Regulatory B1a Cells Suppress Melanoma Tumor Immunity via IL-10 Production and Inhibiting T Helper Type 1 Cytokine Production in Tumor-Infiltrating CD8+ T Cells

See related commentary on pg 1422

ORIGINAL ARTICLE

Regulatory B1a Cells Suppress Melanoma Tumor Immunity via IL-10 Production and Inhibiting T Helper Type 1 Cytokine Production in Tumor-Infiltrating CD8D T Cells Tadahiro Kobayashi1, Kyosuke Oishi1, Ai Okamura1,2, Shintaro Maeda1, Akito Komuro1,2, Yasuhito Hamaguchi1, Manabu Fujimoto3, Kazuhiko Takehara1 and Takashi Matsushita1 In tumor immunity, the participation of IL-10eproducing regulatory B cells (Bregs), which play an important role in suppressing immune responses, is unclear. In this study, we demonstrated an increase in B16F10 melanoma growth and a decrease in the proportion of IFN-ge and TNF-aesecreting tumor-infiltrating CD8þ T cells in B cellespecific PTEN-deficient mice in which Bregs were expanded. The number of tumor-infiltrating Bregs significantly increased in B cellespecific PTEN-deficient mice. More than 50% of tumor-infiltrating B cells consisted of Bregs, predominantly CD19þCD5þCD43þ B1a Bregs, in both B cellespecific PTEN-deficient and control mice. Adoptive B1a B cell transfer, which includes >30% of Bregs, increased melanoma growth, whereas non-B1a B cell transfer, which includes <2% of Bregs, exhibited no effect. In addition, adoptive transfer of B1a B cells from wild-type mice, but not IL-10e/e mice, exacerbated B16F10 melanoma growth. The current study indicates that B1a Bregs negatively regulate anti-melanoma immunity by producing IL-10 and reducing T helper 1 type cytokine production in tumor-infiltrating CD8þ T cells. Therefore, B1a Bregs can be a potentially novel target for immunotherapy of melanomas. Journal of Investigative Dermatology (2019) 139, 1535e1544; doi:10.1016/j.jid.2019.02.016

INTRODUCTION Immune checkpoint inhibitors, which enhance immune responses against malignant tumors by activating cytotoxic T cells, have dramatically improved the overall survival of patients with melanoma (Hodi et al., 2010; Robert et al., 2011). Although T cells play a major role in the immune response, recent assessments of the role of B cells in the immune system indicated that B cells have more essential functions than had been suggested previously (Fillatreau et al., 2002; Mauri et al., 2003). In addition to providing humoral immunity, B cells have other multiple functions and exert both positive and negative effects in immune reactions. B cells can positively modulate the immune response through their roles in antigen presentation, activation of T cells, and cytokine production (Lipsky, 2001; Matsushita et al., 2018; Nelson, 2010; Shen and Fillatreau, 2015). It is also widely known that IL-10eproducing regulatory B cells (Bregs) are 1

Department of Dermatology, Faculty of Medicine, Institute of Medical, Pharmaceutical, and Health Sciences, Kanazawa University, Kanazawa, Japan; 2Department of Plastic Surgery, Kanazawa University Hospital, Kanazawa, Japan; and 3Department of Dermatology, Faculty of Medicine, University of Tsukuba, Tennodai, Tsukuba, Japan Correspondence: Takashi Matsushita, Department of Dermatology, Faculty of Medicine, Institute of Medical, Pharmaceutical, and Health Sciences, Kanazawa University, 13-1 Takaramachi, Kanazawa, Ishikawa 920-8641, Japan. E-mail: [email protected] Abbreviations: Breg, regulatory B cell; Cd19Creþ/ePtenloxp/loxp, B cellespecific PTEN-deficient mice; TIL, tumor-infiltrating lymphocyte; TIL-B, tumor-infiltrating B cell; WT, wild-type Received 1 November 2018; revised 13 February 2019; accepted 17 February 2019; accepted manuscript published online 2 March 2019; corrected proof published online 14 May 2019

