- Email: [email protected]
Cytotoxic T cell responses to Streptococcus are associated with improved prognosis of oral squamous cell carcinoma
Author’s Accepted Manuscript Cytotoxic T cell responses to Streptococcus are associated with improved prognosis of oral squamous cell carcinoma Jie Wang, Feng Sun, Xiaoyu Lin, Zaiye Li, Xiaohe Mao, Canhua Jiang www.elsevier.com/locate/yexcr
PII: DOI: Reference:
S0014-4827(17)30621-3 https://doi.org/10.1016/j.yexcr.2017.11.018 YEXCR10818
To appear in: Experimental Cell Research Received date: 26 September 2017 Revised date: 10 November 2017 Accepted date: 13 November 2017 Cite this article as: Jie Wang, Feng Sun, Xiaoyu Lin, Zaiye Li, Xiaohe Mao and Canhua Jiang, Cytotoxic T cell responses to Streptococcus are associated with improved prognosis of oral squamous cell carcinoma, Experimental Cell Research, https://doi.org/10.1016/j.yexcr.2017.11.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Cytotoxic T cell responses to Streptococcus are associated with improved prognosis of oral squamous cell carcinoma
Jie Wang1,2, Feng Sun2, Xiaoyu Lin1, Zaiye Li1, Xiaohe Mao1, Canhua Jiang1 1Department
of Oral and Maxillofacial Surgery, Xiangya Hospital, Central South
University, Changsha, 410078, China. 2Department
of Immunology, Xiangya School of Medicine,Central South University,
Changsha, 410078,China
Corresponding author: Canhua Jiang Department of Oral and Maxillofacial Surgery, Xiangya Hospital, Central South University, Changsha, 410078, China Tel:+86-731-89753046 Fax:+86-731-89753046 Email: [email protected]
Abstract Several species of Streptococcus, such as S. salivarius, S. mitis, and S. anginosus, are found to extensively colonize the oral cavity and the upper respiratory tract, and have been
shown to increase in patients with oral squamous cell carcinoma (OSCC). Accumulating evidence have revealed that commensal bacteria are involved in antitumor immunity via T cell-mediated mechanisms, but the role of Streptococcus enrichment in OSCC is yet unclear. In this study, we stimulated peripheral blood mononuclear cells from non-cancer controls (NCs) and OSCC patients with S. salivarius, S. mitis, and S. anginosus. We observed that compared to NC subjects, OSCC patients at earlier stages had higher frequencies of granzyme B-expressing CD8 T cells for all Streptococcus species tested, while OSCC patients at more advanced stages had higher frequencies of granzyme B-expressing CD8 T cells for S. anginosus but not other Streptococcus species. In OSCC patients, the Streptococcus-reactive CD8 T cells presented significantly lower levels of PD-1 and TIM-3 expression than Streptococcus-nonreactive CD8 T cells. The clinical outcomes of OSCC patients in our cohort were tracked for 24 months after the resection of the primary tumor. In patients that did not present tumor recurrence, the frequencies of S. salivarius-reactive and S. mitis-reactive CD8 T cells were significantly higher than that in patients that developed recurrent tumor. Furthermore, in patients with tumor recurrence, the duration between primary tumor resection and tumor recurrence was positively associated with the frequencies of S. salivarius-reactive and S. anginosus-reactive CD8 T cells. Together, we demonstrated that Streptococcus-reactive CD8 T cell responses might contribute to antitumor immunity in OSCC patients.
Key words Cytotoxic T cell; bacteria; Streptococcus; cancer; oral squamous cell carcinoma.
Introduction Nearly 400,000 cases of oral squamous cell carcinoma (OSCC) are diagnosed every year, and despite recent advances in surgery, chemotherapy and radiation therapy, the
five-year survival rate remains low at 54%[1,2]. Notably, the prevalence of oral squamous cell carcinoma cannot be attributed to population aging, as the incidence of this disease in young adults (18 - 40 years) is increasing worldwide[3]. Early detection of cancerous legions can increase the cure rate to over 80%, but most cancerous lesions are not detected until the tumor has reached a later stage[2,4]. Moreover, patients who survive the oral cancer are at a significantly elevated risk of developing a second primary cancer at the oral cavity. Treatment and prevention of oral cancer thus remains a significant challenge. In recent years, accumulating evidence suggests that T cell responses toward microbial antigens can present antitumor functions. Commensal bacteria are known to promote the activation of innate antigen-presenting cells and to direct the expansion and differentiation of adaptive T cells[5,6]. Mice hosting different sets of commensal bacteria demonstrated different melanoma growth rates and presented altered tumor-specific T cell responses[7,8]. Bacteria augmented antitumor responses of CD8 T cells, pathogenic Th17 (pTh17) cells and anti-PD-1/PD-L1 immunotherapy[8–10]. Mechanistically, it is thought that various tumor-targeting drugs can cause disruptions to the mucosal epithelium and allow the translocation of commensals. These microbial organisms are then detected by the immune system, resulting in the production of tumor necrosis factor (TNF) and other proinflammatory cytokines with potential antitumor functions, as well as the activation of antitumor pTh17, Th1 and cytotoxic T cell responses[11]. Commensal bacteria may also promote the generation of reactive oxygen species and cause DNA damage in tumor cells[12,13]. Depletion of commensal bacteria abrogated the antitumor efficacy of various tumor-targeting drugs, while adoptive transfer of pTh17 cells partially restored the efficacy[9,10,14]. Many studies have reported that certain bacterial strains could be found in oral cancer lesions and cervical lymph nodes[15–17]. Streptococcus strains are commonly reported to upregulate in oral cancer patients, along with other bacterial species.
Compared to OSCC-free individuals, 80% of OSCC patients presented elevated salivary count of Capnocytophaga gingivalis, Prevotella melaninogenica and Streptococcus mitis[18]. It is currently unknown whether these changes in oral microbial community are incidentally or causally associated with the development of OSCC. Also, it is unclear whether these oral bacteria can contribute to the antitumor responses in OSCC patients. Here, we found that prior to surgical resection, OSCC patients presented an initial upregulation of Streptococcus-reactive cytotoxic T cell response that gradually diminished
with
advancing
tumor,
and
after
resection,
the
frequency
of
Streptococcus-reactive cytotoxic T cell response was positively associated with prognosis. Our findings suggested that Streptococcus-reactive cytotoxic T cell responses presented beneficial roles in OSCC and could potentially serve as a prognostic marker and a treatment strategy.
Methods Subjects All patients were recruited at Xiangya Hospital with the following inclusion criteria: 1) all patients were diagnosed with primary OSCC confirmed by preoperative and postoperative pathologic examination; 2) patients underwent tumor resection; 3) complete follow-up information was available; 4) no other forms of malignancy were found. All OSCC patients were graded using the TNM system in which stages I and II tumors were relatively small tumors that were localized and did not spread to regional lymph nodes or distance sites, whereas stages III and IV tumors were larger and/or had spread to lymph nodes or distant sites[19]. Stages I and II received preoperative chemotherapy and tumor resection. Stages III and IV patients received preoperative chemotherapy, resection of the tumor and/or adjacent sites, and postoperative chemotherapy and/or radiation therapy. Peripheral blood was collected at the time of surgery. Recurrence-free time was defined as the duration between the tumor resection to
recurrence. Non-cancer controls were recruited from age- and sex-matched healthy individuals. All OSCC patients and controls provided written informed consent. Demographic and clinical features of study participants are provided in Table 1.
