Soypeptide lunasin in cytokine immunotherapy for lymphoma
- 6.8k Downloads
Immunostimulatory cytokines can enhance anti-tumor immunity and are part of the therapeutic armamentarium for cancer treatment. We have previously reported that post-transplant lymphoma patients have an acquired deficiency of signal transducer and activator of transcription 4, which results in defective IFNγ production during clinical immunotherapy. With the goal of further improving cytokine-based immunotherapy, we examined the effects of a soybean peptide called lunasin that synergistically works with cytokines on natural killer (NK) cells. Peripheral blood mononuclear cells of healthy donors and post-transplant lymphoma patients were stimulated with or without lunasin in the presence of IL-12 or IL-2. NK activation was evaluated, and its tumoricidal activity was assessed using in vitro and in vivo tumor models. Chromatin immunoprecipitation assay was performed to evaluate the histone modification of gene loci that are regulated by lunasin and cytokine. Adding lunasin to IL-12- or IL-2-stimulated NK cells demonstrated synergistic effects in the induction of IFNG and GZMB involved in cytotoxicity. The combination of lunasin and cytokines (IL-12 plus IL-2) was capable of restoring IFNγ production by NK cells from post-transplant lymphoma patients. In addition, NK cells stimulated with lunasin plus cytokines displayed higher tumoricidal activity than those stimulated with cytokines alone using in vitro and in vivo tumor models. The underlying mechanism responsible for the effects of lunasin on NK cells is likely due to epigenetic modulation on target gene loci. Lunasin represents a different class of immune modulating agent that may augment the therapeutic responses mediated by cytokine-based immunotherapy.
KeywordsLunasin NK Cytokine immunotherapy Lymphoma
Cytokine immunotherapy is one of the therapeutic strategies to harness the power of immunosurveillance to eradicate cancer cells. Numerous cytokines have been used in clinical trials to enhance anti-tumor immunity . Among these cytokines, IFNγ has been recognized as its pivotal role in anti-tumor immunity by enhancing tumor immunogenicity for antigen presentation [2, 3], inducing apoptosis in tumor cells , promoting T helper 1 (Th1) differentiation [5, 6] and augmenting cytotoxicity of CD8+ cytotoxic T lymphocytes (CTLs) . However, administration of recombinant IFNγ had disappointing outcomes in various cancer immunotherapy trials possibly due to its toxicity . Efforts have focused on induction of endogenous IFNγ by natural killer (NK) or T cells following stimulation with cytokines IL-2, IL-12, IL-15, IL-18, and IL-21, which have been used individually or in different combinations [9, 10, 11, 12, 13]. The successful production of IFNγ is required for the efficacy of several immunotherapeutic approaches including IL-12 immunotherapy [14, 15, 16].
IL-12 can be given in biologically active doses to patients with lymphoma after high-dose chemotherapy followed by autologous peripheral blood stem cell transplantation (PBSCT) [17, 18]. However, these heavily treated patients have acquired deficiency of signal transducer and activator of transcription 4 (STAT4), which contributes to impaired production of IFNγ following IL-12 immunotherapy [19, 20]. We subsequently defined the mechanism in which acquired STAT4 deficiency in T and NK populations is caused by the chemotherapeutic regimen . Failure of patient T or NK cells to adequately produce IFNγ would likely compromise the therapeutic effects during cytokine immunotherapy. The goal of this study was to develop an efficacious immune therapy that would enhance anti-tumor activity for chemotherapy-treated lymphoma patients who acquire immune dysfunctions.
In this study, we have identified a new use for the soypeptide lunasin as an immune modulating agent that synergistically works with therapeutic cytokines to enhance NK-mediated anti-tumor activity. Lunasin is a seed peptide containing 43 amino acids , known for its chemopreventive properties capable of suppressing tumor growth . Our results have demonstrated that lunasin in combination with IL-12 or IL-2 exerts a robust synergistic effect on increasing IFNγ and granzyme B expression by NK cells; and this synergism leads to strong NK activation with enhanced cytotoxicity. Notably, the combination of lunasin and cytokines is capable of rescuing IFNγ production by NK cells from heavily treated lymphoma patients who are immune compromised. Our results suggest promise for lunasin in complementing existing modalities with IL-12 or IL-2 to improve therapeutic responses of cytokine-based cancer immunotherapy.