negative regulators and are involved in immune pathologies, such as autoimmunity and inflammation (Bouaziz et al., 2008; Iwata et al., 2011; Lampropoulou et al., 2010; Matsushita et al., 2008). More than a decade ago, Bregs that inhibit inflammatory immune responses were described in a mouse model of inflammatory bowel disease (Mizoguchi et al., 2000). Bregs were identified based on their production of the immunosuppressive cytokine IL-10 ex vivo, following 5 hours of stimulation with phorbol 12-myristate 13-acetate and ionomycin, which stimulate PKC and calcium transport, respectively (Yanaba et al., 2008). In Bregs, PI3K, an important downstream effector of B cell antigen receptor signaling, promotes IL-10 production through Akt activation, and IL-10 production is hyperactivated by knocking down PTEN, as PTEN is an inhibitor of Akt activity (Matsushita et al., 2016). We previously confirmed Bregs were expanded by inactivated PTEN specifically in B cells using a Cre-loxP system (Matsushita et al., 2016). Conditional deletion of PTEN in B cells was achieved by crossing mice expressing loxP-flanked Pten alleles with Cd19Cre mice. B cellespecific PTENdeficient (Cd19Creþ/ePtenloxp/loxp) mice showed expansion of not only B1a and marginal zone B cell populations (Suzuki et al., 2003), but also Bregs in the blood, peripheral lymph nodes, and spleen compared to control (Cd19Creþ/e) mice (Matsushita et al., 2016). Adoptive transfer of PTEN-deficient B cells, which showed a higher proportion of Bregs, significantly improved contact hypersensitivity response (Matsushita et al., 2016). Thus, B cellespecific PTENdeficient mice are a good model for investigating Breg function in tumor immunity.

ª 2019 The Authors. Published by Elsevier, Inc. on behalf of the Society for Investigative Dermatology.

www.jidonline.org

1535

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity

a

RESULTS

Tumor volume (cm³)

40

***

Cd19Cre +/-

30

Cd19Cre

+/-

Pten loxP/loxP

*

20 10 0 0

4

7

11

15

18

21

Time after Tumor Inoculation (Days)

b

% Survival

To determine the effect of Bregs on anti-melanoma immunity, Cd19Creþ/ePtenloxp/loxp and control mice (Cd19Creþ/e mice) were injected subcutaneously with 1.0  106 B16F10 melanoma cells in the flank, and tumor growth was monitored. B16F10 melanoma growth in Cd19Creþ/ePtenloxp/loxp mice was significantly enhanced compared to that in Cd19Creþ/e mice (Figure 1a). Significant differences in tumor volume between Cd19Creþ/ePtenloxp/loxp and Cd19Creþ/e mice were also observed beginning on day 18 after inoculation, and the enhanced tumor growth resulted in decreased survival of Cd19Creþ/ePtenloxp/loxp mice (Figure 1b). Thus, tumor immunity against B16F10 melanoma was attenuated in Cd19Creþ/ePtenloxp/loxp mice. Tumor-infiltrating B cells were significantly increased in B cellespecific PTEN-deficient mice

100 80 60 40 Cd19Cre +/-

20

Cd19Cre

+/-

Pten loxP/loxP

0 0

10

20

30

*

40

Time after Tumor Inoculation (Days) Figure 1. B16F10 melanoma growth was increased in B cellespecific PTEN-deficient mice. (a) Mean tumor volumes  standard error of the mean were determined on the indicated days following B16F10 melanoma inoculation (n ¼ 7 for each group). (b) Kaplan-Meier survival curves of mice following B16F10 melanoma inoculation (n ¼ 7 for each group). Statistical significance was determined by two-way analysis of variance followed by Bonferroni post-test (a) and Mantel-Cox log-rank test (b). Significant differences are indicated: *P < 0.05, ***P < 0.001. Similar results were obtained from two independent experiments. Cd19Creþ/ePtenloxp/loxp, B cellespecific PTEN-deficient mice.