Bacteria Streptococcus strains, including S. salivarius (ATCC 19258), S. mitis (ATCC 49456), and S. anginosus (ATCC 700231) were obtained from ATCC and stored at -70˚C. For stimulation, an aliquot of bacteria was thawed on agar plates for overnight incubation at 37°C. The bacteria was then inoculated in brain heart infusion broth (ATCC medium 44) for logarithmic growth at 37° C and 5% CO2. The bacterial cells were then washed twice in PBS by centrifugation at 9000 g for 3 min each. Immediately before use, the bacterial cells were resuspended in sterile PBS supplemented with 2% BSA to a final concentration of OD 600 nm = 0.1 in a spectrophotometer.
T cell stimulation Peripheral blood mononuclear cells (PBMCs) were harvested from blood by density gradient separation with Ficoll (Sigma-Aldrich), and were then resuspended to a final concentration of 106 cells per mL in a 6-well plate (Corning). 10 µL of bacteria was then added to each mL of PBMCs. The PBMCs were then incubated for 2 h at 37˚C in a humidified CO2 incubator, and if intracellular staining was to be performed, 5 µg/mL of brefeldin A and 5 µg/mL of monensin (BD Pharmingen) was added 5 h before the end of the incubation. Human CD4 T Cell Enrichment kit and Human CD8 T Cell Enrichment kit (Stemcell) was used to magnetically isolate untouched CD4 T cells and CD8 T cells, respectively.
ELISPOT assay The Human Granzyme B ELISPOT kit (R&D Systems) was used with minor
modifications. Briefly, live purified CD4 T cells and CD8 T cells were counted with Trypan Blue solution (Thermo Fisher) and were plated in Human Granzyme B Microplates (R&D Systems) at concentrations indicated per experiment for 6 h at 37˚C in a humidified CO2 incubator. The rest of the procedures were carried out strictly following the manufacturer’s instructions. All experiments were performed in triplicates. The granzyme B-positive spots were counted automatically in ImmunoSpot S6 MICRO analyzer.
Surface and intracellular staining Anti-human antibodies for CD3, CD4, CD8, PD-1, TIM-3, and granzyme B (eBioscience) were used. PBMCs were first incubated with Fixable Violet or Aqua Dead Cell stains (Invitrogen) for 30 min at 4˚C and washed, and the Fc receptors were blocked using anti-human CD16, CD32 and CD64 (BioLegend). PBMCs were then incubated with surface antibodies for 30 min at 4˚C. For intracellular staining, PBMCs were treated with CytoFix / CytoPerm (BD Pharmingen) and intracellular antibodies following the manufacturer’s instructions. The FlowJo software was used for analysis.
Data analysis Differences between datasets were considered if two-tailed p < 0.05. The tests applied were specified in the figure legend. The Prism software was used for all analysis.
Results Streptococcus-reactive cytotoxic T cells were upregulated in OSCC patients in early stages and downregulated in later stages. Given that increased presence of Streptococcus species was observed across many studies[20], we examined the cytotoxic T cell responses toward various oral strains of streptococci in non-cancer subjects and OSCC patients at various stages. Granzyme B
secretion from CD8 T cells and CD4 T cells was examined by ELISPOT assay and was used as a measure of cytotoxicity. Following stimulation with S. salivarius, a common bacterium that colonizes the mouth and upper respiratory tract, Stage I and Stage II (I+II) OSCC patients presented significantly elevated frequencies of granzyme B-expressing CD8 T cells than both non-cancer control (NC) subjects and Stage III and Stage IV (III+IV) OSCC patients (Figure 1A). No significant difference between NC subjects and III+IV patients was seen in terms of the granzyme B expression by CD8 T cells (Figure 1A). Following stimulation with S. mitis, I+II patients presented significantly higher frequencies of granzyme B-expressing CD8 T cells than NC subjects (Figure 1B). No significant differences between NC subjects and III+IV patients, or between I+II patients and III+IV patients, were found (Figure 1B). Following S. anginosus stimulation, both I+II patients and III+IV patients presented significantly higher frequencies of granzyme B-expressing CD8 T cells than NC subjects, while no significant difference between I+II and III+IV patients was observed (Figure 1C). CD4 T cells from NC subjects, I+II patients and III+IV patients did not present significant levels of granzyme B expression following Streptococcus stimulation (data not shown). To examine whether OSCC patients simply presented higher total CD8 T cell activation than NC subjects regardless of antigen, or whether the higher frequencies of granzyme B-expressing CD8 T cells were restricted to Streptococcus antigens, we used CD3/CD28 activator beads to stimulate PBMCs from NC subjects and OSCC patients. In contrast to the observations following Streptococcus stimulation, the NC subjects after CD3/CD28 activation presented significantly higher frequencies of granzyme B-expressing CD8 T cells than I+II and III+IV patients (Figure 2). I+II patients also presented higher frequencies of granzyme B-expressing CD8 T cells than III+IV patients (Figure 2).
Streptococcus-reactive CD8 T cells presented reduced PD-1 and TIM-3 expression than
total CD8 T cells PD-1 and TIM-3 are inhibitory surface molecules that contribute to cell apoptosis and anergy[21,22]. CD8 T cells that express high levels of PD-1 and/or TIM-3 are present at elevated levels in patients with chronic virus infections or cancer, and tend to display an exhausted functional profile. Using flow cytometry, we investigated PD-1 and TIM-3 expression in Streptococcus-reactive CD8 T cells. Granzyme B-expressing (positive) and granzyme B-non-expressing (negative) CD8 T cells were identified by intracellular staining (Figure 3A), and the levels of PD-1 and TIM-3 expression on the surface was then examined on granzyme B-positive vs. granzyme B-negative CD8 T cells (Figure 3B). Only OSCC individuals with greater than 1% of CD8 T cells expressing granzyme B (identified using flow cytometry as shown in Figure 2A) were analyzed because a sizable cell population was required for flow cytometry analysis. We observed that after Streptococcus stimulation, all the granzyme B-positive CD8 T cells from OSCC patients with greater than 1% granzyme B-expressing CD8 T cells presented significantly lower PD-1 than the granzyme B-negative CD8 T cells from the same individuals (Figure 3C). Similarly, the granzyme B-positive CD8 T cells also presented lower TIM-3 expression than the granzyme B-negative CD8 T cells from the same individuals (Figure 3D). For comparison, we also examined the PD-1 and TIM-3 expression levels in granzyme B-expressing and granzyme B-non-expressing CD8 T cells after CD3/CD28 activation using flow cytometry (Figure 4A and 4B). After CD3/CD28 activation, the granzyme B-positive CD8 T cells from OSCC patients presented significantly lower surface PD-1 expression than the granzyme B-negative CD8 T cells from the same individuals (Figure 4C), but unlike the case with Streptococcus stimulation in which granzyme B-positive CD8 T cells presented uniformly lower PD-1 expression, the granzyme B-positive CD8 T cells after CD3/CD28 activation presented higher PD-1 than granzyme B-negative CD8 T cells in a few OSCC subjects. In terms of surface TIM-3 expression, no statistically significant difference was observed between granzyme
B-positive CD8 T cells and granzyme B-negative CD8 T cells (Figure 4D).
The level of Streptococcus-reactive cytotoxic T cells was positively associated with prognosis To examine the role of Streptococcus-reactive cytotoxic T cell responses in OSCC patients, we examined the association between the frequencies of Streptococcus-reactive granzyme B-expressing CD8 T cells and the prognosis of OSCC patients. At 24 months following tumor resection, 11 out of 35 patients (31.4%) presented recurrent tumor. We found that OSCC patients without recurrence presented significantly higher S. salivariusand S. mitis-reactive granzyme B-expressing CD8 T cells than OSCC patients without recurrence (Figure 5A). In patients with recurrence after tumor resection, we examined the frequencies of Streptococcus-reactive granzyme B-expressing CD8 T cells with the recurrence-free time. The length of recurrence-free time was positively associated with the frequencies of S. salivarius- and S. anginosus-reactive granzyme B-expressing CD8 T cells (Figure 5B).
