| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Production In Vitro1
*
Department of Immunology and Medical Zoology, Faculty of Medicine, Fukui Medical University, Yoshida-gun, Fukui, Japan;
Department of Bacterial and Blood Products, National Institute of Infectious Diseases, Musashimurayamashi, Tokyo, Japan; and
Department of Hematology and Clinical Immunology, Kobe City General Hospital, Kobe, Japan.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is mediated mostly by monocytes/macrophages activated by
olig-DNA. However, its action on human IFN--producing cells has not
been well investigated. In the present study, we examined the effect of
oligo-DNAs on highly purified human NK and T cells. Bacillus
Calmette-Guérin-derived or synthetic oligo-DNAs induced NK cells
to produce IFN- with an increased CD69 expression, and the autocrine
IFN- enhanced their cytotoxicity. The response of NK cells to
oligo-DNAs was enhanced when the cells were activated with IL-2, IL-12,
or anti-CD16 Ab. T cells did not produce IFN- in response to
oligo-DNAs but did respond independently of IL-2 when they were
stimulated with anti-CD3 Ab. In the action of oligo-DNAs, the
palindrome sequence containing unmethylated 5'-CpG-3' motif(s) appeared
to play an important role in the IFN--producing ability of NK cells.
The changes of base composition inside or outside the palindrome
sequence altered its activity: The homooligo-G-flanked GACGATCGTC
was the most potent IFN- inducer for NK cells. The CG palindrome was
also important for activated NK and T cells in their IFN-
production, although certain nonpalindromes acted on them. Among the
sequences tested, cell activation- or cell lineage-specific sequences
were likely; i.e., palindrome ACCGGT and nonpalindrome AACGAT were
favored by activated NK cells but not by unactivated NK cells or
activated T cells. These results indicate that oligo-DNAs containing CG
palindrome act directly on human NK cells and activated T cells to
induce IFN- production. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The immunogenicity of oligo-DNAs have also been confirmed by other
investigators with findings that DNA extracted from various strains of
bacteria (other than BCG) and their synthetic counterparts or plasmid
DNA can induce mouse and human immunocompetent cells to produce IFN-
(18, 19, 20, 21, 22, 23, 24), IL-6 (18, 20, 21, 25, 26), IL-12
(18, 19, 21, 22, 23, 28), IL-1ß (29), TNF-
(20, 26, 29, 30), macrophage inflammatory protein-2
(26), type 1 IFN (23, 28), and IL-18
(23). These DNAs also enhance NK activity
(27) and stimulate B cells for their growth and
immunoglobulin production (18, 25, 31, 32, 33, 34, 35). This
cumulative evidence supports the current concept that bacterial DNA
promotes both cellular and humoral responses in protective and/or
defensive immunity in mice and humans (21, 23, 24, 28, 36, 37, 38, 39, 40, 41, 42).
It has been reported that bacterial DNA promotes NK cell function both
directly (18) and indirectly in mice. In the indirect
mode, bacterial DNA-stimulated monocytes/macrophages (Mos/M
s)
produce IL-12, TNF-, and type 1 IFN, and these cytokines induce
IFN- production by NK cells and their enhanced cytotoxicity
(12, 14, 18, 19, 22, 27). Bacterial DNA appears also to
induce T cell activation in Ag-mediated responses in vivo (28, 36, 39, 42) and in vitro in mice (41). However, to
our knowledge, there have been no corresponding studies regarding human
NK and T cells, although Roman et al. (23) showed that
human T cells do not respond to oligo-DNA at the resting state. If
oligo-DNAs directly target human NK and T cells, immunotherapy with the
oligo-DNAs would be more efficient because the direct activation of NK
cells leads to an enhancement of the MHC-nonrestricted cytotoxicity
(43), and IFN- produced by activated NK and/or T cells
induces the generation of Th1 cells (44). Therefore, we
examined whether the oligo-DNAs act directly on human NK or activated T
cells.
Recent studies on the DNA structures that determine the immunogenicity have revealed that there are some differences in the immunogenic sequences of bacterial DNA between those identified by us and those of other investigators; the active sequences determined by us are hexamer palindromes containing the CG motif(s) (12, 13, 14, 15, 16, 17), but theirs are the sequences containing CG in a particular sequence context, with less importance of palindrome sequence (18, 25, 27, 33, 34). In the present study, we also tested whether the CG palindromes and other CG-oligo-DNAs are truly immunogenic in human NK and T cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RPMI 1640 (Nissui, Tokyo, Japan) supplemented with 10%
heat-inactivated FCS (Equitech-Bio, Ingram, TX; endotoxin, <0.05
ng/ml), 100 U/ml penicillin G potassium (Banyu Pharmaceutical, Tokyo,
Japan), and 100 µg/ml streptomycin sulfate (Meiji Seika, Tokyo,
Japan) was used as a complete medium for cell culture. Human rIL-2,
rTNF-, and IFN- were generously provided by Shionogi
Pharmaceutical (Osaka, Japan), Dainippon Pharmaceutical (Osaka, Japan),
and Hayashibara Biochemical Laboratories (Okayama, Japan),
respectively. Human rIL-12 was purchased from R&D Systems (Minneapolis,
MN), and human rIL-15 and rIL-18 were commercially obtained from
BioSource International (Camarillo, CA). mAbs against human IFN-
(IgG2a), TNF- (IgG1), IL-12 (IgG1, clone C8.6), and IL-15 (IgG1)
were purchased from Genzyme (Boston, MA). Anti-IL-18 (IgG2a) and
anti-IFN- mAbs were purchased from R&D Systems and Pestka
Biomedical Laboratories (New Brunswick, NJ), respectively. Polyclonal
rabbit anti-IL-2 Ab was commercially obtained from Collaborative
Research (Bedford, MA). Based on our preliminary experiments, 1 µg of
the anti-IFN- mAb neutralizes 4 ng of human rIFN-, 100 ng of
anti-TNF- mAb neutralizes 200 pg of human rTNF-, and 1
µg/ml anti-IL-12 mAb neutralizes IFN- secretion by human NK
cells induced by 625 pg/ml human rIL-12. Anti-IL-18 (2 µg/ml), 2
µg/ml anti-IL-15, 5 µg/ml anti-IFN-, and 1 µg/ml
anti-IL-2 neutralized 50 ng/ml rIL-18, 5 ng/ml rIL-15, 1000 U/ml
rIFN-, and 100 U/ml IL-2, respectively. Purified mouse myeloma IgG1
and IgG2a proteins and rabbit serum purchased from ICN Pharmaceuticals
(Costa Mesa, CA) were used as an isotype-matched control Ig for the
mAbs and as a control serum for the IL-2 Ab, respectively, and were
shown not to alter the IFN- production or cytotoxicity of NK cells
in our experiments. The following reagents were commercially obtained:
polymyxin B (Sigma Chemical, St. Louis, MO); Dynabeads M-450 CD3, CD14,
CD19, and anti-mouse IgG (Dynal, Oslo, Norway); mouse
anti-human mAbs directed CD3, CD14, CD16, CD19, CD25, CD30, CD38,
CD56, CD69, CD71, CD94, CD97, CD134, CDw137, HLA-DR, and HLA-ABC
(PharMingen Becton Dickinson, San Diego, CA, and/or DAKO, Glostrup,
Denmark); FITC- or PE-labeled anti-CD3, anti-CD14,
anti-CD16, anti-CD19, and anti-CD56 (PharMingen); and goat
anti-mouse Ig (DAKO, Becton Dickinson Immunocytometry Systems, San
Jose, CA, or Caltag, San Francisco, CA).