Materials and methods
Cytokines, antibodies, and lunasin peptides
Recombinant human IL-2 was obtained from Prometheus Laboratories (San Diego, CA) and recombinant human IL-12 from PeproTech (Rocky Hill, NJ). Fluorochrome-conjugated monoclonal antibodies to human CD3, CD4, CD8, CD14, CD56, FasL (CD178), IFNγ, mouse CD3, CD69, NKp46, and human/mouse granzyme B were obtained from BD Biosciences (San Jose, CA). Ficoll-Paque™ PLUS was purchased from GE Healthcare Bio-Sciences (Piscataway, NJ). The lunasin peptide with 43 amino acids was chemically synthesized with 97 % purity by LifeTein (South Plainfield, NJ), which include the following sequences: SKWQHQQDSCRKQLQGVNLTPCEKHIMEKIQGRGDDDDDDDDD. A truncated peptide (32 amino acids) lacking the RGD motif and the poly-D tail was synthesized by LifeTein. A negative control peptide with scrambled sequences (RKMELQEGI HLKKGDQNTQSQSCQPKC IQVWH) that maintains the same molecular weight to the truncated peptide was synthesized. An additional negative control peptide containing an epitope from the influenza matrix protein (M158–66) which binds to human MHC class I molecules was also synthesized. All the peptides were dissolved in sterilized water at stock concentration of 5 mM.
Human blood samples and primary cell cultures
Collection of blood was approved by the Institutional Review Board at Indiana University Medical Center, and written informed consent was obtained from each study subject. Blood samples were obtained from patients with lymphoma after treatment with high-dose chemotherapy and PBSCT. Healthy human blood samples were procured from the Indiana Blood Center (Indianapolis, IN). Peripheral blood mononuclear cells (PBMCs) were isolated using Ficoll-Paque™ PLUS, and aliquots of PBMCs were cryopreserved in liquid nitrogen. Human NK cells were isolated from normal control PBMCs using positive or negative selection kits (Miltenyi Biotech, Auburn, CA). Human B lymphoma cell line Raji cell line was obtained from American Type Culture Collection (ATCC, Manassas, VA, USA).
Evaluation of IFNγ production
IFNγ production at the single-cell levels was evaluated using intracellular cytokine staining from PBMCs following stimulation as indicated . Secreted IFNγ protein collected from the supernatant or mouse serum was measured using ELISA [19, 20].
Analysis of gene expression
Purified human NK cells were stimulated as indicated. One day following stimulation, the cell pellets were subjected to analysis of gene expression using real-time qPCR with Taqman assay primers for IFNG, CSF2, GZMB, TGFB1, and TGFBR2.
The half-maximal effective concentration (EC50) of lunasin
The EC50 of lunasin was calculated from the dose–response curve in IFNγ production using Origin Program (OriginLab, Northampton, MA). The EC50 is presented as mean ± SD averaged from four different normal controls.
In vitro cytotoxicity assays
Purified human NK cells stimulated as indicated for 1 day were washed and co-cultured with target cells (Raji) at the ratio of 10:1 for 4 h at 37 °C in a 5 % CO2 incubator. NK-mediated lysis was analyzed using the CytoTox 96 Non-Radioactive Cytotoxicity Assay Kit (Promega, Madison, WI). For some assays, NK cells were incubated with anti-FasL-blocking antibody (NOK-1 clone, BioLegend) or the isotype control (IgG1) at 10 μg/ml for 2 h  followed by co-culturing with Fas-expressing target cells (Raji).
Adoptive transfer of human NK cells in xenograft model in vivo
NOD/SCID/gcnull (NSG) mice (The Jackson Laboratory, Bar Harbor, Maine) at 2 months old were injected subcutaneously on day 1 with 0.5 × 106 Raji cells in 0.1 ml PBS mixed with 0.1 ml Matrigel (BD Biosciences, San Jose, CA). Human NK cells isolated from healthy control donors were stimulated as indicated for 1 day. On day 2, these pre-treated NK cells were washed and injected into the tumor site (2.5 × 106/mouse). Tumor growth was monitored, and the volumes were measured using standard manual calipers.