While considerable work has been done to elucidate the role of Bregs in tumor immunity, fewer efforts have been made to examine their potential function in melanoma (Lykken et al., 2015). We hypothesized that Bregs inhibit not only the autoimmunity and inflammatory immune response, but also anti-melanoma immunity. In the current study, we found that melanoma growth was increased in B cellespecific PTEN-deficient mice in which the number of tumor-infiltrating Bregs was significantly higher than that in control mice. Most tumorinfiltrating Bregs consist of B1a Bregs, and adoptive B1a B cell transfer exacerbates melanoma tumor immunity. Furthermore, we confirmed that B1a Bregs inhibit anti-melanoma immunity via producing IL-10 and that the proportion of IFN-ge and TNF-aesecreting tumorinfiltrating CD8þ T cells decrease in B cellespecific PTEN-deficient mice. 1536

B16F10 melanoma growth was significantly increased in B cellespecific PTEN-deficient mice

Journal of Investigative Dermatology (2019), Volume 139

We hypothesized that alterations in B cell populations in Cd19Creþ/ePtenloxp/loxp mice affect the attenuation of tumor immunity against B16F10 melanoma. Thus, we assessed tumor-infiltrating lymphocyte (TIL) populations in Cd19Creþ/ePtenloxp/loxp and Cd19Creþ/emice on day 11 after inoculation. Immunohistochemical analysis revealed that the number of tumor-infiltrating B220þ cells was clearly increased in Cd19Creþ/ePtenloxp/loxp mice (Figure 2a, upper panels). In contrast, the number of tumor-infiltrating CD3þ cells in Cd19Creþ/ePtenloxp/loxp mice was similar to that in Cd19Creþ/e mice (Figure 2a, lower panels). Additionally, TILs were assessed for CD3, CD4, CD8, CD19, B220, CD25, and NK1.1 expression by flow cytometry. The phenotypes of the TILs included CD19þB220þ B cells, CD4þ T cells, CD8þ T cells, CD4þ CD25þ regulatory T cells, and NK1.1þCD3e natural killer cells. Consistent with B cell expansion (Suzuki et al., 2003), the percentage and number of CD19þB220þ B cells were significantly increased in Cd19Creþ/ePtenloxp/loxp mice compared to that in Cd19Creþ/e mice (Figure 2b), while no significant difference was observed in the numbers of CD4þ T cells, CD8þ T cells, CD4þCD25þ regulatory T cells, or NK1.1þCD3e natural killer cells (Figure 2ce2e, right bar graphs). These results suggest that alterations in tumor-infiltrating B cell populations contribute to the attenuation of tumor immunity against B16F10 melanoma in Cd19Creþ/ePtenloxp/loxp mice. Most tumor-infiltrating Bregs consisted of B1a Bregs, but not marginal zone Bregs

We previously reported that IL-10eproducing B cells (Bregs) were predominantly found within the splenic marginal zone and B1 B cell subsets (Matsushita et al., 2016). Thus, we assessed the population of tumor-infiltrating B cells and Bregs in Cd19Creþ/ePtenloxp/loxp and Cd19Creþ/e mice on day 11 after B16F10 melanoma inoculation. TILs were stimulated in vitro with lipopolysaccharide, phorbol 12-myristate 13acetate, and ionomycin for 5 hours in the presence of brefeldin A, followed by staining for the cell surface markers CD19, B220, CD1d, CD5, CD21, CD23, and CD43, as well as intracellular IL-10 expression and analyzed by flow cytometry. Tumor-infiltrating B cells (TIL-Bs) were sorted into three fractions—CD1dintCD5e B cells, CD1dhiCD5e B cells,

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity Cd19Cre+/- Pten

Cd19Cre +/-

loxP/loxP

CD3

B220

a

Cd19Cre +/Pten loxP/loxP

Cd19Cre+/10

4

10

103

102

102

101

101

102

103

104

**

% B cells

CD19

103

101

20

10.1%

2.07%

100 100

Percentages

Cell numbers

4

10

100 100

101

102

103

B cells (x104)

b

0

104

** 20

Cd19Cre +/Cd19Cre +/Pten loxP/loxP

Figure 2. Tumor-infiltrating B cell numbers were increased in B cell especific PTEN-deficient mice. (a) Representative immunohistochemistry images of B220þ and CD3þ cells (stained red to differentiate from the brown melanin pigment). All samples were counterstained with hematoxylin (blue/purple). Original images: 100 magnification. Scale bars ¼ 50 mm. The percentages and numbers of (b) tumor-infiltrating CD19þB220þ B cells, (c) CD4þ T cells and CD8þ T cells, (d) CD4þCD25þ regulatory T cells, and (e) NK1.1þCD3e natural killer cells in Cd19Creþ/ePtenloxp/loxp and Cd19Creþ/e mice were determined by immunofluorescent staining followed by flow cytometric analysis. Error bars indicate the mean  standard error of the mean (n ¼ 6 for each group). Statistical significance was determined by unpaired twosided t test. Significant differences are indicated: *P < 0.05, **P < 0.01. Similar results were obtained from two independent experiments. Cd19Creþ/ePtenloxp/loxp, B cell especific PTEN-deficient mice.