Preparation of BCG-derived DNA and the synthetic oligo-DNA
A single-stranded oligo-DNA-rich fraction designated MY-1 was
extracted from BCG as described previously (5). MY-1 does
not contain any detectable cell wall components. We purchased the
oligo-DNAs from Nisshinbo (Tokyo, Japan), who prepared them using an
Expedite Model 8909 Nucleic Acid Synthesis System (PerSeptive
Biosystems, Framingham, MA). The endotoxin level in the synthetic
oligo-DNAs was less than 50 pg/100 µM when measured by the
Limulus test (Seikagaku, Tokyo, Japan), which specifically
detects endotoxin. The sequences of the oligo-DNAs are presented in
Figs. 4
and 9 and in Table II.
|
|
|
Isolation of PBMC.
PBMC were isolated from the venous blood of healthy volunteers by 60%
osmolarity-adjusted Percoll (Pharmacia Fine Chemicals, Uppsala, Sweden)
density gradient centrifugation. Platelets were removed from the PBMC
suspension or the density-fractionated cells by centrifugation on Nyco
Prep 1.063 (Daiichi Pure Chemicals, Tokyo, Japan), throughout the
experiments, because they inhibit MY-1-induced IFN- production by NK
or T cells (our unpublished observation).
Preparation of nonadherent cells (NAC).
PBMC suspended in complete medium were incubated in plastic dishes for
2 h at 37°C in 5% CO2 in a humidified atmosphere.
After the plastic adherence was repeated at least twice, the cells that
floated up by gentle agitation of the dishes were recovered, loaded on
a nylon wool fiber (Polysciences, Warrington, PA) column, and incubated
for 2 h at 37°C in 5% CO2 in humidified air. The
cells that passed through the nylon wool column were collected as NAC
depleted of Mos/Ms and B cells.
Purification of NK and T cells.
NAC were layered over a discontinuous density gradient composed of
42.9% (F1), 46.2% (F2), 50.0% (F3), 54.5% (F4), and 60% (F5) or
F1, F2, and 52.6% (F3/F4) of osmolarity-adjusted Percoll and
centrifuged for 30 min at 1500 or 1250 rpm, respectively, at room
temperature. The cell layers on F3–F4 or F3/F4 were collected as the
large granular lymphocyte (LGL)-rich (morphologically 60–90% of LGL),
and those on F5 were collected as the T cell fraction (>99%
CD3+ determined by flow cytometry (Fig. 10A)).
When T cell purity was not sufficient as determined by flow cytometry,
B cells and Mos/Ms were removed with the use of M-450 CD19 and M-450
CD14, respectively, or by cell sorting (Epics Elite, Beckman Coulter,
Fullerton, CA) with FITC-conjugated CD14, CD16, and CD19 mAbs
(PharMingen). NK cells were purified from the LGL-rich fraction by
negative or positive selection. In the negative selection, the LGL-rich
population was depleted of Mos/Ms and T/B cells by the serial use of
M-450 CD14, M-450 CD3, and M-450 CD19 magnetic beads or by an indirect
method using anti-CD14, anti-CD3, and anti-CD19 mAbs as the
first Abs, then with M-450 goat anti-mouse IgG or M-450 sheep
anti-mouse IgG magnetic beads as the second Abs. The immunomagnetic
depletion was repeated at least twice in each method. In the positive
selection, indirect immunomagnetic separation was performed with a
combination of anti-CD56 mAb and M-450 IgG after the repeated
depletion of Mos/Ms, using M-450 CD14 to avoid the trapping of
Mos/Ms in the NK cell population, which may be caused by
phagocytosis of the immunobeads or nonspecific cell aggregation. Cells
obtained by these methods contained >97% CD56-positive cells as
determined by flow cytometry (Fig. 1A) and <1% Mos/Ms
as evaluated with nonspecific esterase (Muto Pure Chemical, Tokyo,
Japan) or flow cytometric analysis of CD14 expression. In some
experiments, NK cells were isolated with FITC-conjugated CD56 mAb
(PharMingen) by cell sorting (Epics Elite). NK cells purified by
CD56-positive selection in the immunomagnetic separation method were
used after 6 h of incubation at 37°C in 5% CO2 to
detach the beads but were used without removing the beads when cells
were stimulated with anti-CD16 mAb. Both procedures for the NK cell
purification did not alter the responsiveness of the NK cells to
oligo-DNAs.
|
|
NAC, NK, or T cells were placed in 96-well plates (round-bottom
plates for NAC and NK cells and flat-bottom plates for T cells (Corning
Glass Works, Corning, NY) and cultured in complete medium at 37°C in
a humidified atmosphere with 5% CO2 under the conditions
described in Results. IFN- secreted in the culture
supernatants was measured by an ELISA kit (Cytoscreen Immunoassay Kit)
(Biosource International). The lower limit of detection for human
IFN- was 4 pg/ml.