Discussion A role of commensal bacteria in modulating the human immune responses in various inflammatory diseases is being increasingly recognized. At resting state, the mucosal surfaces, where commensal and pathogenic microorganisms may colonize, are protected by intact epithelial barriers, but during an inflammatory response, the epithelial barriers can be disrupted by inflammation and/or drugs, allowing increased contact between microorganisms and the host immune system. In some cases, pathogenic inflammation caused by bacterial colonization and infiltration may contribute to the development of certain
cancers,
such
as
Helicobacter
pylori-associated
gastric
cancer
and
colitis-associated cancer[23]. However, recent investigations found that bacteria-induced inflammation might also participate in antitumor immunity[8–10,13]. In this study, we
provided evidence that bacteria-reactive responses might be a constituent of the antitumor immunity in OSCC patients. Upregulation of Streptococcus strains in OSCC patients were identified across several studies[18,20,24]. To examine their association with the host T cell responses, we used several common Streptococcus species, including S. salivarius, S. mitis, and S. anginosus, to stimulate the PBMCs from NC subjects and OSCC patients. Compared to NC subjects, OSCC patients at earlier stages had higher frequencies of granzyme B-expressing CD8 T cells for all Streptococcus species tested, while OSCC patients at more advanced stages had higher frequencies of granzyme B-expressing CD8 T cells for S. anginosus but not other Streptococcus species. These observations suggested that the Streptococcus-reactive response in OSCC presented an initial upregulation, which was followed by a downregulation. These results might be explained by several non-mutually exclusive mechanisms. First, the elevation of Streptococcus-reactive cytotoxic CD8 T cells might be a corollary of increased Streptococcus colonization. It has been shown that OSCC patients presented increased levels of Streptococci in the salivary microbiota than healthy individuals, and Streptococcus has been identified in resected tumor tissues[18,24]. But within the OSCC group, it is yet unclear whether OSCC individuals at more advanced stages presented lower Streptococcus colonization than OSCC individuals at earlier stages. Second, the downregulation of Streptococcus-reactive cytotoxic CD8 T cells might be a corollary of an overall impairment in CD8 T cell functionality in advanced cancer patients. Supporting this idea, we found that OSCC patients at later stages presented lower granzyme B response than OSCC patients at earlier stages and NC subjects. We subsequently examined the expression of PD-1 and TIM-3, two inhibitory checkpoint markers previously associated with CD8 T cell exhaustion. We found that compared to CD8 T cells that did not respond to Streptococcus stimulation, the CD8 T cells that responded to Streptococcus stimulation by granzyme B secretion presented
lower PD-1 and lower TIM-3 uniformly. This observation first appeared obvious, since less-exhausted CD8 T cells are more likely to present potent cytotoxicity than more-exhausted CD8 T cells. However, we subsequently examined the PD-1 and TIM-3 expression in granzyme B-expressing and granzyme B-non-expressing CD8 T cells following CD3/CD28 stimulation. The CD8 T cells that expressed granzyme B did not present uniformly lower PD-1 than the CD8 T cells that did not express granzyme B. Furthermore, no differences in TIM-3 expression was found between the two groups. Therefore, more research is required to investigate whether bacteria such as Streptococcus could promote T cell responses via mechanisms that result in the downregulation of inhibitory checkpoint molecules. It should also be examined whether blocking PD-1 and TIM-3 could improve the Streptococcus-reactive cytotoxic responses of the CD8 T cells. We also found that the frequencies of Streptococcus-reactive CD8 T cell-mediated granzyme B responses were positively correlated with prognosis. Whether a causal relationship exists between Streptococcus-reactive granzyme B-expressing CD8 T cell responses and clinical outcomes of OSCC patients still require further analysis. If Streptococcus-reactive CD8 T cell can promote antitumor immunity, one can expect that such properties are transferable with the CD8 T cells between different individuals, and Streptococcus strains can be used in the development of future immunotherapies. Extensive animal investigations are required to investigate these hypotheses.
Acknowledgments This work was supported by National Natural Science Foundation of China (No.: 81271154).
Conflict of interest None.
Figure 1. Frequencies of granzyme B-expressing CD8 T cells following Streptococci stimulation. PBMCs were incubated with (A) S. salivarius, (B) S. mitis, or (C) S. anginosus, for 2 days. CD8 T cells was then isolated from Streptococcus-stimulated PBMCs and plated in ELISPOT plates at 1,000 to 10,000 cells per well for 6 h. The percentage of granzyme B-expressing cells was calculated by the number of positive spots / the number of loaded cells * 100%. Kruskal Wallis ANOVA followed by Dunn’s post-test. * p < 0.05. ** p < 0.01.
Figure 2. Frequencies of total granzyme B-expressing CD8 T cells following CD3/CD28 stimulation. PBMCs were incubated with CD3/CD28 activator beads for 2 days. CD8 T cells was then isolated from stimulated PBMCs and plated in ELISPOT plates at 100 to 1,000 cells per well for 6 h. The percentage of granzyme B-expressing cells was calculated by the number of positive spots / the number of loaded cells * 100%. Kruskal Wallis ANOVA followed by Dunn’s post-test. * p < 0.05. *** p < 0.001.
Figure
3.
Expression
of
PD-1
and
TIM-3
in
Streptococcus-reactive
and
Streptococcus-nonreactive cytotoxic CD8 T cells. PBMCs were incubated with S. salivarius, S. mitis, or S. anginosus for 2 days and with brefeldin A and monensin for the final 5 h. Cells were then labeled with fluorescent antibodies for flow cytometry analysis. (A) The gating of granzyme B (GB)-positive and GB-negative cells in CD8 T cells in one representative OSCC individual following S. salivarius stimulation. Dots shown were pregated on live single CD8 T cells. (B) The surface expression of PD-1 (left) and TIM-3 (right) in GB-positive (black) and GB-negative (grey) cells. (C) The mean fluorescence intensity (MFI) of PD-1 on GB-positive and GB-negative CD8 T cells from OSCC individuals following S. salivarius, S. mitis, and S. anginosus stimulation. (D) The MFI of TIM-3 on GB-positive and GB-negative CD8 T cells from OSCC individuals
following S. salivarius, S. mitis, and S. anginosus stimulation. (C) and (D) Wilcoxon matched pairs test. * p < 0.05. ** p < 0.01.
Figure 4. Expression of PD-1 and TIM-3 in granzyme B-positive and granzyme B-negative CD8 T cells following CD3/CD28 stimulation. PBMCs were incubated with CD3/CD28 activator beads for 2 days and with brefeldin A and monensin for the final 5 h. Cells were then labeled with fluorescent antibodies for flow cytometry analysis. (A) The gating of GB-positive and GB-negative cells in CD8 T cells in one representative OSCC individual following CD3/CD28 activation. Dots shown were pregated on live single CD8 T cells. (B) The surface expression of PD-1 (left) and TIM-3 (right) in GB-positive (black) and GB-negative (grey) cells. (C) The mean fluorescence intensity (MFI) of PD-1 on GB-positive and GB-negative CD8 T cells from OSCC individuals following CD3/CD28 activation. (D) The MFI of TIM-3 on GB-positive and GB-negative CD8 T cells from OSCC individuals following CD3/CD28 activation. (C) and (D) Wilcoxon matched pairs test. * p < 0.05.
Figure
5.
