Cytotherapy 24 (2022) 1232 1244

Contents lists available at ScienceDirect

CYTOTHERAPY journal homepage: www.isct-cytotherapy.org

FULL-LENGTH ARTICLE
Translational Research

g c cytokine-aided crosstalk between dendritic cells and natural killer

cells together with doxorubicin induces a healer response in
experimental lymphoma by downregulating FOXP3 and programmed cell
death protein 1
Ranjeet Singh1,*, Uttam Gupta1,*, Prateek Srivastava1, Ankush Paladhi2, Ugir Hossain Sk3,
Sumit Kumar Hira1,2,*, Partha Pratim Manna1,**
1
Immunobiology Laboratory, Department of Zoology, Institute of Science, Banaras Hindu University, Varanasi, India
2
Cellular Immunology Laboratory, Department of Zoology, The University of Burdwan, PurbaBardhhaman, India
3
Chittaranjan National Cancer Institute, Kolkata, India

A R T I C L E I N F O A B S T R A C T

Article History: Background aims: The stimulatory natural killer dendritic cell axis in the tumor microenvironment could
Received 5 May 2022 play a critical role in stimulating cytotoxic T cells and driving immune responses against cancer.
Accepted 31 July 2022 Methods: We established a novel treatment protocol by adroitly combining chemotherapy with doxorubicin
and immunotherapy with dendritic cells and natural killer cells against a highly aggressive and malignant
Key Words: lymphoma called Dalton’s lymphoma.
Binary therapy Results: Our data suggest that binary application of adoptive cell therapy and chemotherapy nearly cures
DC-NK crosstalk (95%) early-stage experimental lymphoma. In the case of mid-stage cancer, the success rate was significantly
Doxorubicin
lower but still impressive (75%). Our results demonstrated that the application of combination therapy in
PD-1
early-stage cancer significantly reduced the tumor volume and extended the lifespan of the experimental
Regulatory T cells
animal in addition to reinvigorating the immune system, including restoring the effector functions of den-
dritic cells and natural killer cells. The novel protocol limits the metastasis of tumor cells in vascularized
organs and rearms the adaptive immune response mediated by dendritic cells and CD4+ and CD8+ T cells.
Conclusions: Combination therapy in the early stage alters the cytokine profile, increases interferon-g and tumor
necrosis factor-a in the serum of treated animals and downregulates programmed cell death protein 1 expression
in CD8+ T cells. Thus, cooperative and cognitive interactions between dendritic cells and natural killer cells in addi-
tion to therapy with doxorubicin promote the immune response and tumoricidal activities against lymphoma.
© 2022 International Society for Cell & Gene Therapy. Published by Elsevier Inc. All rights reserved.

Introduction interactions (DC NK) activate NK cells in vitro, leading to better anti-
tumor responses in vivo [4,5]. The rare intratumoral DC subset was
Dendritic cells (DCs) are considered to be the prime initiator of found to be uniquely potent in restimulating T cells in the tumor
adaptive immune responses [1]. DCs are potent phagocytes that microenvironment and is critical for adoptive T-cell therapy in mouse
internalize pathogens and digest apoptotic or necrotic tumor cells for models [6 10]. These DCs are conventional DCs, and, when derived
the processing and presentation of antigens and are guided by matu- from tumors, they were defined in mice by the surface expression of
ration signals to migrate to the draining lymph nodes, where they the integrin CD103 and by the expression of thrombomodulin
potentially activate T cells [2,3]. DC and natural killer (NK) cell (BDCA-3; also known as CD141) in humans. Studies in lung carci-
noma have shown that conventional DCs are less frequent in tumors
than in nearby normal tissues [11]. In cancer, checkpoint blockade
** Corresponding author: Dr. Partha Pratim Manna, Department of Zoology, Institute
of Science, Immunobiology Laboratory, Banaras Hindu University, Varanasi 221005, therapies such as anti CTLA-4 or anti-programmed cell death protein
India. 1 (PD-1) have become remarkably successful in reinvigorating T lym-
 Yes phocytes against tumors and thus creating an opportunity for long-
E-mail addresses: [email protected] (S.K. Hira), [email protected]
term protection. However, it is a general occurrence of having a sig-
(P.P. Manna).
* These authors contributed equally to this work.
nificant (»80%) portion of patients without objective responses

https://doi.org/10.1016/j.jcyt.2022.07.012
1465-3249/© 2022 International Society for Cell & Gene Therapy. Published by Elsevier Inc. All rights reserved.
 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244 1233