NK activation in mice following short-term and long-term treatment
BALB/c mice received short-term (daily single IP injection for 3 consecutive days) or long-term (daily single IP injection for 5 consecutive days per week for a total of 8 weeks) treatment with PBS (−), IL-2 (1 × 105 U/mouse) without (−) or with (+) lunasin (0.4 mg/kg body weight), or lunasin alone. Blood samples were collected by cardiac puncture, and serum levels of IFNγ were determined using ELISA. NK activation was analyzed from spleens in these mice using staining antibodies for surface activation marker CD69 and intracellular granzyme B. IFNγ production by NK cells was evaluated using intracellular staining from splenocytes that were incubated with GolgiPlug (Brefeldin A) for 4 h at 37 °C in a 5 % CO2 incubator followed by flow cytometry analysis.
Chromatin immunoprecipitation (ChIP)
The ChIP experiment was performed using isolated human NK cells treated 1 day as indicated following the established protocol [26, 27]. Antibodies against acetyl-histone H3 (AcH3), histone H3 trimethyl Lys9 (H3K9me3), and non-immune rabbit serum were obtained from Millipore (Billerica, MA).
Analysis of STAT4 activation by Western blot
Purified human NK cells were stimulated for 22 h. Western blot analysis was performed from total protein extracts of cultured NK cells to measure the activation of STAT4 using an anti-phospho-STAT4 (Y693) antibody (Cell Signaling Technology, Danvers, MA). The same blot was reprobed with an anti-STAT4 monoclonal antibody (BD Biosciences, San Jose, CA) for the total amount of STAT4.
SAS/STAT (SAS Institute Inc., Cary, NC) was used to analyze the data. A mixed model was developed for analyzing the data with within-subject treatments, and the pairwise comparisons among the treatments were performed to determine the P values. Statistical significance between groups of mice was determined using an independent sample Student’s t test.
Lunasin stimulates human NK cells to produce IFNγ
Lunasin regulates gene expression by NK cells
Because of robust synergistic effects of lunasin with IL-12 or IL-2 on inducing IFNG expression, we next evaluated whether lunasin was able to modulate other target genes that are regulated by IL-12 or IL-2. Results of qPCR from samples in Fig. 1g showed that adding lunasin to IL-12 or IL-2 significantly increased expression of GZMB (granzyme B) and CSF2 (granulocyte–macrophage colony-stimulating factor or GM-CSF) as compared to treatment with cytokine alone (Fig. 1h). Cytokine IL-12 or IL-2 stimulation is known to downregulate TGFB1 and TGFBR2 expression by NK cells , and adding lunasin to cytokine-treated NK cultures resulted in further reduction of TGFB1 and TGFBR2 expression as compared to treatment with cytokines alone (Fig. 1i). Thus, it appeared that lunasin exerted synergistic effects imposed by the selected cytokine IL-12 or IL-2 on modulating expression of target genes in NK cells.
Dose-dependent effects of lunasin in combination with cytokines
Rescuing IFNγ production by NK cells from lymphoma patients post-transplant
Lunasin augments cytotoxicity by cytokine-activated NK cells
Granzyme B is constitutively expressed by human periphery NK cells, and its level is associated with the lytic activity. IL-2 appeared to induce a higher GZMB by total NK cells as compared to IL-12 (Fig. 1h, upper panel). We next analyzed the effects of suboptimum IL-2 with or without lunasin on the expression of granzyme B using intracellular staining. In concert with the gene expression result, lunasin increased the protein levels of granzyme B in total NK cells cultured in IL-2 (MFI 140 ± 99 for IL-2-treated vs. 163 ± 104 for IL-2+ lunasin-treated NK; mean ± SD averaged from 3 controls). We also found that both CD56 bright and dim populations had a higher granzyme B when lunasin was included in the culture as compared to IL-2 only (Fig. 4b).