Cytotoxicity assay
NK cells were cultured with K562 cells at the indicated E:T ratios in triplicate, for 4 h at 37°C in 5% CO2. The activity of lactate dehydrogenase released from damaged cells into the culture medium were measured by a cytotoxicity detection kit (lactate dehydrogenase) (Boehringer Mannheim, Mannheim, Germany), and the cytotoxicity is expressed here as a percentage of target cell lysis.
Flow cytometry analysis
Flow cytometry analysis was performed on a EPICS XL (Beckman Coulter). Cells were incubated with mAbs, followed by washing and labeling with FITC or PE-conjugated goat anti-mouse Ig. Data were obtained in a logarithmic scale.
Statistical analysis
Data were analyzed with the Wilcoxon signed rank test, ANOVA, or Student’s t test. Differences in the results were considered significant at p < 5%.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To determine the cell type(s) that is responsive to BCG-DNA, MY-1,
to produce IFN-, we separated LGL and T cells from NAC. When these
two cell fractions were cultured for 24 h at a concentration of
2 x 106 cells/ml, only the LGL fraction produced
IFN- in the presence of MY-1 (Table I,
experiments 1 and 2). NAC which contain 20–30% of NK and 70–80% of
T cells, produced IFN- in response to MY-1 when the cell density was
increased to 4 x 106 cells/ml (experiment 2), whereas
the T cell fraction did not produce IFN- even when cultured at
1 x 107 cells/ml (experiment 3) or for longer periods
(data not shown). Polymyxin B, a LPS inhibitor, did not affect the
MY-1-induced IFN- production, and DNase treatment of MY-1 abolished
the IFN--inducible activity (data not shown).
|
A) and tested their
responsiveness to MY-1. As shown in Table I (experiment 3), NK cells
produced IFN- in response to MY-1. The doses of MY-1 necessary to
induce the maximum amount of IFN- were between 12.5 and 50 µg/ml
in the culture of NK cells (Fig. 1B). IFN- production in
the culture with MY-1 was first observed at 18 h and increased
thereafter (Fig. 2). The amounts of
IFN- produced without MY-1 at 24-h culture were mostly below 4 pg/ml
and did not exceed 13 pg/ml in any NK cell sources examined. These
results show that NK cells are responsive to MY-1 in terms of IFN-
production.
|
-derived cytokines were not involved in the
oligo-DNA-induced IFN- production by NK cells
To prove that the MY-1-induced IFN- production is
caused by a direct action on NK cells, we added neutralizing
concentrations of mAbs against IL-12 or TNF- to the culture of NK
cells in the presence or absence of oligo-DNA. In this experiment,
instead of MY-1, g10GACGA (synthetic oligo-DNA) was used because of its
potent ability to induce IFN- (see below). As shown in Fig. 3, neither anti-IL-12 nor
anti-TNF- mAb influenced the IFN- production by NK cells
cultured with or without g10GACGA. The combined addition of mAbs
against IL-12 and TNF- also did not inhibit the production of
IFN-. No inhibitory effect of these Abs on the IFN- production
was observed when 10 µg/ml g10GACGA was applied for NK cell
stimulation. In addition, 1–10 µg/ml mAbs to IL-18, IL-15, or
IFN- did not alter the level of IFN- production induced by the
oligo-DNA (data not shown).
|
production by purified NK cells
To determine the effective sequence(s) in MY-1 for human
NK cells, 30-mer synthetic oligo-DNAs possessing various palindrome
sequences were first tested. These oligo-DNAs are analogues of BCG-4a
(12), the sequence of which was chosen from the cDNA
encoding 64-kDa heat shock protein AgA of BCG. GACGTC in BCG-4a was
replaced with different types of hexamer palindromes. These included
nine active palindromes which enhanced NK cytotoxicity more strongly
than GACGTC in mice and one inactive form (17). The
ability of each 30-mer oligo-DNAs to induce IFN- production was
compared at a concentration of 5 µM (almost equivalent to
50 µg/ml for each oligo-DNA) with 50 µg/ml MY-1. Among the
oligo-DNAs examined in three separate experiments, the sequences
containing ATCGAT, TCGCGA, CGTACG, CGGCCG, and GACGTC (named ATCGAT-30,
TCGCGA-30, CGTACG-30, CGGCCG-30, and GACGTC-30; BCG-4a,
respectively) were shown to induce IFN- production by NK cells as
potently as did MY-1. The oligo-DNAs containing the CGATCG and AACGTT
sequences (CGATCG-30 and AACGTT-30) also exhibited IFN--inducing
activity, but their activities were less potent than that of MY-1. The
oligo-DNAs with AGCGCT (AGCGCT-30) and GCGCGC (GCGCGC-30) had
a slight ability to induce IFN-. ACCGGT, which we
reported as an exceptionally inactive palindrome sequence
(14, 15, 16, 17), showed no IFN--inducing effect on NK cells.
The representative data are shown in Fig. 4.