following the aforementioned schedule of treatments [12]. Genera- conducted according to AVMA (American Veterinary Medical Associ-
tion of new epitopes for T cells by mutational frequencies is believed ation) Guidelines on Euthanasia (AVMA 2013).
to be the single most important factor for the reduced immune
response [13]. Several other critical parameters and contributions by Mice and cell lines
various other cell types that control the outcome of the treatment
need to be investigated. Details on the mouse strain and cell lines used in the experiments
NK cells are considered a key player in innate immunity against are based on our previous studies [23] and are given in the supple-
virus infections and in neoplasia [14,15]. Recombinant interleukin (rIL)- mentary material.
15 is an all-important cytokine that influences the generation and func-
tions of cells in innate immunity, including NK cells, intestinal gd T Reagents and antibodies
cells, NKT cells, and DCs [16 21]. IL-15 has notable significance for the
functional maturation of DCs in vitro [21]. CpG-stimulated conventional Details are given in the Materials and Methods section of the sup-
DCs, rather than plasmacytoid DCs, are the main producers of both plementary materials. Details of antibodies used in the study are pre-
interleukin (IL)-15 and IL-12 and mediate crosstalk between conven- sented in supplementary Table 1.
tional DCs and plasmacytoid DCs, regulating CpG-mediated immune
activation in vivo [22]. We previously evaluated the immune response Isolation of splenic DCs, T cells and NK cells
of lymphoid DCs against Dalton’s lymphoma (DL) with reference to the
critical role played by the DC-derived tumor necrosis factor (TNF) Splenic DCs, NK cells and T cells from normal, tumor-bearing or
superfamily ligands TNF-related apoptosis-induced ligand (TRAIL) and treated mice were isolated as previously described [23,24]. For isola-
TNF-a. Our results showed that tumor antigens derived from lym- tion of DCs, the single-cell suspension of splenocyte was treated with
phoma-bearing mice downregulate disease-protecting TNF superfamily a cocktail of anti-mouse CD3, CD8, CD14, CD19, CD49b (DX5) anti-
ligands [23,24]. The tumoricidal effects of TRAIL derived from DCs were bodies plus immunomagnetic Dynabeads (Invitrogen Thermo Fisher
augmented by signal transducer and activator of transcription 3 Scientific, Waltham, MA, USA) followed by magnetic separation of
(STAT3) deactivation in tumor cells by cucurbitacin I and made them the bead-bound cells. For NK cell isolation, cocktail of anti-mouse
significantly more susceptible. Cucurbitacin I upregulates TRAIL receptor CD3, CD8, CD14, CD19, DCIR2 (33D1) antibodies plus immunomag-
expression and activates caspases in DL cells [23]. TNF-a levels were netic Dynal beads were used followed by magnetic separation of the
diminished in DCs derived from DL tumor bearing mice, which bead-bound cells. Details are provided in the Materials and Methods
resulted in disease progression. Doxorubicin (DOX)-resistant DL cells section of the supplementary material.
showed increased levels of Mcl-1 and Bcl-2, accompanied by elevated
STAT3 phosphorylation. These cells are refractory to DC-derived TNF-a. DC and CD8+ cell functional assays
DOX-resistant DL cells are susceptible to TNF-a derived from DCs when
tumor cells are pre-treated with cucurbitacin-I, resulting in STAT3 Mouse CD11c+ DCs were obtained from normal, DL or treated ani-
downregulation and ablation of other survival molecules [24].Therapy mals and cultured in 96-well plates. The cells were washed (£3) in
with tumor lysate pulsed, rIL-15 activated DCs plus cucurbitacin I aug- culture medium and used for the expression of costimulatory mole-
mented the survival of tumor-bearing mice significantly, which was fur- cules, class II molecules, etc. CD8+ T cells were obtained from differ-
ther improved by additional treatment with rIL-15 [25]. ent groups and were cocultured with DL or YAC-1 tumor cells at
Here, we investigated the role of binary application of DOX and different Effector:Target ratios depending upon the experimental
adoptively transferred DCs and NK cells stimulated with rIL-15 protocol. CD8+ T-cell cell-mediated growth inhibition (48-hour 3-
against established DL tumors developed in the peritoneum of BALB/ [4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide [MTT]
c mice. Crosstalk between DCs and NK cells depends on TNF-a, which assay) and cytotoxicity (18-hour lactate dehydrogenase release
is critically important for the tumoricidal effect [26]. Our data suggest assay) assays were performed as described previously [23].
that the application of dual therapy (DC+NK cells and a suboptimal
dose of DOX) in early-stage lymphoma significantly increased the In vivo studies: Tumor challenge and treatment experiments
lifespan of tumor-bearing animals with nearly 100% tumor-free sur-
vival up to day 60 compared with the death of untreated litter at DL tumors were maintained in vitro in RPMI 1640 (Invitrogen
days 20 22. The therapeutic efficacy was accompanied by enhanced Thermo Fisher Scientific) supplemented with 10% fetal calf serum
expression of costimulatory molecules in DCs and elevated serum (Invitrogen Thermo Fisher Scientific) and in the peritoneal cavity of
levels of IFN-g and TNF-a. In addition, the combination therapy abro- BALB/c (H2d) mice as mentioned earlier [23]. Female BALB/c mice
gated the dominance of FOXP3-expressing regulatory T cells (Tregs) weighing 18 20 g (n = 6/group) were transplanted with DL tumor
and suppressed enhanced tumor growth factor (TGF)-b secretion. cells (3 £ 104), and after 4, 8 or 14 days, animals were treated with
Binary therapy downplayed disease, exacerbating the role of PD-1 in either a suboptimal dose of DOX (50 mg/mouse) or DC+NK cells
CD8+ T cells and eliminating them from metastasizing organs, thus (DC = 1 £ 105/mouse, NK cell = 5 £ 105/mouse) or in combinations.
improving prognosis by occluding recurrence of the disease. Tumor volume was calculated by measuring the abdominal circum-
ference starting from the day of tumor transplant. Control animals
received vehicle only (phosphate-buffered saline). Mice were
Methods observed until day 80, when final data collection was made for
Kaplan Meier survival analysis.
Ethics statement
Splenocyte stimulation studies
All animal experiments were performed in accordance with the
CPSEA (Committee for the Purpose of Control and Supervision of T cells (CD4+) from healthy control, DL tumor-bearing or surviving
Experiments on Animals), MoEF (Ministry of Environment, Forest vaccinated mice were used as responder cells against whole tumor
and Climate Change) and GOI (Government of India). The study pro- lysate (10 mg/mL)-pulsed and mitomycin C (10 mg/mL)-treated stim-
tocol was approved by the Institutional Ethics Committee, Banaras ulator DCs derived from the treated mice. The responder T cells
Hindu University. Animals were euthanized humanly by cervical (5 £ 105) were cocultured with stimulator DCs (5 £ 103), and the cul-
dislocation, and care was taken to minimize suffering and was ture plates were incubated at 37°C and 5% CO2 for 120 hours followed
1234 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244