Association
between
OSCC
prognosis
and
the
frequencies
of
Streptococcus-reactive CD8 T cells. The recurrence status was followed in OSCC patients for 24 months after tumor resection. (A) The frequencies of granzyme B-expressing CD8 T cells after S. salivarius, S. mitis, and S. anginosus stimulation in OSCC patients with recurrence vs. in OSCC patients without recurrence. Wilcoxon matched pairs test. (B) The correlation between the frequencies of granzyme B-expressing CD8 T cells and the recurrence-free time in patients with recurrence. Pearson correlation test. * p < 0.05. ** p < 0.01.
References [1]
M.D. Mignogna, S. Fedele, L. Lo Russo, The World Cancer Report and the burden
of oral cancer., Eur. J. Cancer Prev. 13 (2004) 139–142. doi:10.1097/00008469-200404000-00008. [2]
S. SILVERMAN, Demographics and occurrence of oral and pharyngeal cancers, J. Am. Dent. Assoc. 132 (2001) 7S–11S. doi:10.14219/jada.archive.2001.0382.
[3]
J.N. Myers, T. Elkins, D. Roberts, R.M. Byers, Squamous cell carcinoma of the tongue in young adults: increasing incidence and factors that predict treatment outcomes, Otolaryngol. Neck Surg. 122 (2000) 44–51.
[4]
J.D. Holmes, E.J. Dierks, L.D. Homer, B.E. Potter, Is detection of oral and oropharyngeal squamous cancer by a dental health care provider associated with a lower stage at diagnosis?, J. Oral Maxillofac. Surg. 61 (2003) 285–291. doi:10.1053/joms.2003.50056.
[5]
L. V. Hooper, D.R. Littman, A.J. Macpherson, Interactions Between the Microbiota and the Immune System, Science (80-. ). 336 (2012) 1268–1273. doi:10.1126/science.1223490.
[6]
M.C. Abt, L.C. Osborne, L.A. Monticelli, T.A. Doering, T. Alenghat, G.F. Sonnenberg, et al., Commensal Bacteria Calibrate the Activation Threshold of Innate Antiviral Immunity, Immunity. 37 (2012) 158–170. doi:10.1016/j.immuni.2012.04.011.
[7]
I.I. Ivanov, K. Atarashi, N. Manel, E.L. Brodie, T. Shima, U. Karaoz, et al., Induction of Intestinal Th17 Cells by Segmented Filamentous Bacteria, Cell. 139 (2009) 485–498. doi:10.1016/j.cell.2009.09.033.
[8]
A. Sivan, L. Corrales, N. Hubert, J.B. Williams, K. Aquino-Michaels, Z.M. Earley, et al., Commensal Bifidobacterium promotes antitumor immunity and facilitates anti-PD-L1 efficacy, Science (80-. ). 350 (2015) 1084–1089. doi:10.1126/science.aac4255.
[9]
S. Viaud, F. Saccheri, G. Mignot, T. Yamazaki, R. Daillère, D. Hannani, et al., The intestinal microbiota modulates the anticancer immune effects of
cyclophosphamide., Science. 342 (2013) 971–6. doi:10.1126/science.1240537. [10]
R. Daillère, M. Vétizou, N. Waldschmitt, T. Yamazaki, C. Isnard, V. Poirier-Colame, et al., Enterococcus hirae and Barnesiella intestinihominis Facilitate Cyclophosphamide-Induced Therapeutic Immunomodulatory Effects, Immunity. 45 (2016) 931–943. doi:10.1016/j.immuni.2016.09.009.
[11]
S. Roy, G. Trinchieri, Microbiota: a key orchestrator of cancer therapy, Nat. Rev. Cancer. 17 (2017) 271–285. doi:10.1038/nrc.2017.13.
[12]
Q.F. Gui, H.F. Lu, C.X. Zhang, Z.R. Xu, Y.M. Yang, Well-balanced commensal microbiota contributes to anti-cancer response in a lung cancer mouse model, Genet. Mol. Res. 14 (2015) 5642–5651. doi:10.4238/2015.May.25.16.
[13] N. Iida, A. Dzutsev, C.A. Stewart, L. Smith, N. Bouladoux, R.A. Weingarten, et al., Commensal Bacteria Control Cancer Response to Therapy by Modulating the Tumor Microenvironment, Science (80-. ). 342 (2013) 967–970. doi:10.1126/science.1240527. [14]
L. Zitvogel, L. Galluzzi, S. Viaud, M. Vétizou, R. Daillère, M. Merad, et al., Cancer and the gut microbiota: an unexpected link., Sci. Transl. Med. 7 (2015) 271ps1. doi:10.1126/scitranslmed.3010473.
[15]
K. Shiga, Presence of Streptococcus infection in extra-oropharyngeal head and neck squamous cell carcinoma and its implication in carcinogenesis, Oncol. Rep. 8 (2001) 245–248.
[16]
K.N. Nagy, I. Sonkodi, I. Szöke, E. Nagy, H.N. Newman, The microflora associated with human oral carcinomas, Oral Oncol. 34 (1998) 304–308. doi:10.1016/S1368-8375(98)00008-6.
[17]
S.J. Hooper, S.J. Crean, M.J. Fardy, M.A. Lewis, D.A. Spratt, W.G. Wade, et al., A molecular analysis of the bacteria present within oral squamous cell carcinoma, J Med Microbiol. 56 (2007) 1651–1659. doi:10.1099/jmm.0.46918-0.
[18] D.L. Mager, A.D. Haffajee, P.M. Devlin, C.M. Norris, M.R. Posner, J.M. Goodson,
The salivary microbiota as a diagnostic indicator of oral cancer: a descriptive, non-randomized study of cancer-free and oral squamous cell carcinoma subjects., J. Transl. Med. 3 (2005) 27. doi:10.1186/1479-5876-3-27. [19]
B.W. Neville, T. a. Day, Oral Cancer and Precancerous Lesions, CA. Cancer J. Clin. 52 (2002) 195–215. doi:10.3322/canjclin.52.4.195.
[20]
N. Khajuria, R. Metgud, Role of bacteria in oral carcinogenesis., Indian J. Dent. 6 (2015) 37–43. doi:10.4103/0975-962X.151709.
[21]
M.E. Keir, M.J. Butte, G.J. Freeman, A.H. Sharpe, PD-1 and its ligands in tolerance and immunity., Annu. Rev. Immunol. 26 (2008) 677–704. doi:10.1146/annurev.immunol.26.021607.090331.
[22]
A.C. Anderson, Tim-3: an emerging target in the cancer immunotherapy landscape., Cancer Immunol. Res. 2 (2014) 393–8. doi:10.1158/2326-6066.CIR-14-0039.
[23]
A. Dzutsev, R.S. Goldszmid, S. Viaud, L. Zitvogel, G. Trinchieri, The role of the microbiota in inflammation, carcinogenesis, and cancer therapy, Eur. J. Immunol. 45 (2015) 17–31. doi:10.1002/eji.201444972.
[24]
M. Sasaki, C. Yamaura, Y. Ohara-Nemoto, S. Tajika, Y. Kodama, T. Ohya, et al., Streptococcus anginosus infection in oral cancer and its infection route., Oral Dis. 11 (2005) 151–156. doi:10.1111/j.1601-0825.2005.01051.x.