The oligo-G introduced at the position of the extrapalindrome sequence
potentiates the ability of the palindrome sequence to enhance NK
activity in mice (17). To examine whether the IFN-
induction abilities of the weakly active palindromes, CGATCG and
AACGTT, are augmented by inducing the homooligo-G extrapalindrome, we
flanked these palindromes with 12-mer oligo-G at both 5'- and 3'-ends
(g12CGA and g12AAC, respectively). As shown in Fig. 4, the ability of
CGATCG, but not AACGTT, was augmented 2-fold by oligo-G flanking of the
original sequence. An elongated palindrome (GACGATCGTC) that flanked
with 10-mer oligo-G (g10GACGA) exhibited the most potent ability to
induce IFN- production, but ACAACGTTGT (g10ACAAC) did not. The doses
of g10GACGA capable of inducing the plateau level of IFN- production
by NK cells were from 6.3 to 50 µg/ml (10 µg/ml is almost equal to
1 µM) (Fig. 1). The level of IFN- production at 24-h culture with
10 µg/ml of g10GACGA was comparable with those induced by 10 U/ml
IL-2 or 10 pg/ml IL-12 (see Fig. 8). The amount of IFN- detected in
the culture with 30-mer oligo-G (g30) appeared slightly higher than
that in the control culture, but the increase was not significant.
|
production by NK
cells
The sequences of immunostimulatory DNAs that have been
extensively studied by other investigators do not have the hexamer
palindrome. In mice, gagaacgctcgaccttcgat (1643) is mitogenic to B
cells (33), and tccatgacgttcctgatgct (1668) induces not
only B cell activation (33) but also the production of
inflammatory cytokines by lymphocytes (18, 20).
Furthermore, tctccagcgtgcgccat (1758, antisense hBcl-2) activates NK
cells (37). In humans, ttgcttccatcttcctcgtc (2105)
directly activates B cells (34). We tested whether these
sequences are effective for human NK cells to induce IFN-
production. Unlike the active sequences mentioned above, they did not
induce IFN- production of human NK cells; these sequences were
tested at concentrations ranging from 0.2 to 20 µM, by extending the
culture periods to 5 days. The representative data (IFN- amount,
pg/ml) from three separate experiments with NK cells, which were
purified by cell sorting and cultured at 3 x 106/ml
for 3 days in the presence of 1643, 1668, 1758, 2105, AACGTT-30,
g10GACGA, or medium alone, were 20.0 ± 3.6 (mean
± SD, n = 3), 25.4 ± 4.8, 26.2 ± 5.1,
18.4 ± 1.8, 75.4 ± 15.4, 163.7 ± 25.2, and 18.8
± 3.6, respectively. Further, replacement of the AACGTT motif of
AACGTT-30 with the immunostimulatory core sequence, PuPuCGPyPy, i.e.,
AACGCT, AACGTC, or AACGCC, and with other sequences containing one
thymine at the 3'-side of AACG, such as AACGAT, AACGGT, AACGTA, and
AACGTG, did not induce IFN- production (data not shown).
Oligo-DNA-induced IFN- production participates in the
enhancement of NK activity
We tested whether oligo-DNAs can affect the cytotoxicity
of purified NK cells. As shown in Fig. 5,
when NK cells were cultured with MY-1 or g10GACGA, the ability to lyse
K562 cells was enhanced and the enhancement was prominent in the
culture with g10GACGA. The 30-mer homooligo-G, g30, used as a control
DNA, did not alter the cytotoxic activity. These results indicate that
the enhanced cytotoxicity is directly elicited by purified NK cells
cultured with synthetic oligo-DNAs. To examine the role of IFN-
induced by oligo-DNAs in the augmentation of NK activity, we added a
neutralizing mAb against IFN- to the NK cell culture in the presence
or absence of g10GACGA. The ability of g10GACGA to enhance NK activity
was diminished by the addition of anti-IFN- mAb (Fig. 6). Similarly, MY-1 enhanced NK cells and
the enhancement was inhibited in the presence of the anti-IFN-
mAb (data not shown).
|
|
To identify the activation-associated molecule(s) which
would be induced by oligo-DNA, we examined the expression of CD25,
CD69, and CD94 molecules and HLA-ABC on cultured NK cells. As shown in
Fig. 7, NK cells strongly expressed the
CD69 molecule when cultured with g10GACGA with 2.44 ± 0.56
(mean ± SD, n = 3) times more in the percent
positive cells than in those cultured with medium alone. Expression of
other molecules such as CD25 and CD94 was unchanged by the culture with
g10GACGA. In these experiments, the fluorescence of NK cells stained
with HLA-ABC was always intensified by the culture with oligo-DNA
(e.g., medium: 502 ± 17, g10GACGA: 590 ± 19, and IL-2 as a
positive control: 705 ± 17 as expressed by the mean
intensity ± SD), indicating the autocrine stimulation of NK cells
by IFN- induced by g10GACGA.
|
production by NK cells is enhanced
in the presence of IL-2, IL-12, or anti-CD16 mAb
We then tested the ability of NK cells to produce IFN-
in response to oligo-DNAs in the presence of IL-2, IL-12, or
anti-CD16 mAb to examine the influence of the activation status of
NK cells on their responsiveness to oligo-DNAs. g10GACGA was used in
this experiment because of its potent activity. As shown in Fig. 8, g10GACGA could induce IFN-
production by NK cells in the absence of the stimuli. The addition of
IL-2, IL-12, or anti-CD16 mAb to this culture significantly
enhanced the IFN- production. The increase was synergistic in the
culture with IL-2, whereas in the culture with IL-12 or anti-CD16
mAb, the increases were additive. Therefore, the activated NK cells
appear to be more susceptible to oligo-DNAs in terms of IFN-
production, especially with IL-2 stimulation.
Effect of different palindrome sequences on IL-2-activated NK cells
With IL-2, MY-1 also enhanced IFN- production by NK
cells (Fig. 9). We then examined the
effect of different palindrome sequences, which often occur in MY-1
(17), on the IFN- production by NK cells in the
presence of IL-2, in a manner similar to that used for unactivated NK
cells. The synthetic oligo-DNAs that induced IFN- production by
unactivated NK cells, i.e., ATCGAT-30, GACGTC-30, TCGCGA-30, CGTACG-30,
and CGGCCG-30, and those that showed weak or modest abilities to induce
IFN- production by the unactivated NK cells, i.e., AGCGCT-30,
CGATCG-30, GCGCGC-30, and AACGTT-30 all enhanced the IFN- production
by NK cells in the presence of IL-2. When the IFN--inducing activity
of these palindromes was expressed as a percentage of the control in
six separate experiments, the order of potency was as follows: AACGTT
(432 ± 95 pg/ml, mean ± SE), ACCGGT (408 ± 48),
CGTACG (376 ± 42), AGCGCT (376 ± 72), GCGCGC (320 ±
32), CGATCG (259 ± 37), TCGCGA (256 ± 35), ATCGAT (249
± 37), CGGCCG (246 ± 22), and GACGTC (238 ± 40). That of
MY-1 was 397 ± 34. These values were not statistically different.