by the MTT assay to determine cell proliferation as described previ- and based on the mean survival of the untreated tumor-bearing
ously [25]. mice, which was approximately 20 22 days. Combination therapy
with DOX plus DC+NK cells significantly enhances the survival of
Flow cytometry and enzyme-linked immunosorbent assay (ELISA) tumor-bearing mice with 50% survival beyond day 80 with reduced
tumor volume and tumor weight (P < 0.0001). The effect was also
Flow cytometry and ELISA for IFN-g , TNF-a, IL-2, TGF-b, IL-6 and impressive at day 8 PTT mice, registering 30% survival at day 80
IL-10 were performed as described previously [26]. A detailed (Figure 1). In addition, binary treatment at this advanced stage
description is given in the Materials and Methods section of the sup- reduces the tumor volume and weight (P < 0.0001) (Figure 1).
plementary materials. Gating strategies for T cells, Tregs and DC are Detailed statistical analysis for Kaplan Meier curves is presented in
presented in supplementary Figure 1. supplementary Tables 2 and 3. DOX only treatment augments the
lifespan of the tumor-bearing mice only up to 28 days, when the
Tissue processing and immunohistochemical analysis entire animal succumbs to death with high tumor load, including
uncontrolled growth of the tumor volume and weight (supplemen-
Tissues (liver, lung, spleen and lymph node) were collected from tary Figure 2A-D). Therapy with crude DL tumor antigen-pulsed DCs
healthy control, DL tumor-bearing and vaccinated mice, fixed in nor- in combination with NK cells augmented survival up to 50 days (sup-
mal buffered formalin and processed. Formalin-fixed, paraffin- plementary Figure 2E-H). Therapy started at day 14 PTT did not pro-
embedded tissue sections were stained with hematoxylin and eosin duce similar protection as that observed on day 4 or day 8 PTT with
and examined for regions of tumor metastasis. Immunohistochemical respect to an increase in life span and a reduction in tumor volume/
staining was performed on paraffin-embedded lymph node and weight. Day 14 PTT is considered late-stage cancer when the tumor
spleen samples. Immunohistochemical staining was performed using mass increases uncontrollably and the weight/volume of the tumor
5 mm tissue sections using Novolink Polymer Detection Systems also increases significantly, leading to death of the litter (supplemen-
(Leica, Wetzlar, Germany) according to the manufacturer’s protocol. tary Figure 3). Rigorous statistical analysis for Kaplan Meier curves
To summarize, after antigen retrieval with EDTA buffer (pH 8.0), sam- is presented in supplementary Tables 4 and 5.
ples were treated with anti-mouse primary antibodies (CD4, FOXP3,
CD8, PD-1) and incubated overnight at 4°C. A secondary antibody Designing the best protocol for therapy
was applied for 30 minutes at room temperature. Chromogenic
detection of protein expression was determined in the presence of The aforementioned data suggest that the introduction of binary
diaminobenzidine, and images were captured by a Nikon 80i research therapy with DOX plus DC+NK cells produced a significant advance-
microscope (£400). ment in therapeutic success and better prognostic outcomes against
deadly lymphoma. We realized that DOX therapy followed by cell
Statistical analysis therapy may be insufficient to elicit a stronger antitumor immune
response against the disease. Thus, an alternative therapeutic sched-
An unpaired, 2-tailed t test was used to compare pairs of groups, ule was designed. Alternative DOX therapy with intermittent DC
and 2-way analysis of variance with the Tukey test was performed +NK-cell therapy produced a significantly greater survival rate (P <
for comparisons in multiple group experiments using Prism 7 soft- 0.0001) accompanied by reduced tumor mass and volume and
ware (GraphPad, San Diego, CA, USA). The log-rank (Mantel Cox) increased survival of the treated mice (Figure 2). Kaplan Meier sta-
test was performed to define significant differences in survival tistical analysis is presented in supplementary Tables 6-9.
between the two groups in the therapeutic studies. P values less than The immunogenicity of CD11c+DCs used in the therapy also was
0.05 were considered significant. All values in the graphs are reported assessed before their adoptive transfer to DL mice along with DOX in
as the mean § standard deviation. The values were calculated for combination therapy. The expression of class II, the activation marker
each experimental group (n = 3 5). All experiments were performed CD69, and the costimulatory molecules CD40 and CD86 was signifi-
in triplicate and represent summary data of five independent experi- cantly upregulated in DCs treated with whole DL tumor antigen (Ag)
ments (mice = 5) from each group. Each plot indicates the mean § +IL-15 compared with DCs treated with or without Ag stimulation
standard deviation value and was calculated for each experiment (supplementary Figure 4). Mean fluorescence intensity quantification
(n = 3 5 for each mouse, and a total of 5 mice were used for each was also significantly greater in Ag+IL-15 treated DCs used for adop-
group). Multivariate analysis of ELISA data (e.g., PCA and correlation tive transfer, suggesting that optimally activated DCs were delivered
matrix) was performed using GraphPad Prism 9. Correlations for better outcomes in binary therapy (supplementary Figure 4).
between FOXP3:CD4 and CD62 L:CD8 ratios were analyzed using the
nonparametric Spearman test [27]. Flow cytometry data were ana- Binary therapy improved histopathology and enhanced cure response
lyzed with the use of FlowJo software (version 10.0.5; Treestar,
Woodburn, OR). Statistical analysis of Kaplan Meier curves are pre- The application of DOX with alternate DC+NK-cell therapy signifi-
sented in supplementary Tables 2-9. cantly enhanced the survival of tumor-bearing animals, suggesting
its unique advantage in fighting against the disease (P < 0.0001). DL
Results and Discussion tumor-bearing animals were either left untreated (DL) or treated
with DOX (T1), antigen-pulsed DC+NK cells (T2) or antigen-pulsed
Establishment of an effective DC-NK therapy protocol in lymphoma DC+NK cells+DOX (T3) (Figure 3A). Treatment of DL mice with the T3
combination enhanced survival>85% in treated animals with no
We have introduced a binary treatment schedule for experimental relapse when data were last recoded. The overall appearance and the
DL lymphoma where we have coadministered DOX, alternating with size of the vascularized organs returned to normal when the organ
DC and NK cells. We categorized murine lymphoma into three differ- texture was compared with the tumor-bearing mice (Figure 3B). His-
ent stages, mimicking the stages of the disease and simulating similar topathologic parameters of vascularized organs (spleen, liver and
conditions in patients with lymphoma. Stage 1 is day 4 post-tumor lung) in the DC+NK+Ag+DOX (T3)-treated group demonstrated a sig-
transplant (PTT), representing the early stage of the disease, stage 2 is nificant advantage of the binary treatment over DOX only or therapy
day 8 PTT, representing middle/advanced stage cancer, and stage 3 is with T2 combination. DL mice showed extensive metastasis of the
day 14 PTT, representing late-stage cancer. All these categorizations tumor cells in the length and breadth of the vascularized organs,
were reasonable approximations with reference to human disease such as the spleen and liver, with large areas of metastatic foci in the
 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244 1235

Figure 1. Binary therapeutic protocol and survival analysis. (A and B) Therapeutic schedule for binary therapy started at day 4 and day 8 PTT. (C and D) Kaplan Meier survival anal-
ysis for the animal receiving the indicated treatment started at day 4 and day 8 PTT. DL, untreated DL tumor bearing animal; DL+DOX, DOX-treated DL mice; DC, DL mice treated
with DC only; DOX+DC, DL mice treated with DOX+DC; NK, DL mice treated with NK-cell only; DOX+NK, DL mice treated with DOX+NK; DC+NK, DL mice treated with DC+NK; DOX
+DC+NK, DL mice treated with indicated combination; DC(Ag)+NK, DL mice treated with antigen pulsed DC & NK cells; DOX+DC (Ag)+NK, DL mice treated with indicated combina-
tion. (Color version of figure is available online).

liver. In addition to increased survival, treatment with the T3 combi- indicating the protective role of the cytokine, which plummeted to
nation significantly reduced tumor weight and volume compared a significantly low level in the DL group of animals (P < 0.001)
with the untreated control (Figure 3C-E). T3 treatment eliminated (Figure 3H). In addition, TNF-a secretion in serum was reduced in
tumor cells in the liver significantly compared with the untreated DL DL animals compared with the T2 and T3 groups, indicating a sig-
mice with no or sparsely visible tumor cells, suggesting the ability to nificant surge in the cytokine level (P < 0.001) (Figure 3H). The sig-
prevent cancer cell metastasis and dissemination (Figure 3F). T1 and nificant elevation in IFN-g and TNF-a in treated mice, specifically in
T2 treatment partially eliminated the tumor cells, with T2 showing the T3 group, could be related to the critical role played by DCs and
better performance compared with T1. In the lung histology, T3 treat- NK cells, which triggered cytokine-mediated protection in binary
ment restored the alveolar architecture to a normal level from severe therapy against lymphoma.
infiltration of the tumor cells, as observed in the untreated group
(Figure 3F). The splenic cortical architecture was severely distorted, Restoration of the immune response by binary therapy in early-stage
and the medullary regions were infested with tumor cells in DL mice. cancer
T3 treatment restored the normal features of the splenic archi-
tecture, which were very similar to those of naive spleens from We then examined the potential effect of T3 treatment on the res-
healthy animals (Figure 3F). Similarly, kidney architecture returns toration of the immunological competence of DCs and CD8+ T cells as
to its normal feature following T3 treatment. The number of meta- judged by their effector functions. DCs were analyzed for the expres-
static foci in the liver was quantified in DL livers and compared sion of class II, and costimulatory molecules showed significant
with the treatment groups. T3 treatment significantly reduced the enhancement in the expression of CD40, CD80 and CD86 derived
number of tumor foci, which overwhelmed the length and breadth from the T3 group (Figure 4A). Mean fluorescence intensity was
of the organs in the DL group (Figure 3G). The serum interferon determined for the indicated treatment and showed significant upre-
(IFN)-g level in the T3 treatment group was significantly enhanced, gulation of the costimulatory and class II molecules in T3 treatment,
1236 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244