Table 1. Demographic and clinical information of the study participants. Healthy
Patient (Stages I Patient + II)
III + IV)
(Stages p-value
Number of patients
16
19
16
Female (N, %)
6 (37%)
7 (37%)
5 (31%)
> 0.05
Age in years
61.0
61.4
63.8
> 0.05
(median, range)
(55.3 - 65.2)
(54.1 - 66.3)
(57.8 - 67.7) >0.05
Location Tongue
11
8
Oral floor
4
5
Lower gingiva
4
3 < 0.001
Tumor size T1
5
0
T2
12
1
T3
0
9
T4
0
6 < 0.01
Lymph node metastasis N0
19
8
N1
0
4
N2
0
4 > 0.05
Distant metastasis Negative
19
16
Positive
0
0
PII: DOI: Reference:
S0014-4827(17)30621-3 https://doi.org/10.1016/j.yexcr.2017.11.018 YEXCR10818
To appear in: Experimental Cell Research Received date: 26 September 2017 Revised date: 10 November 2017 Accepted date: 13 November 2017 Cite this article as: Jie Wang, Feng Sun, Xiaoyu Lin, Zaiye Li, Xiaohe Mao and Canhua Jiang, Cytotoxic T cell responses to Streptococcus are associated with improved prognosis of oral squamous cell carcinoma, Experimental Cell Research, https://doi.org/10.1016/j.yexcr.2017.11.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Cytotoxic T cell responses to Streptococcus are associated with improved prognosis of oral squamous cell carcinoma
Jie Wang1,2, Feng Sun2, Xiaoyu Lin1, Zaiye Li1, Xiaohe Mao1, Canhua Jiang1 1Department
of Oral and Maxillofacial Surgery, Xiangya Hospital, Central South
University, Changsha, 410078, China. 2Department
of Immunology, Xiangya School of Medicine,Central South University,
Changsha, 410078,China
Corresponding author: Canhua Jiang Department of Oral and Maxillofacial Surgery, Xiangya Hospital, Central South University, Changsha, 410078, China Tel:+86-731-89753046 Fax:+86-731-89753046 Email: [email protected]
Abstract Several species of Streptococcus, such as S. salivarius, S. mitis, and S. anginosus, are found to extensively colonize the oral cavity and the upper respiratory tract, and have been
shown to increase in patients with oral squamous cell carcinoma (OSCC). Accumulating evidence have revealed that commensal bacteria are involved in antitumor immunity via T cell-mediated mechanisms, but the role of Streptococcus enrichment in OSCC is yet unclear. In this study, we stimulated peripheral blood mononuclear cells from non-cancer controls (NCs) and OSCC patients with S. salivarius, S. mitis, and S. anginosus. We observed that compared to NC subjects, OSCC patients at earlier stages had higher frequencies of granzyme B-expressing CD8 T cells for all Streptococcus species tested, while OSCC patients at more advanced stages had higher frequencies of granzyme B-expressing CD8 T cells for S. anginosus but not other Streptococcus species. In OSCC patients, the Streptococcus-reactive CD8 T cells presented significantly lower levels of PD-1 and TIM-3 expression than Streptococcus-nonreactive CD8 T cells. The clinical outcomes of OSCC patients in our cohort were tracked for 24 months after the resection of the primary tumor. In patients that did not present tumor recurrence, the frequencies of S. salivarius-reactive and S. mitis-reactive CD8 T cells were significantly higher than that in patients that developed recurrent tumor. Furthermore, in patients with tumor recurrence, the duration between primary tumor resection and tumor recurrence was positively associated with the frequencies of S. salivarius-reactive and S. anginosus-reactive CD8 T cells. Together, we demonstrated that Streptococcus-reactive CD8 T cell responses might contribute to antitumor immunity in OSCC patients.
Key words Cytotoxic T cell; bacteria; Streptococcus; cancer; oral squamous cell carcinoma.
Introduction Nearly 400,000 cases of oral squamous cell carcinoma (OSCC) are diagnosed every year, and despite recent advances in surgery, chemotherapy and radiation therapy, the
five-year survival rate remains low at 54%[1,2]. Notably, the prevalence of oral squamous cell carcinoma cannot be attributed to population aging, as the incidence of this disease in young adults (18 - 40 years) is increasing worldwide[3]. Early detection of cancerous legions can increase the cure rate to over 80%, but most cancerous lesions are not detected until the tumor has reached a later stage[2,4]. Moreover, patients who survive the oral cancer are at a significantly elevated risk of developing a second primary cancer at the oral cavity. Treatment and prevention of oral cancer thus remains a significant challenge. In recent years, accumulating evidence suggests that T cell responses toward microbial antigens can present antitumor functions. Commensal bacteria are known to promote the activation of innate antigen-presenting cells and to direct the expansion and differentiation of adaptive T cells[5,6]. Mice hosting different sets of commensal bacteria demonstrated different melanoma growth rates and presented altered tumor-specific T cell responses[7,8]. Bacteria augmented antitumor responses of CD8 T cells, pathogenic Th17 (pTh17) cells and anti-PD-1/PD-L1 immunotherapy[8–10]. Mechanistically, it is thought that various tumor-targeting drugs can cause disruptions to the mucosal epithelium and allow the translocation of commensals. These microbial organisms are then detected by the immune system, resulting in the production of tumor necrosis factor (TNF) and other proinflammatory cytokines with potential antitumor functions, as well as the activation of antitumor pTh17, Th1 and cytotoxic T cell responses[11]. Commensal bacteria may also promote the generation of reactive oxygen species and cause DNA damage in tumor cells[12,13]. Depletion of commensal bacteria abrogated the antitumor efficacy of various tumor-targeting drugs, while adoptive transfer of pTh17 cells partially restored the efficacy[9,10,14]. Many studies have reported that certain bacterial strains could be found in oral cancer lesions and cervical lymph nodes[15–17]. Streptococcus strains are commonly reported to upregulate in oral cancer patients, along with other bacterial species.
Compared to OSCC-free individuals, 80% of OSCC patients presented elevated salivary count of Capnocytophaga gingivalis, Prevotella melaninogenica and Streptococcus mitis[18]. It is currently unknown whether these changes in oral microbial community are incidentally or causally associated with the development of OSCC. Also, it is unclear whether these oral bacteria can contribute to the antitumor responses in OSCC patients. Here, we found that prior to surgical resection, OSCC patients presented an initial upregulation of Streptococcus-reactive cytotoxic T cell response that gradually diminished
with
advancing
tumor,
and
after
resection,
the
frequency
of
Streptococcus-reactive cytotoxic T cell response was positively associated with prognosis. Our findings suggested that Streptococcus-reactive cytotoxic T cell responses presented beneficial roles in OSCC and could potentially serve as a prognostic marker and a treatment strategy.
Methods Subjects All patients were recruited at Xiangya Hospital with the following inclusion criteria: 1) all patients were diagnosed with primary OSCC confirmed by preoperative and postoperative pathologic examination; 2) patients underwent tumor resection; 3) complete follow-up information was available; 4) no other forms of malignancy were found. All OSCC patients were graded using the TNM system in which stages I and II tumors were relatively small tumors that were localized and did not spread to regional lymph nodes or distance sites, whereas stages III and IV tumors were larger and/or had spread to lymph nodes or distant sites[19]. Stages I and II received preoperative chemotherapy and tumor resection. Stages III and IV patients received preoperative chemotherapy, resection of the tumor and/or adjacent sites, and postoperative chemotherapy and/or radiation therapy. Peripheral blood was collected at the time of surgery. Recurrence-free time was defined as the duration between the tumor resection to
recurrence. Non-cancer controls were recruited from age- and sex-matched healthy individuals. All OSCC patients and controls provided written informed consent. Demographic and clinical features of study participants are provided in Table 1.
Bacteria Streptococcus strains, including S. salivarius (ATCC 19258), S. mitis (ATCC 49456), and S. anginosus (ATCC 700231) were obtained from ATCC and stored at -70˚C. For stimulation, an aliquot of bacteria was thawed on agar plates for overnight incubation at 37°C. The bacteria was then inoculated in brain heart infusion broth (ATCC medium 44) for logarithmic growth at 37° C and 5% CO2. The bacterial cells were then washed twice in PBS by centrifugation at 9000 g for 3 min each. Immediately before use, the bacterial cells were resuspended in sterile PBS supplemented with 2% BSA to a final concentration of OD 600 nm = 0.1 in a spectrophotometer.