Unlike those in the culture of unactivated NK cells, however, AACGTT
was the most potent and GACGTC the weakest palindrome in the culture of
IL-2-activated NK cells. In these results, there was a striking
difference in the sequence pattern of the induction of IFN-
production from those observed in the unactivated NK cells. That is, an
oligo-DNA with the ACCGGT palindrome (which was inactive in unactivated
NK cells) was able to induce IFN- in the presence of IL-2. One of
these data is shown in Fig. 9 (experiment 1) as the amount of IFN-
produced in the cell culture supernatant. A palindrome that contains GC
instead of CG, AAGCTT, showed no effect on the IFN- production
(experiment 2).
Effect of nonpalindrome sequences on IL-2-activated NK cells
It has been reported by Chace et al. (22) that
oligo-DNAs act on mouse NK cells in the presence but not in the absence
of IL-12. Activated NK cells may be more susceptible for oligo-DNA
stimulation, regardless of the particular contexts such as a palindrome
with internal CG or the PuPuCGPyPy sequence. Then two bases of AACGTT
were replaced at the 3'-side with theoretically possible dinucleotides
containing one thymine, to test for the IFN--inducing ability using
IL-2-activated NK cells. As shown in Fig. 9 (experiment 3), activated
NK cells responded to the sequences containing CG irrespective of
particular contexts such as the palindrome or PuPuCGPyPy. Among them,
the sequences with TT, CT, AT, TC, and TA at the 3'-side of AACG were
more potent.
Activated T cells can be induced by MY-1 and oligo-DNAs to produce
IFN-
Resting T cells did not respond to MY-1 (Table I).
However, the synergism between oligo-DNAs and IL-2 observed in NK cells
prompted us to examine the responsiveness of activated T cells to MY-1
and synthetic oligo-DNAs. We stimulated purified T cells (Fig. 10A) with Dynabeads M-450
CD3, which is able to activate T cells (manufacturer’s information),
and evaluated their IFN- production in response to MY-1 and
g10GACGA. As shown in Fig. 10B, T cells produced IFN- in
the presence of M-450 CD3, and this production was significantly
enhanced by the addition of MY-1 or g10GACGA. A 10-µg/ml
concentration of g10GACGA was almost equipotent to 100 U/ml IL-2 for
the induction of IFN- in M-450 CD3-stimulated T cells. The effect of
g10GACGA on the IFN- production was not influenced by the addition
of anti-IL-2 Ab to these cultures (Fig. 10C), indicating
that oligo-DNA-induced IFN- production is independent of IL-2
production by activated T cells.
The expression of CD25, CD69, CD94, HLA-ABC, CD30, CD38, CD71, CD94, CD97, CD134, CDw137, and HLA-DR was also tested as to whether specific activation marker(s) are induced by oligo-DNA. Anti-CD3 stimulation of T cells expressed higher levels of these molecules, and further enhancement was not observed when examined at 24 and 48 h of cultures with MY-1 or g10GACGA (data not shown).
Effective sequences to induce IFN- production by activated T
cells
To seek out the effective sequences for activated T cells,
the sequences involved in the induction of IFN- production were
examined in a manner similar to that used for NK cells. All oligo-DNAs
that contained hexamer palindromes with CG motif(s), except for ACCGGT,
induced IFN- production by M-450 CD3-stimulated T cells. When the
activity was presented as a percentage of the control in six
independent experiments, the order of potency among the active
palindromes was as follows: CGGCCG (273 ± 57 pg/ml, mean ±
SE); TCGCGA (223 ± 45); AGCGCT (220 ± 38); AACGTT (218
± 13); ATCGAT (173 ± 19); CGATCG (166 ± 19); GCGCGC
(163 ± 6); CGTACG (160 ± 21); and GACGTT (132 ± 6).
That of MY-1 was 233 ± 12. These values were not statistically
different, but one of the weakly active sequences for IL-2-activated NK
cells, CGGCCG, was distinctly active for anti-CD3-stimulated T
cells. A representative result of these experiments is shown as the
amount of IFN- in the culture supernatant in Fig. 11. A non-CG palindrome, CG of which
was replaced with GC, i.e., AAGCTT-30, showed no effect on IFN-
production (Fig. 11 and Table II).
|
inducing activity, when
replaced by AACGGT or AACGTG, the activity was very weak or completely
undetectable. AACGAT remained inactive (Table II). The substitution of
GC for CG in the active motifs abolished their activities (Table II).
However, the oligo-DNA, 1643, which contains AACGCT, did not induce
IFN- production. Then, AACGTT, AACGCT, AACGTC, and AACGTA were
flanked by oligo-G to determine whether a backbone sequence changes the
activity of these sequences. In the oligo-G-flanked sequences, only
AACGTT (g12AAC) showed potent activity (Table II), and this seemed most
potent for activated T cells among the sequences tested. To examine
whether the IFN--inducing activity of g12AAC is modified by the
methylation of CG, we synthesized methylated g12AAC in which CG inside
the palindrome was methylated. As shown in Table II, the methylated
g12AAC did not induce the IFN- production. Other well-investigated
immunostimulatory oligo-DNAs, 1668, 1758, and 2105, which activate
mouse spleen cells or human B cells, had little activity for
anti-CD3-stimulated T cells (Table II). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and that the autocrine IFN-
enhances NK activity. Our present study reveals that the oligo-DNA
responsiveness of human IFN--producing cells differ with that of
mice in terms of the sequence requirement and that the effective
sequences are different according to the cell types and/or the
activation status. Furthermore, the IFN--inducing activity of the
oligo-DNA was somehow interdependent on the presence of CG, the context
of the core motif of CG, and its outer flanking sequences.