Figure 2. Day 4 and 8 alternate combination therapy. (A and B) Therapeutic schedule for combination therapy given on alternate days starting at day 4 and at day 8 PTT. (C and D)
Kaplan Meier survival analysis for the animals receiving the indicated treatment as described in Figure 1 at day 4 and day 8 PTT. (Color version of figure is available online).

which was substantially downregulated in DL mice (Figure 4B). DC- similar to those observed in DC effector function (P < 0.01)
mediated growth inhibition of DL and YAC-1 cells showed improved (Figure 4D). Similar results were observed with YAC-1 cells (Figure 4E
effector function by DCs derived from the T3 group (P < 0.01) and F).
(Figure 4C). We analyzed the tumoricidal functions of DCs and CD8+ T Responder T cells obtained from the T3 group showed signifi-
cells derived from normal (DC1 and N-CD8), DL (DC2 and DL-CD8) cantly greater antigen-specific proliferation than T cells obtained
and treated groups (T1, T2, T3) (DC3 and T1-CD8, DC4 and T2-CD8, from DL mice (Figure 5). Normal T cells did not respond appreciably
DC5 and T3-CD8) of mice against DL and YAC-1 cells. to antigen (crude soluble DL antigen)-specific proliferation
Growth inhibition of DL tumor cells was augmented by splenic DCs (Figure 5A). The antigen-specific T-cell response generated by DL
generated from T3 animals treated with DOX plus DC+NK cells. Splenic mouse T cells was very poor, indicating the presence of impaired
DCs caused »40% growth inhibition of DL tumor cells compared with responder T cells, likely due to neoplastic development (Figure 5B).
18% inhibition by DCs derived from DL mice at an E:T ratio of 1:1 (P < CD4+ T cells isolated from the treated groups demonstrated an
0.01). In contrast, DCs derived from animals treated with DC+NK (T1), increasing antigen-specific proliferative response in the order of T3 >
DC+NK+Ag (T2) or DC+NK+Ag+DOX (T3) showed significant increases T2 > T1, suggesting that antigen-pulsed cytokine-activated DCs in
in growth inhibition to 50%, 55% and 65%, respectively, at an E:T ratio combination with NK cells and DOX strongly induced an immune
of 1:1 (P < 0.01). At a greater E:T ratio (5:1), the growth inhibition of response with greater therapeutic results than DC+NK cells alone
DL tumor cells by DCs from the T3 group increased substantially and (Figure 5C-E). The best antigen-specific T-cell response (higher stim-
reached 80% compared to 55% with DCs from the control animal (P < ulation index) was demonstrated by T cells generated from the T3-
0.01) (Figure 4C). Similar results were observed with YAC-1 cells treated group (P < 0.05 and P < 0.01) (Figure 5E). These data suggest
(Figure 4E).CD8+ T cells derived from the treated animals showed aug- that binary therapy of drug- and antigen-pulsed adoptively trans-
mented cytotoxicity against DL cells compared with the T cells derived ferred DCs produced a durable immune response to counter the
from healthy control animals at E:T ratios of 1:1 and 5:1 (P < 0.01) malignant pathogenesis of lymphoma. The advantage of combining
(Figure 4D). CD8+ T cells from DL mice showed impaired functions DOX with adoptive DC therapy also was evident in significantly
 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244 1237

Figure 3. Enhanced tumor-free survival and reduced DL tumors in BALB/c mice treated with binary therapy given on alternate days started at day 4 PTT. (A) Therapeutic schedule
for combination therapy given on alternate days starting at day 4. (B) Photographic evidence supporting therapeutic success and corresponding organ sizes following the indicated
treatment. (C, D and E) Kaplan Meier survival analysis, body weight and measurement of abdominal circumferences (tumor volume) of untreated and treated groups in mice
receiving the treatment for the indicated time period. (F) Histopathologic analysis of the liver, lung, spleen and kidney of treated and untreated animals compared with healthy con-
trols. (G) Quantification of metastatic foci in the livers of the untreated and treated groups. **p < 0.01, and ***p < 0.001 value. (H) Serum IFN-g and TNF-a levels for the animals
treated with or without combined therapy at day 22 when the untreated DL animal succumbs to death. Data are presented as the mean § standard deviation, n = 6 of all the animals
in individual groups. (Two-way analysis of variance, Holm Sidak posthoc test, *P < 0.05, ***P < 0.001). (Color version of figure is available online).

greater splenic CD8+ T-cell activity against DL and YAC-1 tumor cells T regulatory cells. In contrast, the percentage of CD8+ T cells was sig-
(Figure 4D and F). nificantly lower in DL mice. Treatment with T1 partially reduced CD4
expression, which was augmented following binary therapy with DC
Binary therapy downregulates PD-1 and FOXP3 expression in T cells +NK cells, as observed in T2 treatment, likely due to the mobilization
of CD4-expressing DCs. T3 treatment significantly enhanced the con-
Next, we analyzed the expression of FOXP3 and PD-1 in the dition and nearly restored the normal histopathological architecture,
spleens of untreated and treated DL animals and compared them which was also reflected in the increased survival of the animals
with those of naive control mice. CD4+ T cells were increased in DL (Figure 6A). DL mice expressed very high levels of FOXP3, indicating
mice compared with naive control mice, suggesting the occurrence of the opulence of Tregs in tumor-bearing animals (Figure 6A). These
1238 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244

Figure 4. Fluorescence-activated cell sorting analysis and quantification of costimulatory and class II molecules in infiltrated CD11c+ DCs in the spleen of untreated or mice that
received the indicated treatment (T1-T3) as described in Figure 3. (A and B) Restoration of Class II and costimulatory molecules (CD40, CD80 and CD86) in animals subjected to
binary therapy compared with untreated DL mice. Representative of one experiment out of three performed is shown. (C and E) Growth inhibition against DL or YAC-1 tumor cells
by DCs derived from treated groups DC3 (DOX), DC4 (DC-Ag+NK) and DC5 (DC-Ag+NK+DOX) compared with DCs derived from naive control or untreated DL mice. (D and F) Cyto-
toxicity potential of CD8+ T cells derived from T1- to T3-treated animals compared with naive control or untreated DL mice against DL and YAC-1 tumor cells. DCs from healthy ani-
mals were used as controls. (Color version of figure is available online).