T cell stimulation Peripheral blood mononuclear cells (PBMCs) were harvested from blood by density gradient separation with Ficoll (Sigma-Aldrich), and were then resuspended to a final concentration of 106 cells per mL in a 6-well plate (Corning). 10 µL of bacteria was then added to each mL of PBMCs. The PBMCs were then incubated for 2 h at 37˚C in a humidified CO2 incubator, and if intracellular staining was to be performed, 5 µg/mL of brefeldin A and 5 µg/mL of monensin (BD Pharmingen) was added 5 h before the end of the incubation. Human CD4 T Cell Enrichment kit and Human CD8 T Cell Enrichment kit (Stemcell) was used to magnetically isolate untouched CD4 T cells and CD8 T cells, respectively.
ELISPOT assay The Human Granzyme B ELISPOT kit (R&D Systems) was used with minor
modifications. Briefly, live purified CD4 T cells and CD8 T cells were counted with Trypan Blue solution (Thermo Fisher) and were plated in Human Granzyme B Microplates (R&D Systems) at concentrations indicated per experiment for 6 h at 37˚C in a humidified CO2 incubator. The rest of the procedures were carried out strictly following the manufacturer’s instructions. All experiments were performed in triplicates. The granzyme B-positive spots were counted automatically in ImmunoSpot S6 MICRO analyzer.
Surface and intracellular staining Anti-human antibodies for CD3, CD4, CD8, PD-1, TIM-3, and granzyme B (eBioscience) were used. PBMCs were first incubated with Fixable Violet or Aqua Dead Cell stains (Invitrogen) for 30 min at 4˚C and washed, and the Fc receptors were blocked using anti-human CD16, CD32 and CD64 (BioLegend). PBMCs were then incubated with surface antibodies for 30 min at 4˚C. For intracellular staining, PBMCs were treated with CytoFix / CytoPerm (BD Pharmingen) and intracellular antibodies following the manufacturer’s instructions. The FlowJo software was used for analysis.
Data analysis Differences between datasets were considered if two-tailed p < 0.05. The tests applied were specified in the figure legend. The Prism software was used for all analysis.
Results Streptococcus-reactive cytotoxic T cells were upregulated in OSCC patients in early stages and downregulated in later stages. Given that increased presence of Streptococcus species was observed across many studies[20], we examined the cytotoxic T cell responses toward various oral strains of streptococci in non-cancer subjects and OSCC patients at various stages. Granzyme B
secretion from CD8 T cells and CD4 T cells was examined by ELISPOT assay and was used as a measure of cytotoxicity. Following stimulation with S. salivarius, a common bacterium that colonizes the mouth and upper respiratory tract, Stage I and Stage II (I+II) OSCC patients presented significantly elevated frequencies of granzyme B-expressing CD8 T cells than both non-cancer control (NC) subjects and Stage III and Stage IV (III+IV) OSCC patients (Figure 1A). No significant difference between NC subjects and III+IV patients was seen in terms of the granzyme B expression by CD8 T cells (Figure 1A). Following stimulation with S. mitis, I+II patients presented significantly higher frequencies of granzyme B-expressing CD8 T cells than NC subjects (Figure 1B). No significant differences between NC subjects and III+IV patients, or between I+II patients and III+IV patients, were found (Figure 1B). Following S. anginosus stimulation, both I+II patients and III+IV patients presented significantly higher frequencies of granzyme B-expressing CD8 T cells than NC subjects, while no significant difference between I+II and III+IV patients was observed (Figure 1C). CD4 T cells from NC subjects, I+II patients and III+IV patients did not present significant levels of granzyme B expression following Streptococcus stimulation (data not shown). To examine whether OSCC patients simply presented higher total CD8 T cell activation than NC subjects regardless of antigen, or whether the higher frequencies of granzyme B-expressing CD8 T cells were restricted to Streptococcus antigens, we used CD3/CD28 activator beads to stimulate PBMCs from NC subjects and OSCC patients. In contrast to the observations following Streptococcus stimulation, the NC subjects after CD3/CD28 activation presented significantly higher frequencies of granzyme B-expressing CD8 T cells than I+II and III+IV patients (Figure 2). I+II patients also presented higher frequencies of granzyme B-expressing CD8 T cells than III+IV patients (Figure 2).
Streptococcus-reactive CD8 T cells presented reduced PD-1 and TIM-3 expression than
total CD8 T cells PD-1 and TIM-3 are inhibitory surface molecules that contribute to cell apoptosis and anergy[21,22]. CD8 T cells that express high levels of PD-1 and/or TIM-3 are present at elevated levels in patients with chronic virus infections or cancer, and tend to display an exhausted functional profile. Using flow cytometry, we investigated PD-1 and TIM-3 expression in Streptococcus-reactive CD8 T cells. Granzyme B-expressing (positive) and granzyme B-non-expressing (negative) CD8 T cells were identified by intracellular staining (Figure 3A), and the levels of PD-1 and TIM-3 expression on the surface was then examined on granzyme B-positive vs. granzyme B-negative CD8 T cells (Figure 3B). Only OSCC individuals with greater than 1% of CD8 T cells expressing granzyme B (identified using flow cytometry as shown in Figure 2A) were analyzed because a sizable cell population was required for flow cytometry analysis. We observed that after Streptococcus stimulation, all the granzyme B-positive CD8 T cells from OSCC patients with greater than 1% granzyme B-expressing CD8 T cells presented significantly lower PD-1 than the granzyme B-negative CD8 T cells from the same individuals (Figure 3C). Similarly, the granzyme B-positive CD8 T cells also presented lower TIM-3 expression than the granzyme B-negative CD8 T cells from the same individuals (Figure 3D). For comparison, we also examined the PD-1 and TIM-3 expression levels in granzyme B-expressing and granzyme B-non-expressing CD8 T cells after CD3/CD28 activation using flow cytometry (Figure 4A and 4B). After CD3/CD28 activation, the granzyme B-positive CD8 T cells from OSCC patients presented significantly lower surface PD-1 expression than the granzyme B-negative CD8 T cells from the same individuals (Figure 4C), but unlike the case with Streptococcus stimulation in which granzyme B-positive CD8 T cells presented uniformly lower PD-1 expression, the granzyme B-positive CD8 T cells after CD3/CD28 activation presented higher PD-1 than granzyme B-negative CD8 T cells in a few OSCC subjects. In terms of surface TIM-3 expression, no statistically significant difference was observed between granzyme
B-positive CD8 T cells and granzyme B-negative CD8 T cells (Figure 4D).
The level of Streptococcus-reactive cytotoxic T cells was positively associated with prognosis To examine the role of Streptococcus-reactive cytotoxic T cell responses in OSCC patients, we examined the association between the frequencies of Streptococcus-reactive granzyme B-expressing CD8 T cells and the prognosis of OSCC patients. At 24 months following tumor resection, 11 out of 35 patients (31.4%) presented recurrent tumor. We found that OSCC patients without recurrence presented significantly higher S. salivariusand S. mitis-reactive granzyme B-expressing CD8 T cells than OSCC patients without recurrence (Figure 5A). In patients with recurrence after tumor resection, we examined the frequencies of Streptococcus-reactive granzyme B-expressing CD8 T cells with the recurrence-free time. The length of recurrence-free time was positively associated with the frequencies of S. salivarius- and S. anginosus-reactive granzyme B-expressing CD8 T cells (Figure 5B).