To exclude an indirect action caused by contaminated Mos/Ms, we
purified the NK and T cell fractions to more than 97%
CD56+ and 99% CD3+ cells, respectively, and
less than 1% Mos/Ms. We also added mAbs against IL-12, TNF-,
IL-15, IL-18, or IFN- to these cell cultures with oligo-DNAs to
determine whether these cytokines produced by contaminated Mos/Ms
(<1%) were involved in the IFN- production. However, the levels of
IFN- production remained unchanged. Indeed, none of these cytokines
was detected in the culture supernatants of the NK or T cell population
(data not shown). For instance, the concentration of IL-12 measured by
ELISA was less than the detectable dose (1 pg/ml) which was not
sufficient to induce IFN- production by these cells in our
preliminary experiments. Furthermore, g10GACGA induced TNF-
production in the same culture conditions; however, the level was too
low (2.86 ± 1.70 pg/ml in 40-h culture, mean ± SD) to
induce IFN- production (Ref. 45 and our observation). Our
unpublished data suggest that TNF- detected in the NK cell culture
with g10GACGA may be produced by NK cells rather than by Mos/Ms.
These facts thus indicate that the effect of Mos/Ms contaminating
the NK or T cell fraction at <1% was negligible in the IFN-
production.
It has been reported in mice that bacterial DNA or oligo-DNA does not
induce IFN- production or NK enhancement when purified NK cells or
Mo/M-depleted nonadherent cells are used without additional stimuli
as the responders (19, 22, 27). In humans, these DNAs
directly activated NK cells (present study). This implies that mouse
and human NK cells behave differently in response to the DNA
stimulation. It is unclear as to what caused the difference in the
responsiveness of NK cells to oligo-DNAs between our present study and
the others. In B cells, a differential requirement regarding the
oligo-DNA sequence between humans and mice has been observed (33, 34). The sensitivity of NK cells to oligo-DNA sequences may
also, therefore, be different between mice and humans. The PuPuCGPyPy
sequences that were immunostimulatory for mouse spleen cells
(33), such as AACGCT or AACGTC, did not directly activate
mouse NK cells (27). In the present study, these sequences
did not act directly on human NK cells either. Although the hexamer
palindromes with CG dinucleotide(s) also did not activate mouse NK
cells (9, 12, 14, 19, 27), they were active for human NK
cells in our study. Human NK cells thus appear sensitive to oligo-DNA,
especially when the particular sequences such as the CG palindromes are
present. For Mos/Ms, hexamer palindromes behave actively regardless
of the species (12, 14, 15, 19, 23, 28, 46). In this
study, these sequences were shown to be effective also for activated
human T cells to enhance IFN- production. The palindrome sequences
containing the CG motif may therefore be some of the most potent
sequences for immunocompetent cells involved in the induction of
IFN- production in humans.
Among the CG-oligo-DNAs tested in this study, the favorable sequences
for the IFN- induction differed with the cell lineage and/or its
activation status. For example, in palindromes, ATCGAT was more
effective in unactivated NK cells than in activated NK cells, and vice
versa in AACGTT. Further, the weakly active CGGCCG in activated NK
cells was distinctly active for activated T cells. More importantly,
ACCGGT, one of the palindrome sequences that was inactive for
unactivated NK cells turned out to be active for IL-2-activated NK
cells. Because in humans ACCGGT is inactive for other types of cells
such as Mos/Ms (15), B cells (34), and
even M-450 CD3-activated T cells (Fig. 11), this sequence seems to be
specific for NK cells in relation to their activation status. In
addition to the palindromes, certain nonpalindrome sequences acted
effectively when NK or T cells were activated (Figs. 9 and 11 and Table II). Also in these sequences, however, the action differed with the
targeted cells as well, and AACGAT seemed favorable for activated NK
cells but not for activated T cells. From these facts, cell activation-
and/or cell lineage-specific sequences are likely in terms of the
IFN--inducing activity. Oligo-DNA-binding molecules which would be
expressed differently by the cell lineage and/or the activation status
may cause the different responsiveness of IFN--producing cells to
the sequences.
The CG palindrome oligo-DNAs used as the IFN- inducer in this study
are analogues of BCG-4a (GACGTC-30 in this study) which was randomly
selected from the cDNA-encoding 64-KDa protein (Ag A) of M.
bovis BCG (12). These analogues contain 4 CG
dinucleotides besides the CG(s) inside palindrome. In this study, the
change of CG to GC or to methylated CG in palindrome diminished the
IFN--inducing activity (Figs. 9 and 11 and Table II). Therefore,
unmethylated CG palindrome in these oligo-DNAs appears critical for
their activity. However, it should be also noted that the flanking
bases outside the CG palindrome seemed to cooperate with the active
core motif to implement the activity. This indicates that the
immunostimulatory sequences, PuPuCGPyPy, such as AACGCT, exhibited the
IFN--inducing activity when inserted into BCG-4a instead of the
palindrome motif (Table II). This sequence was inactive when used as
the 1643 oligo-DNA itself or by being flanked with oligo-G (Table II).
These facts signify that the activity of the CG-oligo-DNA is somehow
interdependent on the presence of CG, the context of the core motif of
CG, and its outer flanking sequences.