FOXP3+ cells are predominantly CD4+ T cells committed to exacerba- levels of expression in the T1 and T2 treatments (Figure 6A). Linear
tion of the disease. FOXP3-expressing cells were dramatically regression analysis for CD4+ vs FOXP3 showed a positive correlation
reduced in T3 treatment, with intermediate levels of FOXP3 expres- in DL mice (P = 0.0001) compared with negative or no correlation
sion in T2 treatment. T1 treatment also reduced FOXP3 expression between the same group in T3-treated animals (P = 0.2) (Figure 6B
compared to that in the untreated DL group. We also assessed the and C). Linear regression analysis for CD8+ vs PD-1 indicated a posi-
role of the disease-exacerbating checkpoint inhibitor PD-1 in splenic tive correlation in DL mice (P = 0.0001), whereas a negative correla-
sections from DL mice. Significantly greater expression of PD-1 was tion was documented in T3-treated animals (P = 0.108) (Figure 6D
observed in DL spleen compared with the naive control mice and E). FOXP3 expression was also assessed in CD3+CD25+ T cells of
(Figure 6A). PD-1 expression was significantly downregulated in the DL and treated animals. CD25+/FOXP3+ T cells were significantly
T3 treatment compared with the DL treatment, with intermediate reduced from 48% in the DL group to 18% in the T3-treated group
 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244 1239

Figure 5. Antigen-specific proliferation of CD4+ T cells derived from mice that received binary therapy. (A-E) CD4+ T cells obtained from naive control (Normal), untreated DL mice (DL) or DL
mice treated with T1, T2 or T3 were used as responder cells for proliferation in response to unstimulated or stimulated DCs (Ag or rIL-15) and assessed by MTT assay. Data are presented as
the mean § standard deviation of triplicate determinations in each case. Representative of one experiment out of three performed is shown. (Color version of figure is available online).

(Figure 6F). The reduction in FOXP3 expression in the T1 and T2 show any surge in serum IFN-g levels (Figure 7A). Similar to IFN-g ,
groups was 36% and 22%, respectively, whereas the naive control TNF-a levels also were augmented following T3 therapy compared
group showed 14% positive FOXP3 cells (Figure 6F). with untreated DL mice (P < 0.001). T2 treatment also demonstrated
The expression of PD-1 in CD8+ T cells was significantly aug- a significant surge in TNF-a synthesis compared with the DL group (P
mented from 1.69% in the naive control group to 11.9% in untreated < 0.05) but less than T3 treatment (Figure 7B). In contrast to the
DL animals (Figure 6G). T3 treatment significantly diminished PD-1 aforementioned two cytokines, the serum IL-12 level only surged in
expression, suggesting elimination of disease exacerbating T cells antigen-pulsed DC+NK-cell cell therapy (T2) (P < 0.01), which
following binary therapy (Figure 6G). became more significant in T3 treatment (P < 0.001). DL mice showed
lower IL-12 secretion than naive control mice, indicating severe sup-
Cytokine profiling in binary therapy pression of important inflammatory cytokines that stimulate IFN-g
secretion from CD8+ T cells and NK cells (Figure 7C). In contrast to
Serum levels of inflammatory and anti-inflammatory cytokines in IFN-g and TNF-a, the anti-inflammatory cytokine TGF-b in DL mice
the untreated and treated groups were assessed. T3 treatment dem- was significantly greater compared with treatment in T2 (P < 0.05)
onstrated a significant surge in IFN-g secretion, which was found to or with T3 (P < 0.01) (Figure 7D). In contrast to TGF-b, the IL-6 level
be negligible in DL mice. T2 treatment showed a slight increase, did not show a significant change between DL and T2 or T3 treat-
whereas T1 treatment was unable to make any difference compared ment; however, a significant surge was observed following treatment
with untreated DL mice. As expected, naive control mice did not with T1 (P < 0.001) (Figure 7E). T1 treatment showed a greater level
1240 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244

Figure 6. Analysis of FOXP3 and PD1 in T cells treated with binary therapy. (A) Immunohistochemical localization of CD4+ and CD8+ T cells, FOXP3+Treg cells and PD-1+ exhausted T
cells in the spleen of healthy control, DL tumor-bearing, or treated animals with the indicated dual therapy (magnification £400). (B and C) Linear regression analysis for the rela-
tionship of CD4:FOXP3 ratios in the spleen of the untreated DL and T3 treated groups. (D and E) Linear regression analysis of CD8/PD-1 ratios in the spleen of the untreated DL- and
T3-treated mice. (F) Percentage of CD25+/FOXP3+ T cells in the gated CD3+/CD4+ T cells. (G) Percentage of CD8+/PD-1+ cells in the gated CD3+/CD8+ T-cell populations following the
indicated therapy. (Color version of figure is available online).

of IL-6 secretion, likely due to the effect of DOX, which causes the level considerably but was unable to reach the basal level, suggesting
release of inflammatory cytokines. In contrast to IL-6, serum IL-10 that chemotherapy alone is not sufficient to treat lymphoma (P <
levels were significantly greater in DL mice than in naive control 0.01). Treatment with T2 and T3 dramatically reduced IL-10 levels to
mice (P < 0.001) (Figure 7F). DOX treatment alone reduced the IL-10 basal levels, with better results observed following treatment with
 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244 1241

Figure 7. Profiling of inflammatory and anti-inflammatory cytokines in treated mice. (A-F) Serum levels of IFN-g , TNF-a, IL-2, TGF-b, IL-6 and IL-10 in mice treated with T1, T2 or T3
compared with naive control and untreated DL mice at day 22. Data represent the mean § standard deviation of six mice from three independent experiments. (G) Median survival
of various treatment groups and serum level cytokine responses were analyzed by PCA and (H) correlation matrix. *p < 0.05, **p < 0.01, and ***p < 0.001 value. (Color version of
figure is available online).

T3 (P < 0.0001) (Figure 7F). Our results suggested that therapy with explain huge sources of variation that are available beyond the first
DOX and crude soluble tumor antigen-pulsed DC+ NK cells had a sig- PC and lie orthogonal to it. PCA identified Treg instability and Th1-
nificant impact in lymphoma-bearing mice by reducing disease-exac- type immune responses as potential predictive biomarkers for the
erbating cytokines such as IL-10 and TGF-b with >90% survival. long-term immune protection of DOX+DC(Ag)+NK treatment. To bet-
Animals treated with binary therapy survive without any loss of life ter resolve the specific immune response profile, we performed PCA
up to 60 days when untreated, and DOX-treated mice succumb to using Prism 9 software. In the early-stage analysis, we incorporated
death at days 22 and 29, respectively. data derived from all the experimental groups and explored the sour-
Principal-component analysis (PCA) is an unsupervised ces of variation within the dataset using PCA. This could become
dimensionality-reduction method that generates latent variables indicative about the cytokines accountable for variation between the
designated Principal Components (PC). PC1 indicates linear combina- groups over time, independent of any assumptions about which ani-
tions for individual original variables that package a great source of mal or which cytokine in each observation produced the results. The
variation in a dataset. Subsequent PCs indicate latent variables that loading plots (Figure 7G) represent the contribution made by each
1242 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244