To investigate whether lunasin could affect the FasL-induced apoptosis by IL-2-cultured NK cells, we first evaluated the surface expression of FasL following 1 day stimulation. Flow cytometry analysis showed minimum effects of lunasin on surface expression of FasL in NK cells cultured in suboptimum IL-2 (3 ± 1.2 % of FasL + NK in IL-2-treated vs. 2.7 ± 0.8 % in IL-2+ lunasin-treated, n = 2). In addition, cultured NK cells treated with FasL-blocking antibody or the isotype control had similar killing activity against Fas-expressing Raji target cells (Fig. 4c), suggesting that lunasin had no effect on FasL-mediated killing by NK cells activated with suboptimum IL-2.
Cellular therapy using NK cells activated in vitro has been tested clinically against several tumors [31, 32, 33]. In a Raji lymphoma xenograft model, tumor growth increased over time in the control mice without transferred human NK cells (Fig. 4d). While tumor growth was attenuated at day 27 in mice receiving cytokine-activated NK cells, the group receiving NK cells activated with lunasin and cytokine had lowest tumor size when compared with the control group without transferred human NK cells (Fig. 4d).
In vivo effect of lunasin on serum IFNγ secretion
NK activation in vivo following short-term treatment
The effects of lunasin in vivo on NK activation were examined. Given the toxicity-related death caused by cytokine storm with high levels of IFNγ , we thus chose the regimen of IL-2 at suboptimal dose in order to keep animal alive for analysis. We analyzed NK cells gated on CD3−NKp46+ populations (Fig. 5b) from spleens of mice receiving short-term treatment (daily single IP injection for 3 days). The combination of lunasin and IL-2 resulted in significant increase in the percentage of CD69+ or granzyme B + NK cells as compared to those treated with PBS (Fig. 5C, upper and middle panels). Intracellular staining identified NK cells to be responsible for IFNγ production in mice treated with both lunasin and IL-2 (Fig. 5c, lower panel). While IL-2 at suboptimal dose had undetectable effects in this short-term treatment, adding lunasin to IL-2 was able to induce NK activation.
NK activation in vivo following long-term treatment
Following long-term treatment (daily single IP injection for 5 days per week with a total of 8 weeks), these mice exhibited similar percentage of NK cells (CD3−NKp46+) and spleen cellularity (data not shown). Long-term treatment with IL-2 in this setting resulted in NK activation (evidenced by increased expression of CD69, granzyme B, and IFNγ; Fig. 5d) as well as increased levels of serum IFNγ (Fig. 5e). In the presence of lunasin, however, the in vivo effects of IL-2 on NK activation and serum IFNγ production were compromised (Fig. 5d, e).
Mechanisms of synergistic effects mediated by lunasin
It has been shown that lunasin is capable of inhibiting the acetylation of histone H3 by p300/CBP-associated factor (PCAF), a histone acetylase enzyme . Epigenetic regulation by chromatin modification is known to alter gene expression . The effects of lunasin on epigenetic regulation of NK cells were examined. Consistent with gene expression profiles of TGFB1 (Fig. 1i, upper panel), the level of acetylated histone H3 (AcH3) was negatively associated with TGFB1 locus, and less DNA was pulled down by the anti-AcH3 antibody in NK cells treated with lunasin and IL-12 as compared to that with IL-12 alone (Fig. 6b, lower panel). Conversely, the level of AcH3 was positively associated with IFNG locus to a greater degree in NK cells treated with lunasin plus IL-12 than that treated with cytokine alone (Fig. 6b, upper panel). Results suggest the ability of lunasin for modulating the levels of AcH3 bound to the target genes, resulting in gene regulation.
Histone mark with tri-methylated histone H3 at lysine 9 (H3K9me3) is associated with transcriptional repression . We next tested whether changes in the repressive epigenetic marker H3K9me3 were also associated with IL-12-mediated gene regulation in NK cells. Results in Fig. 6c showed a very limited binding of H3K9me3 in the gene loci including TGFB1 and IFNG, suggesting that H3K9me3 was not involved in IL-12-mediated gene regulation in NK cells, and adding lunasin had no effect on levels of H3K9me3 bound to these loci.