The regulatory role of the flanking bases could be further suggested by
the following results. For instance, when the palindrome was flanked
with oligo-G, CGATCG acquired more ability to induce IFN- production
by human NK cells, but AACGTT failed to do so (Fig. 4). Oligo-G has an
increased affinity to cellular membranes (46), and it has
been hypothesized that the G quartet formed by four contiguous
guanosine residues reduces the degree of rotational freedom of
oligo-DNA (47). In addition, as for the biological action,
oligo-G itself is not only mitogenic for murine B lymphocytes
(32) but also inhibitory for NK enhancement
(46) and IFN- production in the phosphorothioate form
(48). Therefore, oligo-G may act as a regulator of
palindromes by promoting the cell entry and binding to targeted
intracellular molecules; thus, the efficacy of palindromes to modulate
the targeted molecules may be augmented by being flanked with the
oligo-G. This could result in the enhancement of IFN- production
with some types of palindrome such as CGATCG (g10GACGA) or could
conversely cause the suppression of IFN- production by perturbing
the interacting signals toward the IFN- production in other types of
palindrome such as AACGTT in NK cells. This may also be one of the
reasons why oligo-G-flanked AACGTT (19) or TCAACGTTGA
(27) did not activate the murine NK cells. Changes of the
IFN--inducing activity by oligo-G flanking was further observed when
activated T cells were targeted: the activity of the palindrome AACGTT
was greatly enhanced by the oligo-G flanking; and that of the
nonpalindrome AACGCT, AACGTC, or AACGTA was decreased to negligible
levels. The flanking sequence thus appears to influence the oligo-DNA
activity in different ways according to the sequence of the core motif
and also to the lineage and/or activation status of the target cells,
in terms of IFN- production.
We previously reported that the biological activity of palindromes is
triggered after their entry into the cells (46, 49). As
shown in other types of cells (50, 51), oligo-DNA
containing palindrome with CG may be located in endosomes and in the
nucleus once it is taken up by NK or activated T cells. However, the
mechanisms by which oligo-DNA induces IFN- production in these cells
remain to be identified. Yi et al. (25) showed that
oligo-DNA directly increases the transcriptional activity of IL-6
promoter, suggesting an interaction of oligo-DNA with responsive
elements. Stacey et al. (29) and Sparwasser et al.
(30) demonstrated that oligo-DNA modulates the activity of
transcription factors. As an alternative mechanism, a certain structure
of oligo-DNA may interact, as a charged structure, with second
messenger-delivered signals which are involved in IFN- production,
because the calf thymus-DNA structure activates p68 kinase, which has
specific ATP-binding sites (52). This mechanism should be
examined in the oligo-DNA-induced IFN- production in human NK or
activated T cells, because the expression of CD69 Ag, which was
reported to be involved in signal transduction (53), was
enhanced by oligo-DNA in NK cells.
A clinical trial with MY-1 has been performed in Japan to assess its efficacy as an immunotherapeutic agent for malignant diseases, with positive results (54). Our present results imply that not only MY-1 but also the immunogenic synthetic oligo-DNAs induce multiple immune responses in vivo including NK cell activation and the possible induction of cytotoxic T cells, both of which are major components of the immune defense system against neoplasms. If NK or T cells are activated under some circumstances, the in vivo action of these oligo-DNAs may be augmented under those circumstances. Further studies are required to identify physiological potentiators of oligo-DNAs for the better therapeutic efficacy of these agents.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Correspondence address to: Dr. Sumiko Iho, 23 Shimoaizuki, Matsuoka-cho, Yoshida-gun, Fukui 910-1193, Japan. E-mail address:
3 Abbreviations used in this paper: BCG, bacillus Calmette-Gúerin; CG, CpG; oligo-DNA, oligodeoxynucleotide; Mo, monocyte; M, macrophage; LGL, large granular lymphocyte; NAC, nonadherent cells.
Received for publication September 4, 1998. Accepted for publication July 19, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/ß and - with deoxyribonucleic acid fraction from Mycobacterium bovis BCG. Jpn. J. Cancer Res. 79:866.[Medline]
. Proc. Natl. Acad. Sci. USA 93:2879. production by stimulation of interleukin-12 and tumor necrosis factor-. Cell. Immunol. 167:72.[Medline]
in vivo and increases the toxicity of lipopolysaccharides. J. Immunol. 156:4570.[Abstract]
production is dependent on macrophage secretion of IL-12. Clin. Immunol. Immunopathol. 84:185.[Medline]
-mediated shock. Eur. J. Immunol. 27:1671.[Medline]
interferon production by natural killer cell in scid mice: roles of tumor necrosis factor and bacterial stimuli. Infect. Immun. 59:1709. production by phosphorothioate deoxyguanosine oligomers. Immunopharmacology 29:47.[Medline]
This article has been cited by other articles:
|
|
 |
|