Discussion A role of commensal bacteria in modulating the human immune responses in various inflammatory diseases is being increasingly recognized. At resting state, the mucosal surfaces, where commensal and pathogenic microorganisms may colonize, are protected by intact epithelial barriers, but during an inflammatory response, the epithelial barriers can be disrupted by inflammation and/or drugs, allowing increased contact between microorganisms and the host immune system. In some cases, pathogenic inflammation caused by bacterial colonization and infiltration may contribute to the development of certain
cancers,
such
as
Helicobacter
pylori-associated
gastric
cancer
and
colitis-associated cancer[23]. However, recent investigations found that bacteria-induced inflammation might also participate in antitumor immunity[8–10,13]. In this study, we
provided evidence that bacteria-reactive responses might be a constituent of the antitumor immunity in OSCC patients. Upregulation of Streptococcus strains in OSCC patients were identified across several studies[18,20,24]. To examine their association with the host T cell responses, we used several common Streptococcus species, including S. salivarius, S. mitis, and S. anginosus, to stimulate the PBMCs from NC subjects and OSCC patients. Compared to NC subjects, OSCC patients at earlier stages had higher frequencies of granzyme B-expressing CD8 T cells for all Streptococcus species tested, while OSCC patients at more advanced stages had higher frequencies of granzyme B-expressing CD8 T cells for S. anginosus but not other Streptococcus species. These observations suggested that the Streptococcus-reactive response in OSCC presented an initial upregulation, which was followed by a downregulation. These results might be explained by several non-mutually exclusive mechanisms. First, the elevation of Streptococcus-reactive cytotoxic CD8 T cells might be a corollary of increased Streptococcus colonization. It has been shown that OSCC patients presented increased levels of Streptococci in the salivary microbiota than healthy individuals, and Streptococcus has been identified in resected tumor tissues[18,24]. But within the OSCC group, it is yet unclear whether OSCC individuals at more advanced stages presented lower Streptococcus colonization than OSCC individuals at earlier stages. Second, the downregulation of Streptococcus-reactive cytotoxic CD8 T cells might be a corollary of an overall impairment in CD8 T cell functionality in advanced cancer patients. Supporting this idea, we found that OSCC patients at later stages presented lower granzyme B response than OSCC patients at earlier stages and NC subjects. We subsequently examined the expression of PD-1 and TIM-3, two inhibitory checkpoint markers previously associated with CD8 T cell exhaustion. We found that compared to CD8 T cells that did not respond to Streptococcus stimulation, the CD8 T cells that responded to Streptococcus stimulation by granzyme B secretion presented
lower PD-1 and lower TIM-3 uniformly. This observation first appeared obvious, since less-exhausted CD8 T cells are more likely to present potent cytotoxicity than more-exhausted CD8 T cells. However, we subsequently examined the PD-1 and TIM-3 expression in granzyme B-expressing and granzyme B-non-expressing CD8 T cells following CD3/CD28 stimulation. The CD8 T cells that expressed granzyme B did not present uniformly lower PD-1 than the CD8 T cells that did not express granzyme B. Furthermore, no differences in TIM-3 expression was found between the two groups. Therefore, more research is required to investigate whether bacteria such as Streptococcus could promote T cell responses via mechanisms that result in the downregulation of inhibitory checkpoint molecules. It should also be examined whether blocking PD-1 and TIM-3 could improve the Streptococcus-reactive cytotoxic responses of the CD8 T cells. We also found that the frequencies of Streptococcus-reactive CD8 T cell-mediated granzyme B responses were positively correlated with prognosis. Whether a causal relationship exists between Streptococcus-reactive granzyme B-expressing CD8 T cell responses and clinical outcomes of OSCC patients still require further analysis. If Streptococcus-reactive CD8 T cell can promote antitumor immunity, one can expect that such properties are transferable with the CD8 T cells between different individuals, and Streptococcus strains can be used in the development of future immunotherapies. Extensive animal investigations are required to investigate these hypotheses.
Acknowledgments This work was supported by National Natural Science Foundation of China (No.: 81271154).
Conflict of interest None.
Figure 1. Frequencies of granzyme B-expressing CD8 T cells following Streptococci stimulation. PBMCs were incubated with (A) S. salivarius, (B) S. mitis, or (C) S. anginosus, for 2 days. CD8 T cells was then isolated from Streptococcus-stimulated PBMCs and plated in ELISPOT plates at 1,000 to 10,000 cells per well for 6 h. The percentage of granzyme B-expressing cells was calculated by the number of positive spots / the number of loaded cells * 100%. Kruskal Wallis ANOVA followed by Dunn’s post-test. * p < 0.05. ** p < 0.01.
Figure 2. Frequencies of total granzyme B-expressing CD8 T cells following CD3/CD28 stimulation. PBMCs were incubated with CD3/CD28 activator beads for 2 days. CD8 T cells was then isolated from stimulated PBMCs and plated in ELISPOT plates at 100 to 1,000 cells per well for 6 h. The percentage of granzyme B-expressing cells was calculated by the number of positive spots / the number of loaded cells * 100%. Kruskal Wallis ANOVA followed by Dunn’s post-test. * p < 0.05. *** p < 0.001.
Figure
3.
Expression
of
PD-1
and
TIM-3
in
Streptococcus-reactive
and
Streptococcus-nonreactive cytotoxic CD8 T cells. PBMCs were incubated with S. salivarius, S. mitis, or S. anginosus for 2 days and with brefeldin A and monensin for the final 5 h. Cells were then labeled with fluorescent antibodies for flow cytometry analysis. (A) The gating of granzyme B (GB)-positive and GB-negative cells in CD8 T cells in one representative OSCC individual following S. salivarius stimulation. Dots shown were pregated on live single CD8 T cells. (B) The surface expression of PD-1 (left) and TIM-3 (right) in GB-positive (black) and GB-negative (grey) cells. (C) The mean fluorescence intensity (MFI) of PD-1 on GB-positive and GB-negative CD8 T cells from OSCC individuals following S. salivarius, S. mitis, and S. anginosus stimulation. (D) The MFI of TIM-3 on GB-positive and GB-negative CD8 T cells from OSCC individuals
following S. salivarius, S. mitis, and S. anginosus stimulation. (C) and (D) Wilcoxon matched pairs test. * p < 0.05. ** p < 0.01.
Figure 4. Expression of PD-1 and TIM-3 in granzyme B-positive and granzyme B-negative CD8 T cells following CD3/CD28 stimulation. PBMCs were incubated with CD3/CD28 activator beads for 2 days and with brefeldin A and monensin for the final 5 h. Cells were then labeled with fluorescent antibodies for flow cytometry analysis. (A) The gating of GB-positive and GB-negative cells in CD8 T cells in one representative OSCC individual following CD3/CD28 activation. Dots shown were pregated on live single CD8 T cells. (B) The surface expression of PD-1 (left) and TIM-3 (right) in GB-positive (black) and GB-negative (grey) cells. (C) The mean fluorescence intensity (MFI) of PD-1 on GB-positive and GB-negative CD8 T cells from OSCC individuals following CD3/CD28 activation. (D) The MFI of TIM-3 on GB-positive and GB-negative CD8 T cells from OSCC individuals following CD3/CD28 activation. (C) and (D) Wilcoxon matched pairs test. * p < 0.05.
Figure
5.
Association
between
OSCC
prognosis
and
the
frequencies
of
Streptococcus-reactive CD8 T cells. The recurrence status was followed in OSCC patients for 24 months after tumor resection. (A) The frequencies of granzyme B-expressing CD8 T cells after S. salivarius, S. mitis, and S. anginosus stimulation in OSCC patients with recurrence vs. in OSCC patients without recurrence. Wilcoxon matched pairs test. (B) The correlation between the frequencies of granzyme B-expressing CD8 T cells and the recurrence-free time in patients with recurrence. Pearson correlation test. * p < 0.05. ** p < 0.01.