STAT4 activation in lunasin-cultured NK cells
STAT4 is required for IL-12-induced IFNγ production by T and NK cells [20, 39, 40], and its activation is involved in H3 hyperacetylation at Ifng locus . In this study, we observed activation of STAT4 in human NK cells following IL-2 stimulation, albeit stronger phospho-STAT4 was induced by IL-12 (Fig. 6d). The levels of pSTAT4 were higher in the cytokine-cultured NK cells containing lunasin (Fig. 6d), suggesting that adding lunasin to cytokine-cultured NK cells enhanced STAT4 activation, which may also contribute to induction of IFNG by NK cells.
Lunasin was first described for its chemopreventive properties that have been demonstrated in cell cultures and mice models . In this study, we have found that lunasin exerts synergistic effects with cytokine IL-12 or IL-2 on modulating expression of a number of genes in NK cells. This synergism results in strong NK activation with enhanced cytotoxicity, which is associated with higher levels of IFNγ and granzyme B expressed by both CD56 bright and dim populations. Adding lunasin to cytokine cocktails with both IL-12 and IL-2 was capable of rescuing the production of IFNγ by NK cells from post-transplant lymphoma patients (Fig. 3), suggesting its potential application as an alternative strategy to improve the clinical outcomes by circumventing chemotherapy-induced immune dysfunction. Taken together, our results demonstrate that the combination of lunasin and selected cytokine (designated as lunakine) is superior to cytokine alone for harnessing NK-mediated anti-tumor functions.
In addition to cytokine stimulation, NK cells can be activated by engagement of co-stimulatory molecules through direct contact with other innate immune cells . Interaction between activated γδT and NK cells increases the cytotoxicity of NK cells via the CD137 engagement . Thus, there is a possibility that γδT or other innate immune cells remaining in the NK cultures following negative selection (97 % purity) may respond to treatment, which in turn activate NK cytotoxicity. Future studies will need to define cell types that can respond to lunasin-based treatment, as well as target genes that are regulated by such treatment. Our results suggest that enhanced cytotoxicity of NK cells following combination treatment with lunasin and cytokine is at least in part due to up-regulation of granzyme B that involves in the granule exocytosis pathway.
Using the Raji B lymphoma xenograft model, we showed that lunakine-treated NK cells could be used in cellular therapy following adoptive transfer (Fig. 4d). However, lunasin did not increase the number of NK cells cultured in cytokines (data not shown), suggesting its limited effect on survival or expansion of NK cells in cellular therapy. Nonetheless, we verified lunasin’s in vivo effect on IFNγ production and NK activation from mice following short-term treatment with IL-2 and lunasin (Fig. 5a–c). The suppressive effect following the long-term treatment (Fig. 5d, e) was not likely caused by the reduced number of NK cells systemically as these mice exhibited similar percentage of NK cells and total spleen cellularity (data not shown). Although the 8-week treatment used in our study is not a current practice, this result raises precautions in the future use of lunasin peptide, in which immune suppression is induced following immune activation in order to maintain homeostasis and avoid the unwanted immune responses. It will be imperative to optimize the doses of each agent and duration of treatment and to evaluate the bioavailability, biodistribution, and pharmacokinetic (PK) profile for lunasin in order to fully realize its clinical potential.
The synergistic effects of lunasin with selected cytokine such as IL-12 are in part due to reducing the levels of acetyl-H3 (Fig. 6b, lower panel), which further suppresses the expression of TGFB1 (Fig. 1i, upper panel). However, acetyl-H3 is associated with IFNG locus to a greater degree in NK cells treated with IL-12 plus lunasin than those treated with IL-12 only (Fig. 6b, upper panel). If the role of lunasin is solely dependent on inhibiting acetylation of histone, how can lunasin coordinately up- and down-regulate expression of different gene loci? One possibility is that lunasin can bind not only deacetylated but also acetylated histone proteins depending upon the preconditioned chromatin structure created by the selected cytokine such as IL-12. To identify histones that can physically interact with lunasin, we screened over 384 unique histone modifications using histone peptide assays (Active Motif, Carlsbad, CA). Indeed, lunasin is capable of binding to deacetylated H3 as well as H3 modifications that are associated with both gene activation and repression (data not shown). It has been shown that the level of acetyl-H3 is augmented following stimulation with IL-12, which contributes to induction of IFNG by NK cells . We thus speculate that lunasin binds to the histone mark (acetyl-H3) created by initial cytokine exposure such as IL-12 , and this binding protects the target gene loci from changing its epigenetic state, and results in maintaining and stabilizing a nucleosome structure favorable for promoting transcription of target genes.
