars, SEM.
profile of nylon wool-treated nonadherent murine splenocytes
in culture containing IL-2 plus retinoic acid in comparison with
IL-2 alone.
When the freshly prepared nylon wool-treated nonadherent
mouse splenocytes were compared, the phenotyping examina
tion using cell surface markers revealed a decrease in percentage
of the cells carrying Lyt-2, no change in the cells carrying Thy1.2 and L3T4, and an increase in the cells expressing IL-2
receptors and asialo GM, in IL-2-containing culture. Similar
r, 12
0
12
24
48
72
Retinoic ecid in IL-2 culture ( hr )
Fig. 8. The effect of the length of incubation time with retinoic acid on LAK
cellular serine protease activity. LAK cells were generated from nylon wooltreated nonadherent BALB/c mouse splenocytes (1 x 10') with IL-2 (3000 IU/
ml). Retinoic acid (1 x 10'' M) was added to the IL-2-containing complete
medium at the time indicated prior to the end of the 3-day culture. LAK cellular
BLT-esterase activity was determined. Control indicates the addition of DMSO
(0.001'V) alone, the solvent used in preparation of retinole acid solution. Points.
averages of duplicate assays.
3017
Downloaded from cancerres.aacrjournals.org on April 14, 2017. © 1990 American Association for Cancer Research.
AUGMENTATION
OF MURINE LAR ACTIVITY BY RETINOIC ACID
murine NK and nonspecific T-killer cells (30). Additionally,
three retinole acid receptors («,/i, and 7) have been identified
in both mouse and human cells and tissues (31). These retinoic
acid receptors are responsible for biological function of retinoic
acid by mediating a mechanism of action similar to that of
steroid/thyroid hormones and their receptors (32, 33). In light
of these reports along with our data on enhancement of LAK
cellular BLT-esterase activity by retinoic acid, it is likely that
an increase in IL-2-induced expression of serine protease activ
ity is the major mechanism underlying the augmentation of IL2-induced LAK cell activity by retinoic acid.
In summary, our study has revealed that retinoic acid, an
active metabolite of vitamin A and a differentiation-enhancing
agent, augmented IL-2-induced LAK cell activity in vitro in
both a dose- and time-dependent manner and was accompanied
by an increase in serine protease activity in LAK cells. Our
results therefore indicate that the augmentation may be me
diated by an increase in IL-2-induced expression of serine
protease activity. This study also suggests that, in addition to
its use in chemoprevention of cancer, retinoic acid and retinoids
may be of potential in IL-2-related immunotherapy.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
ACKNOWLEDGMENTS
21.
We would like to thank D. Sheedy for fluorescence activated cell
sorter analysis, J. Ogledzinski for secretarial assistance, and the De
partment of Laboratory Animal Resources for providing the animals
and their care.
22.
23.
24.
25.
REFERENCES
1. I man R. Effects of vitamin A and its analogs (retinoids) on normal and
neoplastic cells. Biochim. Biophys. Acta, 60S: 33-91, 1980.
2. Chopra, D. P., and Wilkoff, L. J. Inhibition and reversal by fi-retinoic acid
of hyperplasia induced in cultured mouse prostate tissue by 3-methylcholanthrene or .V-methyl-A/'-nitro-A'-nitrosoguanidine. J. Nati. Cancer Inst.. 56:
583-587, 1976.
3. Todaro, G. J.. DeLarco, J. E., and Sporn, M. B. Retinoids block phenotypic
cell transformation produced by sarcoma growth factor. Nature (Lond.). 276:
272-274, 1978.
4. Maiorana. A., and Cullino. P. M. Effect of retimi acetate on the incidence
of mammary carcinomas and hepatomas in mice. J. Nati. Cancer Inst.. 64:
655-660, 1980.
5. Seifter, E., Rettura, G.. Padawer, J.. and Levenson. S. M. Moloney murine
sarcoma virus tumors in CBA/J mice: chemopreventive and chemotherapeutic actions of supplemental fi-carotene. J. Nail. Cancer Inst.. 68: 835-840.
1982.
6. Sporn. M. B., and Roberts, A. B. Role of retinoids in differentiation and
carcinogenesis. Cancer Res.. 43: 3034-3040. 1983.
7. Eccles, S. A. Effects of retinoids on growth and dissemination of malignant
tumors: immunological considerations. Biochem. Pharmacol.. 34: 15991610, 1985.
8. Floersheim, G. L., and Bollag. W. Accelerated rejection of skin homografts
by vitamin A acid. Transplantation (Baltimore). 15: 564-567, 1972.
9. Malkovsky, M., Dorè,C., Hunt, R.. Palmer. L., Chandler, P.. and Medawar.
26.
27.
28.
29.
30.
31.
32.
33.
P. B. Enhancement of specific antitumor immunity in mice fed a diet enriched
in vitamin A acetate. Proc. Nati. Acad. Sci. USA. SO:6322-6326. 1983.
Dennert, G.. and Lotan. R. Effects of retinoic acid on the immune system:
stimulation of T killer cell induction. Eur. J Immunol.. 8: 23-29. 1978.
Lotan. R.. and Dennert. G. Stimulatory effects of vitamin A analogs on
induction of cell-mediated cytotoxicity in vivo. Cancer Res.. 39: 55-58. 1979.