References [1]
M.D. Mignogna, S. Fedele, L. Lo Russo, The World Cancer Report and the burden
of oral cancer., Eur. J. Cancer Prev. 13 (2004) 139–142. doi:10.1097/00008469-200404000-00008. [2]
S. SILVERMAN, Demographics and occurrence of oral and pharyngeal cancers, J. Am. Dent. Assoc. 132 (2001) 7S–11S. doi:10.14219/jada.archive.2001.0382.
[3]
J.N. Myers, T. Elkins, D. Roberts, R.M. Byers, Squamous cell carcinoma of the tongue in young adults: increasing incidence and factors that predict treatment outcomes, Otolaryngol. Neck Surg. 122 (2000) 44–51.
[4]
J.D. Holmes, E.J. Dierks, L.D. Homer, B.E. Potter, Is detection of oral and oropharyngeal squamous cancer by a dental health care provider associated with a lower stage at diagnosis?, J. Oral Maxillofac. Surg. 61 (2003) 285–291. doi:10.1053/joms.2003.50056.
[5]
L. V. Hooper, D.R. Littman, A.J. Macpherson, Interactions Between the Microbiota and the Immune System, Science (80-. ). 336 (2012) 1268–1273. doi:10.1126/science.1223490.
[6]
M.C. Abt, L.C. Osborne, L.A. Monticelli, T.A. Doering, T. Alenghat, G.F. Sonnenberg, et al., Commensal Bacteria Calibrate the Activation Threshold of Innate Antiviral Immunity, Immunity. 37 (2012) 158–170. doi:10.1016/j.immuni.2012.04.011.
[7]
I.I. Ivanov, K. Atarashi, N. Manel, E.L. Brodie, T. Shima, U. Karaoz, et al., Induction of Intestinal Th17 Cells by Segmented Filamentous Bacteria, Cell. 139 (2009) 485–498. doi:10.1016/j.cell.2009.09.033.
[8]
A. Sivan, L. Corrales, N. Hubert, J.B. Williams, K. Aquino-Michaels, Z.M. Earley, et al., Commensal Bifidobacterium promotes antitumor immunity and facilitates anti-PD-L1 efficacy, Science (80-. ). 350 (2015) 1084–1089. doi:10.1126/science.aac4255.
[9]
S. Viaud, F. Saccheri, G. Mignot, T. Yamazaki, R. Daillère, D. Hannani, et al., The intestinal microbiota modulates the anticancer immune effects of
cyclophosphamide., Science. 342 (2013) 971–6. doi:10.1126/science.1240537. [10]
R. Daillère, M. Vétizou, N. Waldschmitt, T. Yamazaki, C. Isnard, V. Poirier-Colame, et al., Enterococcus hirae and Barnesiella intestinihominis Facilitate Cyclophosphamide-Induced Therapeutic Immunomodulatory Effects, Immunity. 45 (2016) 931–943. doi:10.1016/j.immuni.2016.09.009.
[11]
S. Roy, G. Trinchieri, Microbiota: a key orchestrator of cancer therapy, Nat. Rev. Cancer. 17 (2017) 271–285. doi:10.1038/nrc.2017.13.
[12]
Q.F. Gui, H.F. Lu, C.X. Zhang, Z.R. Xu, Y.M. Yang, Well-balanced commensal microbiota contributes to anti-cancer response in a lung cancer mouse model, Genet. Mol. Res. 14 (2015) 5642–5651. doi:10.4238/2015.May.25.16.
[13] N. Iida, A. Dzutsev, C.A. Stewart, L. Smith, N. Bouladoux, R.A. Weingarten, et al., Commensal Bacteria Control Cancer Response to Therapy by Modulating the Tumor Microenvironment, Science (80-. ). 342 (2013) 967–970. doi:10.1126/science.1240527. [14]
L. Zitvogel, L. Galluzzi, S. Viaud, M. Vétizou, R. Daillère, M. Merad, et al., Cancer and the gut microbiota: an unexpected link., Sci. Transl. Med. 7 (2015) 271ps1. doi:10.1126/scitranslmed.3010473.
[15]
K. Shiga, Presence of Streptococcus infection in extra-oropharyngeal head and neck squamous cell carcinoma and its implication in carcinogenesis, Oncol. Rep. 8 (2001) 245–248.
[16]
K.N. Nagy, I. Sonkodi, I. Szöke, E. Nagy, H.N. Newman, The microflora associated with human oral carcinomas, Oral Oncol. 34 (1998) 304–308. doi:10.1016/S1368-8375(98)00008-6.
[17]
S.J. Hooper, S.J. Crean, M.J. Fardy, M.A. Lewis, D.A. Spratt, W.G. Wade, et al., A molecular analysis of the bacteria present within oral squamous cell carcinoma, J Med Microbiol. 56 (2007) 1651–1659. doi:10.1099/jmm.0.46918-0.
[18] D.L. Mager, A.D. Haffajee, P.M. Devlin, C.M. Norris, M.R. Posner, J.M. Goodson,
The salivary microbiota as a diagnostic indicator of oral cancer: a descriptive, non-randomized study of cancer-free and oral squamous cell carcinoma subjects., J. Transl. Med. 3 (2005) 27. doi:10.1186/1479-5876-3-27. [19]
B.W. Neville, T. a. Day, Oral Cancer and Precancerous Lesions, CA. Cancer J. Clin. 52 (2002) 195–215. doi:10.3322/canjclin.52.4.195.
[20]
N. Khajuria, R. Metgud, Role of bacteria in oral carcinogenesis., Indian J. Dent. 6 (2015) 37–43. doi:10.4103/0975-962X.151709.
[21]
M.E. Keir, M.J. Butte, G.J. Freeman, A.H. Sharpe, PD-1 and its ligands in tolerance and immunity., Annu. Rev. Immunol. 26 (2008) 677–704. doi:10.1146/annurev.immunol.26.021607.090331.
[22]
A.C. Anderson, Tim-3: an emerging target in the cancer immunotherapy landscape., Cancer Immunol. Res. 2 (2014) 393–8. doi:10.1158/2326-6066.CIR-14-0039.
[23]
A. Dzutsev, R.S. Goldszmid, S. Viaud, L. Zitvogel, G. Trinchieri, The role of the microbiota in inflammation, carcinogenesis, and cancer therapy, Eur. J. Immunol. 45 (2015) 17–31. doi:10.1002/eji.201444972.
[24]
M. Sasaki, C. Yamaura, Y. Ohara-Nemoto, S. Tajika, Y. Kodama, T. Ohya, et al., Streptococcus anginosus infection in oral cancer and its infection route., Oral Dis. 11 (2005) 151–156. doi:10.1111/j.1601-0825.2005.01051.x.
Table 1. Demographic and clinical information of the study participants. Healthy
Patient (Stages I Patient + II)
III + IV)
(Stages p-value
Number of patients
16
19
16
Female (N, %)
6 (37%)
7 (37%)
5 (31%)
> 0.05
Age in years
61.0
61.4
63.8
> 0.05
(median, range)
(55.3 - 65.2)
(54.1 - 66.3)
(57.8 - 67.7) >0.05
Location Tongue
11
8
Oral floor
4
5
Lower gingiva
4
3 < 0.001
Tumor size T1
5
0
T2
12
1
T3
0
9
T4
0
6 < 0.01
Lymph node metastasis N0
19
8
N1
0
4
N2
0
4 > 0.05
Distant metastasis Negative
19
16
Positive
0
0