  E. Marcenaro, B. Ferranti, M. Falco, L. Moretta, and A. Moretta Human NK cells directly recognize Mycobacterium bovis via TLR2 and acquire the ability to kill monocyte-derived DC Int. Immunol., September 1, 2008; 20(9): 1155 - 1167. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Hoelzer, L. A. Shackelton, and C. R. Parrish Presence and role of cytosine methylation in DNA viruses of animals Nucleic Acids Res., May 1, 2008; 36(9): 2825 - 2837. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Montoya, H.-B. Jie, L. Al-Harthi, C. Mulder, P. J. Patino, M. T. Rugeles, A. M. Krieg, A. L. Landay, and S. B. Wilson Activation of Plasmacytoid Dendritic Cells with TLR9 Agonists Initiates Invariant NKT Cell-Mediated Cross-Talk with Myeloid Dendritic Cells J. Immunol., July 15, 2006; 177(2): 1028 - 1039. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Gauvreau, E. M. Hessel, L.-P. Boulet, R. L. Coffman, and P. M. O'Byrne Immunostimulatory Sequences Regulate Interferon-inducible Genes but not Allergic Airway Responses Am. J. Respir. Crit. Care Med., July 1, 2006; 174(1): 15 - 20. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Roda, R. Parihar, and W. E. Carson III CpG-Containing Oligodeoxynucleotides Act through TLR9 to Enhance the NK Cell Cytokine Response to Antibody-Coated Tumor Cells J. Immunol., August 1, 2005; 175(3): 1619 - 1627. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Schlaepfer, A. Audige, B. von Beust, V. Manolova, M. Weber, H. Joller, M. F. Bachmann, T. M. Kundig, and R. F. Speck CpG Oligodeoxynucleotides Block Human Immunodeficiency Virus Type 1 Replication in Human Lymphoid Tissue Infected Ex Vivo J. Virol., November 15, 2004; 78(22): 12344 - 12354. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Desmet, P. Gosset, B. Pajak, D. Cataldo, M. Bentires-Alj, P. Lekeux, and F. Bureau Selective Blockade of NF-{kappa}B Activity in Airway Immune Cells Inhibits the Effector Phase of Experimental Asthma J. Immunol., November 1, 2004; 173(9): 5766 - 5775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Chalifour, P. Jeannin, J.-F. Gauchat, A. Blaecke, M. Malissard, T. N'Guyen, N. Thieblemont, and Y. Delneste Direct bacterial protein PAMP recognition by human NK cells involves TLRs and triggers {alpha}-defensin production Blood, September 15, 2004; 104(6): 1778 - 1783. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Scheller, A. Ullrich, K. McPherson, B. Hefele, J. Knoferle, S. Lamla, A. R. M. Olbrich, H. Stocker, K. Arasteh, V. t. Meulen, et al. CpG Oligodeoxynucleotides Activate HIV Replication in Latently Infected Human T Cells J. Biol. Chem., May 21, 2004; 279(21): 21897 - 21902. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.-K. Yi, J.-G. Yoon, and A. M. Krieg Convergence of CpG DNA- and BCR-mediated signals at the c-Jun N-terminal kinase and NF-{kappa}B activation pathways: regulation by mitogen-activated protein kinases Int. Immunol., May 1, 2003; 15(5): 577 - 591. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kambayashi, E. Assarsson, A. E. Lukacher, H.-G. Ljunggren, and P. E. Jensen Memory CD8+ T Cells Provide an Early Source of IFN-{gamma} J. Immunol., March 1, 2003; 170(5): 2399 - 2408. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-J. Yeo, J.-G. Yoon, S.-C. Hong, and A.-K. Yi CpG DNA Induces Self and Cross-Hyporesponsiveness of RAW264.7 Cells in Response to CpG DNA and Lipopolysaccharide: Alterations in IL-1 Receptor-Associated Kinase Expression J. Immunol., January 15, 2003; 170(2): 1052 - 1061. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Takauji, S. Iho, H. Takatsuka, S. Yamamoto, T. Takahashi, H. Kitagawa, H. Iwasaki, R. Iida, T. Yokochi, and T. Matsuki CpG-DNA-induced IFN-{alpha} production involves p38 MAPK-dependent STAT1 phosphorylation in human plasmacytoid dendritic cell precursors J. Leukoc. Biol., November 1, 2002; 72(5): 1011 - 1019. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Yu, E. R. Kandimalla, L. Bhagat, J.-Y. Tang, Y. Cong, J. Tang, and S. Agrawal 'Immunomers'--novel 3'-3'-linked CpG oligodeoxyribonucleotides as potent immunomodulatory agents Nucleic Acids Res., October 15, 2002; 30(20): 4460 - 4469. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Hornung, S. Rothenfusser, S. Britsch, A. Krug, B. Jahrsdorfer, T. Giese, S. Endres, and G. Hartmann Quantitative Expression of Toll-Like Receptor 1-10 mRNA in Cellular Subsets of Human Peripheral Blood Mononuclear Cells and Sensitivity to CpG Oligodeoxynucleotides J. Immunol., May 1, 2002; 168(9): 4531 - 4537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. van Crevel, T. H. M. Ottenhoff, and J. W. M. van der Meer Innate Immunity to Mycobacterium tuberculosis Clin. Microbiol. Rev., April 1, 2002; 15(2): 294 - 309. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Miconnet, S. Koenig, D. Speiser, A. Krieg, P. Guillaume, J.-C. Cerottini, and P. Romero CpG Are Efficient Adjuvants for Specific CTL Induction Against Tumor Antigen-Derived Peptide J. Immunol., February 1, 2002; 168(3): 1212 - 1218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ronaghy, B. J. Prakken, K. Takabayashi, G. S. Firestein, D. Boyle, N. J. Zvailfler, S. T. A. Roord, S. Albani, D. A. Carson, and E. Raz Immunostimulatory DNA Sequences Influence the Course of Adjuvant Arthritis J. Immunol., January 1, 2002; 168(1): 51 - 56. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. E. Britigan, T. S. Lewis, M. Waldschmidt, M. L. McCormick, and A. M. Krieg Lactoferrin Binds CpG-Containing Oligonucleotides and Inhibits Their Immunostimulatory Effects on Human B Cells J. Immunol., September 1, 2001; 167(5): 2921 - 2928. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Verthelyi, K. J. Ishii, M. Gursel, F. Takeshita, and D. M. Klinman Human Peripheral Blood Cells Differentially Recognize and Respond to Two Distinct CpG Motifs J. Immunol., February 15, 2001; 166(4): 2372 - 2377. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Warren, S. K. Bhatia, A. M. Acosta, C. E. Dahle, T. L. Ratliff, A. M. Krieg, and G. J. Weiner APC Stimulated by CpG Oligodeoxynucleotide Enhance Activation of MHC Class I-Restricted T Cells J. Immunol., December 1, 2000; 165(11): 6244 - 6251. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. FUJIEDA, S. IHO, Y. KIMURA, H. YAMAMOTO, H. IGAWA, and H. SAITO Synthetic Oligodeoxynucleotides Inhibit IgE Induction in Human Lymphocytes Am. J. Respir. Crit. Care Med., July 1, 2000; 162(1): 232 - 239. [Abstract] [Full Text]
|
|
|
|
|
|
B. M. Segal, J. T. Chang, and E. M. Shevach CpG Oligonucleotides Are Potent Adjuvants for the Activation of Autoreactive Encephalitogenic T Cells In Vivo J. Immunol., June 1, 2000; 164(11): 5683 - 5688. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