cytokine to the corresponding PC, illustrated in the score plot. The the inhibitory receptor CD96, which is expressed on T cells and NK
loading plots enable us to explore the largest sources of variation cells, the results in increased NK-cell functions and has been reported
within the dataset and identify clusters of cytokines that are closely to act synergistically with anti-CTLA-4 and anti-PD-1 checkpoint
covaried. Three cytokines (IL-6, TNF and IFN) are generally involved blockade [36,37]. DCs have the property to cooperate with NK cells,
in antitumor immunity. These cytokines were located in the same resulting in reciprocal activation of both IFN-g producing NK cells
quadrant for both PCA loading plots, indicating concomitant release. and IL-12-expressing DCs. The innate immune NK-cell repertoire
Conversely, TGF-b and IL-10 are housed in the opposite quadrant of exerts significant effector functions by producing mediators/cyto-
the loading plot (bottom left in Figure 7G). Interestingly, IL-2 kines and by direct lysis of the target cells without previous sensitiza-
appeared adjacent to IFN and therefore covaries with IFN closely. The tion and independent of MHC restriction [5,38,39]. NK cells also have
multivariate correlation matrix also strongly correlated (correlation been reported to increase their presence in the tumor microenviron-
coefficient 0.82) with IFN and TNF responses (Figure 7H). The correla- ment and become activated via enhanced production of FLT3 ligand
tion between IFN and IL-2 was strong (Figure 7H). IL-10 and TGF in cancer cells and in turn exponentially increase protective DCs at
were negatively correlated ( 0.42) with IFN (Figure 7H). Collectively, the site of malignancy [40]. This could be achieved either by skewed
PCA data revealed that induction of a strong Th1 immune response differentiation of DC precursors in the tumor microenvironment or
(IFN-g ) is the best predictor for a durable antilymphoma immune enhanced survival of differentiated DCs, providing increased protec-
response in therapy with DOX+DC(Ag)+NK cells. tion via checkpoint (PD-1/CTLA-4)-mediated immunotherapy [40].
DCs have been reported to interact with a variety of cells, including T
Discussion and Conclusion cells, macrophages and NK cells, in vivo after vaccination [33]. NK
cells eliminate antigen-loaded DCs and facilitate an adaptive T-cell
We have established a comprehensive therapeutic protocol for response [41]. NK cells target immature DCs and can kill them via
the binary application of DOX plus DC+ NK cells in DL mice dur- TRAIL due to reduced expression of MHC-I in immature DCs [33]. DCs
ing the early stage of cancer for better prognosis and results. contribute to cytokine signaling and initiate cell cell contact, which
Although studied well in the literature, immunotherapy has sel- is important for the activation of NK cells. In the coculture of stimu-
dom been given priority as a serious therapeutic option in early- lated DCs and NK cells, mature DCs are activated, and NK-induced
stage cancer. Often, it is considered a last option after several cytotoxicity is augmented. Cytokines secreted following the interac-
rounds of chemotherapy, which is always the number one choice tion of DC NK coculture include IL-2, IL-12, TNF-a and IFN-g also
for therapy. Adoptive DC therapy is considered a last option could inhibit the effect of prostaglandin E on DC NK interactions
when it is too late for any successful outcome. In most cases, DCs [42]. NK cell depletion and reconstitution experiments showed
generated from autologous bone marrow or peripheral blood important role of IFN-g for polarization of T-helper cells upon stimu-
stem cells are given as an alternative option, which generally pro- lation with mature DCs [43]. Activation of NK cells by MHC class I low
duces less than the desired result. In addition, the immune sys- targets prime DCs to produce IL-12 and induces protective CD8+ T
tem becomes impaired and compromised in patients with cell memory responses [44]. Human NK cells can induce membrane-
middle- or advanced-stage cancer, when physiological bound form of IL-15 on DC via IFN-g secretion promoting specific
impairment in immune cells results in anergy and poor antitumor CD8+T lymphocyte response in the absence of DC-derived IL-12 [45].
response. Thus, we believe that the application of DC-mediated Early activation of NK cells and DCs are critical for effective T-cell
immunotherapy in the early stage of the disease may produce response in murine models of WEHI-164 fibrosarcoma and C51 colon
better results and provide an option for the management of the carcinoma, inducing significant reduction of M2-type infiltrating
disease. macrophages [46].
DC NK crosstalk in binary therapy with DOX was studied in Our results demonstrated that DL mice immunized with tumor
experimental metastatic lymphoma to design an effective thera- antigen pulsed and IL-15-stimulated DC+ NK cells enumerate a spe-
peutic protocol. The idea of binary application of DC+NK cells in cific advantage in tackling the disease when given in combination
combination with DOX may have a profound impact in patient with DOX at an early and mid-stages of the disease. Antigen-stimu-
care. This could specifically identify an effective therapeutic pro- lated DC NK coculture could elicit strong cytotoxic T lymphocyte
tocol connecting innate NK cells and antigen-driven DCs against responses [41], which is also consistent with our results; however,
cancers, including lymphoma. The DC NK axis as an immuno- the exact mechanisms require further study. DC NK crosstalk bene-
therapeutic combination in addition to DOX as a coherent strat- fits the respective functions of DCs and NK cells and has potential in
egy for lymphoma treatment has not been tested. DC and NK clinical applications for immunotherapy against lymphoma. When
cells are rare, constituting less than a few percent of the total administered in the simulated early phase of clinical tumorigenesis,
immune population. However, they constitute a dynamic range binary application of DC NKs and a low dose of DOX in an alternate
that predicts the outcome of the disease course. Several earlier schedule resulted in a cure response. In the simulated advanced stage
studies have implicated DC NK interactions in stimulating NK of the disease, the same recipe of binary components does not pro-
cells in vitro, resulting in progressive antitumor immune duce results similar to those observed in early phase treatment. How-
responses in vivo [28,29].This interaction likely results from the ever, it still demonstrated a significant advantage over the singular
chemokine-mediated rendezvous between NK cells and DCs [11], application of either chemotherapy or adoptive cell therapy. One
and it also implies that these cells exist in unique anatomical important advantage of the proposed binary therapy is the reduction
sites, perhaps having affinity for tertiary lymphoid organs. in cardiotoxicity compared with chemotherapy only with DOX with
NK cells are critical components in maintaining homeostasis and significantly low accumulation of the drug in heart tissue (supple-
preventing tumor formation and virus infection [30]. Although NK mentary Figure 5). In the spleen, we also recorded low drug accumu-
cells could eliminate transformed cells without stimulation, several lation in animals treated with binary therapy compared with DOX
studies in animal models of tumorigenesis showed the requirement only (supplementary Figure 5).
of additional activation in addition to NK cells for protection against
lethal tumor invasion [31,32]. The crosstalk between DCs and NK cells
has been closely interdependent, and it has been reported that DC Declaration of Competing Interest
vaccines have the potential to activate NK cells [31,33,34]. It has been
acknowledged that NK cells have a tumoricidal effect on malignant The authors have no commercial, proprietary or financial interest
cells in addition to their ability to kill cancer cells [28,35]. Blocking in the products or companies described in this article.
 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244 1243