Capone. P. M.. Kadohama. N.. and Chu. T. M. Immunotherapy in a spon
taneously developed murine mammary carcinoma with syngeneic monoclonal
antibody. Cancer Immunol. Immunother.. 25: 93-99. 1987.
Chao, T. Y., and Chu. T. M. Characterization of a new spontaneously
developed murine mammary adenocarinoma in syngeneic BALB/c hosts. In
Vitro Cell. Dev. Biol.. 25: 621-626. 1989.
Chao, T. V.. Ohnishi. H., and Chu. T. M. Augmentation of murine lymphokine (rlL-2)-activated killer cell activity by indomethacin. Mol. Biother.. I:
318-322. 1989.
Ohnishi. H.. Lin, K. M.. and Chu. T. M. Prolongation of serum half-life of
interleukin-2 and augmentation of lymphokine activated killer cell activity
by pepstatin in mouse. Cancer Res.. 50: 1107-1112, 1990.
Pasternack. M. S.. and Eisen. H. N. A novel serine esterase expressed by
cytotoxic T lymphocytes. Nature (Lond.), 314: 743-745, 1985.
Chao, T. Y., and Chu. T. M. Effect of indomethacin on tumor-infiltrating
lymphocytes of a spontaneously developed murine mammary adenocarcinoma. Cancer Immunol. Immunother., 30: 158-164, 1989.
Trinchieri, G., Matsumoto-Kobayashi, M.. Clark, S. C., Seehra, J., London,
L.. and Perussia, B. Response of resting human peripheral blood natural
killer cells to interleukin 2. J. Exp. Med.. 160: 1147-1169. 1984.
Colizzi, V., and Malkovsky. M. Augmentation of interleukin-2 production
and delayed hypersensitivity in mice infected with \fycobacterium hovis and
fed a diet supplemented with vitamin A acetate. Infect. Immun., 48: 581583. 1985.
Sidell. N., and Ramsdell. F. Retinoic acid upregulates interlcukin-2 receptors
on activated human thymocytes. Cell. Immunol., 115: 299-309. 1988.
Abb. J., and Deinhardt. F. Effects of retinoic acid on the human lymphocyte
response to mitogens. Exp. Cell Biol.. 48: 169-179. 1980.
Valone, F. H., and Payan, D. G. Potentiation of mitogen-induced human Tlymphocyle activation by retinoic acid. Cancer Res.. 45: 4128-4131, 1985.
Sidell, N., Rieber. P.. and Golub. S. H. Immunological aspects of retinoids
in humans. Cell. Immunol.. 87: 118-125. 1984.
Henkart, P., and Yue, C. C. The role of cytoplasmic granules in lymphocyte
cytotoxicity. Prog. Allergy. 40: 82-110. 1988.
Zagury. D. Direct analysis of individual killer T cells: susceptibility of target
cells to lysis and secretion of hydrolytic enzymes by CTL. Adv. Exp. Med.
Biol., 146: 149-163, 1982.
Hcnkart, P. A., Yue, C. C., Y'ang, J.. and Rosenberg. S. A. Cytolytic and
biochemical properties of cytoplasmic granules of murine lymphokine-activated killer cells. J. Immunol.. 137: 2611-2617. 1986.
Young, J. D-E., Nathan, C. F.. Podack. E. R.. Palladino, M. A., and Cohn,
Z. A. Functional channel formation associated with cytotoxic T-cell granules.
Proc. Nati. Acad. Sci. USA. 83: 150-154, 1986.
Lavie, G., Leib. Z., and Servadio. C. The mechanism of human NK cell
mediated cytotoxicity: mode of action of surface-associated proteases in the
early stages of the lytic reaction. J. Immunol.. 135: 1470-1476, 1985.
Simon, M. M., Fruth, U., Simon, H. G.. and Kramer, M. D. Evidence for
the involvement of a T-cell-associated serine protease (TSP-1) in cell killing.
Ann. Inst. Pasteur Immunol.. 138: 309-314, 1987.
Manyak, C. L., Norton, G. P., Lobe, C. G., Bleackley, R. C., Gershenfeld,
H. K.. V\eissman, I. L.. Kumar. V., Sigal, N. H., and Koo. G. C. IL-2 induces
expression of serine protease enzymes and genes in natural killer and non
specific T killer cells. J. ImmunoL 142: 3707-3713. 1989.
Krust. A.. Kastner, P., Petkovich, M., Zelent. A., and Chambón, P. A third
human retinoic acid receptor, h RAR->. Proc. Nati. Acad. Sci. USA, 86:
5310-5314, 1989.
Evans, R. M. The steroids and thyroid hormone receptor supcrfamily. Science
(Wash. DC). 240: 889-895. 1988.
Petkovich. M., Brand. N. J., Krust. A., and Chambón. P. A human relinoic
acid receptor which belongs to the family of nuclear receptors. Nature
(Lond.). 33: 444-450, 1987.
3018
Downloaded from cancerres.aacrjournals.org on April 14, 2017. © 1990 American Association for Cancer Research.
Enhancement of Murine Lymphokine-activated Killer Cell
Activity by Retinoic Acid
Tsung-hsing Lin and T. Ming Chu
Cancer Res 1990;50:3013-3018.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/50/10/3013
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on April 14, 2017. © 1990 American Association for Cancer Research.