The molecular mechanisms mediated by lunasin require further investigation. Nonetheless, our results suggest that lunasin likely acts as an epigenetic modulator, which synergistically works with cytokines on regulating expression of susceptible genes in NK cells. Our discovery for the novel property of lunasin represents a different class of immune modulating agent that may augment the therapeutic responses by cytokine-based immunotherapy.
This work was supported in part by American Cancer Society IRG (Hua-Chen Chang, Shivani Srivastava), Walther Scholar Grant (Shivani Srivastava) from the Indiana University Simon Cancer Center (P30 CA82709), and National Institutes of Health Grant RO1 CA118118 (Michael J. Robertson). The authors thank the nursing staff in the Indiana CTSI Clinical Research Center for collection of blood samples. We also thank Drs. Randy Brutkiewicz and Mark Kaplan for their invaluable suggestions.
Conflict of interest
- 8.Dunn GP, Koebel CM, Schreiber RD (2006) Interferons, immunity and cancer immunoediting. Nat Rev Immunol 6(11):836–848Google Scholar
- 13.Srivastava S, Salim N, Robertson MJ (2010) Interleukin-18: biology and role in the immunotherapy of cancer. Curr Med Chem 17 (29):3353–3357 Google Scholar
- 21.Lupov IP, Voiles L, Han L, Schwartz A, De La Rosa M, Oza K, Pelloso D, Sahu RP, Travers JB, Robertson MJ, Chang HC (2011) Acquired STAT4 deficiency as a consequence of cancer chemotherapy. Blood 118 (23):6097–6106Google Scholar
- 30.Yu J, Wei M, Becknell B, Trotta R, Liu S, Boyd Z, Jaung MS, Blaser BW, Sun J, Benson DM Jr, Mao H, Yokohama A, Bhatt D, Shen L, Davuluri R, Weinstein M, Marcucci G, Caligiuri MA (2006) Pro- and antiinflammatory cytokine signaling: reciprocal antagonism regulates interferon-gamma production by human natural killer cells. Immunity 24(5):575–590PubMedCrossRefGoogle Scholar
- 31.Miller JS, Soignier Y, Panoskaltsis-Mortari A, McNearney SA, Yun GH, Fautsch SK, McKenna D, Le C, Defor TE, Burns LJ, Orchard PJ, Blazar BR, Wagner JE, Slungaard A, Weisdorf DJ, Okazaki IJ, McGlave PB (2005) Successful adoptive transfer and in vivo expansion of human haploidentical NK cells in patients with cancer. Blood 105(8):3051–3057PubMedCrossRefGoogle Scholar
- 33.Ljunggren HG, Malmberg KJ (2007) Prospects for the use of NK cells in immunotherapy of human cancer. Nat Rev Immunol 7(5):329–339Google Scholar
- 34.Carson WE, Dierksheide JE, Jabbour S, Anghelina M, Bouchard P, Ku G, Yu H, Baumann H, Shah MH, Cooper MA, Durbin J, Caligiuri MA (2000) Coadministration of interleukin-18 and interleukin-12 induces a fatal inflammatory response in mice: critical role of natural killer cell interferon-gamma production and STAT-mediated signal transduction. Blood 96(4):1465–1473PubMedGoogle Scholar
- 42.Fernandez NC, Lozier A, Flament C, Ricciardi-Castagnoli P, Bellet D, Suter M, Perricaudet M, Tursz T, Maraskovsky E, Zitvogel L (1999) Dendritic cells directly trigger NK cell functions: cross-talk relevant in innate anti-tumor immune responses in vivo. Nat Med 5(4):405–411PubMedCrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.