Funding [16] Waldmann TA, Tagaya Y. The multifaceted regulation of interleukin-15 expres-
sion and the role of this cytokine in NK cell differentiation and host response to
intracellular pathogens. Annu Rev Immunol 1999;17:19–49. https://doi.org/
This work was supported by a minor IOE grant, No. R/Dev/D/IOE/ 10.1146/annurev.immunol.17.1.19.
Incentive/2021-22/32275 to Partha Pratim Manna from BHU, [17] Ogasawara K, Hida S, Azimi N, Tagaya Y, Sato T, Yokochi-Fukuda T, et al. Require-
Varanasi. ment for IRF-1 in the microenvironment supporting development of natural killer
cells. Nature 1998;391:700–3. https://doi.org/10.1038/35636.
[18] Ohteki T, Yoshida H, Matsuyama T, Duncan GS, Mak TW, Ohashi PS. The transcrip-
Author Contributions tion factor interferon regulatory factor 1 (IRF-1) is important during the matura-
tion of natural killer 1.1+ T cell receptor-alpha/beta+ (NK1+ T) cells, natural killer
cells, and intestinal intraepithelial T cells. J Exp Med 1998;187:967–72. https://
Conception and design of the study: PPM. Acquisition of data: RS, doi.org/10.1084/jem.187.6.967.
UG, PS, AP, UHS and SKH. Analysis and interpretation of data: RS, UG [19] Kennedy MK, Glaccum M, Brown SN, Butz EA, Viney JL, Embers M, et al. Reversible
and SKH. Drafting or revising the manuscript: PPM. All authors have defects in natural killer and memory CD8 T cell lineages in interleukin 15-defi-
cient mice. J Exp Med 2000;191:771–80. https://doi.org/10.1084/jem.191.5.771.
approved the final article.
[20] Lodolce JP, Boone DL, Chai S, Swain RE, Dassopoulos T, Trettin S, et al. IL-15 recep-
tor maintains lymphoid homeostasis by supporting lymphocyte homing and pro-
Acknowledgments liferation. Immunity 1998;9:669–76. https://doi.org/10.1016/s1074-7613(00)
80664-0.
[21] Ohteki T, Suzue K, Maki C, Ota T, Koyasu S. Critical role of IL-15-IL-15R for anti-
RS thanks UGC, India (19/06/2016(i) EU-V) for senior research fel- gen-presenting cell functions in the innate immune response. Nat Immunol
lowship. The authors are thankful to Md. Kashif for help in maintain- 2001;2:1138–43. https://doi.org/10.1038/ni729.
[22] Kuwajima S, Sato T, Ishida K, Tada H, Tezuka H, Ohteki T. Interleukin 15-depen-
ing the mouse colony.
dent crosstalk between conventional and plasmacytoid dendritic cells is essential
for CpG-induced immune activation. Nat Immunol 2006;7:740–6. https://doi.org/
Supplementary materials 10.1038/ni1348.
[23] Hira SK, Mondal I, Bhattacharya D, Manna PP. Downregulation of endogenous
STAT3 augments tumoricidal activity of interleukin 15 activated dendritic cell
Supplementary material associated with this article can be found against lymphoma and leukemia via TRAIL. Exp Cell Res 2014;327:192–208.
in the online version at doi:10.1016/j.jcyt.2022.07.012. https://doi.org/10.1016/j.yexcr.2014.08.012.
[24] Hira SK, Mondal I, Bhattacharya D, Gupta KK, Manna PP. Downregulation of STAT3
phosphorylation enhances tumoricidal effect of IL-15-activated dendritic cell
References against doxorubicin-resistant lymphoma and leukemia via TNF-a. Int J Biochem
Cell Biol 2015;67:1–13. https://doi.org/10.1016/j.biocel.2015.08.002.
[1] Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, Liu YJ, et al. Immunobiology [25] Hira SK, Mondal I, Manna PP. Combined immunotherapy with whole tumor
of dendritic cells. Annu Rev Immunol 2000;18:767–811. https://doi.org/10.1146/ lysate-pulsed interleukin-15-activated dendritic cells and cucurbitacin I pro-
annurev.immunol.18.1.767. motes strong CD8(+) T-cell responses and cures highly aggressive lymphoma.
[2] Rescigno M, Borrow P. The host-pathogen interaction: new themes from den- Cytotherapy 2015;17:647–64. https://doi.org/10.1016/j.jcyt.2015.01.006.
dritic cell biology. Cell 2001;106:267–70. https://doi.org/10.1016/s0092-8674 [26] Gupta U, Hira SK, Singh R, Paladhi A, Srivastava P, Pratim Manna P. Essential role
(01)00454-8. of TNF-a in gamma c cytokine aided crosstalk between dendritic cells and natural
[3] Kokhaei P, Choudhury A, Mahdian R, Lundin J, Moshfegh A, Osterborg A, et al. killer cells in experimental murine lymphoma. Int Immunopharmacol
Apoptotic tumor cells are superior to tumor cell lysate, and tumor cell RNA in 2020;78:106031. https://doi.org/10.1016/j.intimp.2019.106031.
induction of autologous T cell response in B-CLL. Leukemia 2004;18:1810–5. [27] Hira SK, Rej A, Paladhi A, Singh R, Saha J, Mondal I, et al. Galunisertib drives Treg
https://doi.org/10.1038/sj.leu.2403517. fragility and promotes dendritic cell-mediated immunity against experimental
[4] Altfeld M, Fadda L, Frleta D, Bhardwaj N. DCs and NK cells: critical effectors in the lymphoma. IScience 2020;23:101623. https://doi.org/10.1016/j.isci.2020.101623.
immune response to HIV-1. Nat Rev Immunol 2011;11:176–86. https://doi.org/ [28] Zitvogel L. Dendritic and natural killer cells cooperate in the control/switch of
10.1038/nri2935. innate immunity. J Exp Med 2002;195. https://doi.org/10.1084/jem.20012040.
[5] Gerosa F, Baldani-Guerra B, Nisii C, Marchesini V, Carra G, Trinchieri G. Reciprocal F9-14.
activating interaction between natural killer cells and dendritic cells. J Exp Med [29] Fernandez NC, Lozier A, Flament C, Ricciardi-Castagnoli P, Bellet D, Suter M, et al.
2002;195:327–33. https://doi.org/10.1084/jem.20010938. Dendritic cells directly trigger NK cell functions: cross-talk relevant in innate
[6] Broz M, Binnewies M, Boldajipour B, Nelson A, Pollock J, Erle D, et al. Dissecting anti-tumor immune responses in vivo. Nat Med 1999;5:405–11. https://doi.org/
the tumor myeloid compartment reveals rare activating antigen presenting cells, 10.1038/7403.
critical for T cell immunity. Cancer Cell 2014;26:638–52. https://doi.org/10.1016/ [30] Gasser S, Raulet D. The DNA damage response, immunity and cancer. Semin Can-
j.ccell.2014.09.007. cer Biol 2006;16:344–7. https://doi.org/10.1016/j.semcancer.2006.07.004.
[7] Salmon H, Idoyaga J, Rahman A, Leboeuf M, Remark R, Jordan S, et al. Expansion [31] van den Broeke LT, Daschbach E, Thomas EK, Andringa G, Berzofsky JA. Dendritic
and activation of CD103(+) dendritic cell progenitors at the tumor site enhances cell-induced activation of adaptive and innate antitumor immunity. J Immunol
tumor responses to therapeutic PD-L1 and BRAF inhibition. Immunity 2003;171:5842–52. https://doi.org/10.4049/jimmunol.171.11.5842.
2016;44:924–38. https://doi.org/10.1016/j.immuni.2016.03.012. [32] Karimi K, Boudreau JE, Fraser K, Liu H, Delanghe J, Gauldie J, et al. Enhanced anti-
[8] S
anchez-Paulete AR, Cueto FJ, Martínez-Lo
pez M, Labiano S, Morales-Kastresana tumor immunity elicited by dendritic cell vaccines is a result of their ability to
A, Rodríguez-Ruiz ME, et al. Cancer immunotherapy with immunomodulatory engage both CTL and IFN gamma-producing NK cells. Mol Ther 2008;16:411–8.
anti-CD137 and anti-PD-1 monoclonal antibodies requires BATF3-dependent https://doi.org/10.1038/sj.mt.6300347.
dendritic cells. Cancer Discov 2016;6:71–9. https://doi.org/10.1158/2159-8290. [33] Degli-Esposti MA, Smyth MJ. Close encounters of different kinds: dendritic cells
CD-15-0510. and NK cells take centre stage. Nat Rev Immunol 2005;5:112–24. https://doi.org/
[9] Hildner K, Edelson BT, Purtha WE, Diamond M, Matsushita H, Kohyama M, et al. 10.1038/nri1549.
Batf3 deficiency reveals a critical role for CD8alpha+ dendritic cells in cytotoxic T [34] Boudreau JE, Bridle BW, Stephenson KB, Jenkins KM, Brunellie re J, Bramson JL,
cell immunity. Science 2008;322:1097–100. https://doi.org/10.1126/sci- et al. Recombinant vesicular stomatitis virus transduction of dendritic cells
ence.1164206. enhances their ability to prime innate and adaptive antitumor immunity. Mol
[10] Spranger S, Dai D, Horton B, Gajewski TF. Tumor-Residing Batf3 dendritic cells are Ther 2009;17:1465–72. https://doi.org/10.1038/mt.2009.95.
required for effector T cell trafficking and adoptive T cell therapy. Cancer Cell [35] Goding SR, Yu S, Bailey LM, Lotze MT, Basse PH. Adoptive transfer of natural killer
2017;31:711–23. https://doi.org/10.1016/j.ccell.2017.04.003. .e4. cells promotes the anti-tumor efficacy of T cells. Clin Immunol 2017;177:76–86.
[11] Bo€ ttcher JP, Bonavita E, Chakravarty P, Blees H, Cabeza-Cabrerizo M, Sammicheli https://doi.org/10.1016/j.clim.2016.06.013.
S, et al. NK cells stimulate recruitment of cDC1 into the tumor microenvironment [36] Blake SJ, Stannard K, Liu J, Allen S, Yong MCR, Mittal D, et al. Suppression of
promoting cancer immune control. Cell 2018;172:1022–37. https://doi.org/ metastases using a new lymphocyte checkpoint target for cancer immunother-
10.1016/j.cell.2018.01.004. .e14. apy. Cancer Discov 2016;6:446–59. https://doi.org/10.1158/2159-8290.CD-15-
[12] Topalian SL, Drake CG, Pardoll DM. Immune checkpoint blockade: a common 0944.
denominator approach to cancer therapy. Cancer Cell 2015;27:450–61. https:// [37] Dougall WC, Kurtulus S, Smyth MJ, Anderson AC. TIGIT and CD96: new checkpoint
doi.org/10.1016/j.ccell.2015.03.001. receptor targets for cancer immunotherapy. Immunol Rev 2017;276:112–20.
[13] Rizvi NA, Hellmann MD, Snyder A, Kvistborg P, Makarov V, Havel JJ, et al. Muta- https://doi.org/10.1111/imr.12518.
tional landscape determines sensitivity to PD-1 blockade in non small cell lung [38] Piccioli D, Sbrana S, Melandri E, Valiante NM. Contact-dependent stimulation and
cancer. Science 2015;348:124–8. https://doi.org/10.1126/science.aaa1348. inhibition of dendritic cells by natural killer cells. J Exp Med 2002;195:335–41.
[14] Moretta A, Bottino C, Mingari MC, Biassoni R, Moretta L. What is a natural killer https://doi.org/10.1084/jem.20010934.
cell? Nat Immunol 2002;3:6–8. https://doi.org/10.1038/ni0102-6. [39] Ferlazzo G, Tsang ML, Moretta L, Melioli G, Steinman RM, Mu € nz C. Human den-
[15] Langers I, Renoux VM, Thiry M, Delvenne P, Jacobs N. Natural killer cells: role in dritic cells activate resting natural killer (NK) cells and are recognized via the
local tumor growth and metastasis. Biologics 2012;6:73–82. https://doi.org/ NKp30 receptor by activated NK cells. J Exp Med 2002;195:343–51. https://doi.
10.2147/BTT.S23976. org/10.1084/jem.20011149.
1244 R. Singh et al. / Cytotherapy 24 (2022) 1232 1244

[40] Barry KC, Hsu J, Broz ML, Cueto FJ, Binnewies M, Combes AJ, et al. A natural for T(H)1 priming. Nat Immunol 2004;5:1260–5. https://doi.org/10.1038/
killer-dendritic cell axis defines checkpoint therapy-responsive tumor microen- ni1138.
vironments. Nat Med 2018;24:1178–91. https://doi.org/10.1038/s41591-018- [44] Mocikat R, Braumu € ller H, Gumy A, Egeter O, Ziegler H, Reusch U, et al. Natural
0085-8. killer cells activated by MHC class I(low) targets prime dendritic cells to induce
[41] Morandi B, Mortara L, Chiossone L, Accolla RS, Mingari MC, Moretta L, et al. Den- protective CD8 T cell responses. Immunity 2003;19:561–9. https://doi.org/
dritic cell editing by activated natural killer cells results in a more protective can- 10.1016/s1074-7613(03)00264-4.
cer-specific immune response. PLoS One 2012;7:e39170. https://doi.org/10.1371/ [45] Morandi B, Mortara L, Carrega P, Cantoni C, Costa G, Accolla RS, et al. NK cells pro-
journal.pone.0039170. vide helper signal for CD8+ T cells by inducing the expression of membrane-
[42] Van Elssen CHMJ, Vanderlocht J, Oth T, Senden-Gijsbers BLMG, Germeraad WTV, bound IL-15 on DCs. Int Immunol 2009;21:599–606. https://doi.org/10.1093/
Bos GMJ. Inflammation-restraining effects of prostaglandin E2 on natural intimm/dxp029.
killer-dendritic cell (NK-DC) interaction are imprinted during DC maturation. [46] Balza E, Zanellato S, Poggi A, Reverberi D, Rubartelli A, Mortara L. The therapeutic
Blood 2011;118:2473–82. https://doi.org/10.1182/blood-2010-09-307835. T-cell response induced by tumor delivery of TNF and melphalan is dependent on
[43] Martín-Fontecha A, Thomsen LL, Brett S, Gerard C, Lipp M, Lanzavecchia A, et al. early triggering of natural killer and dendritic cells. Eur J Immunol 2017;47:743–
Induced recruitment of NK cells to lymph nodes provides IFN-gamma 53. https://doi.org/10.1002/eji.201646544.