10 0

104

104

5.17%

12.4% 103

103

102

102

% CD4+ cells

c

*

CD4

10

101

15.3%

100 100

101

102

103

104

101

100 100

9.57% 101

102

103

5 0

104

CD4+ cells (x104)

B220 20 10

Cd19Cre +/Cd19Cre +/Pten loxP/loxP

0

d

104

104

1.86%

103

102

CD4

102

101

10

1.59%

103

101

0

100

101

102

103

104

100 100

101

102

103

104

CD25 104

3

3

10

10

102

102

16.7%

14.8%

101

101

100 100

101

102

103

104

100 100

4

2

0

20 % NK cells

104

CD3

e

% CD4+ CD25+ cells

0

101

102

103

104

10

0

CD4+ CD25+ cells (x104)

15

NK cells (x104)

% CD8+ cells

30

CD8+ cells (x104)

CD8 40

Cd19Cre +/-

20

Cd19Cre +/Pten loxP/loxP

0 8

4

Cd19Cre +/Cd19Cre +/Pten loxP/loxP

0

20

Cd19Cre +/-

10

Cd19Cre +/Pten loxP/loxP

0

NK1.1

www.jidonline.org

1537

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity Cd19Cre +/Pten loxP/loxP

Cd19Cre +/105

104

104

103

103

102 0

5.81%

60 CD1dint CD5- Fo B cell

40

102

0

0

92.6% 103

104

105

CD1dhi CD5- MZ B cell

20

102

6.94%

75.1%

18.5% 0

102

103

104

CD1dint CD5+ B1a B cell

0 105

Cd19Cre +/Pten loxP/loxP

Cd19Cre +/65.6%

IL-10

105

isotype

59.1%

105

104

104

103

103

103

102

102

102

0

0

102

103

104

105

0.22%

105

104

0

0

0

102

103

104

105

0

102

103

104

105

105

CD1d

0.02%

2.62%

104

104

103

103

102

102

0

0

98.2%

1.62% 0

102

103

104

105

90.3%

6.59% 0

102

103

104

105

CD5

% Tumor-infiltrating IL-10+ B cells

CD19

105

Cd19Cre +/Pten loxP/loxP

Cd19Cre +/-

CD5

b

*** ***

80 % TIL-Bs

0.17%

*** ***

100

*** ***

100

Percentages 80

40

0

Tumor-infiltrating IL-10+ B cells (x104)

CD1d

105

% Tumor-infiltrating IL-10+ B cells

a

Cell numbers 40

20

*

Cd19Cre +/Cd19Cre +/Pten loxP/loxP

0

*** ***

80 60 CD1dint CD5- Fo IL-10+ B cells

40

CD1dhi CD5- MZ IL-10+ B cells 20

CD1dint CD5+ B1a IL-10+ B cells

0 Cd19Cre +/-

Cd19Cre +/Pten loxP/loxP

Figure 3. Most tumor-infiltrating regulatory B cells consisted of B1a regulatory B cells. (a) The subsets (CD1dintCD5e follicular [Fo], CD1dhiCD5e marginal zone [MZ], and CD1dintCD5þ B1a [B1a] B cells) of tumor-infiltrating B cells. (b) The percentages, numbers, and subsets of tumor-infiltrating IL-10þ B cells. Lower panels show only the IL-10þ B cells. Error bars indicate the mean  standard error of the mean (n ¼ 6 for each group). Two-way analysis of variance followed by Bonferroni post-test was used to compare the subsets of tumor-infiltrating regulatory B cells (a, lower right bar graphs in b). Unpaired two-sided t test was used to compare the percentages and numbers of tumor-infiltrating regulatory B cells (upper right bar graphs in b). Significant differences are indicated: *P < 0.05, ***P < 0.001. Similar results were obtained from two independent experiments. Cd19Creþ/ePtenloxp/loxp, B cellespecific PTEN-deficient mice.

and CD1dintCD5þ B cells. We confirmed that these populations respectively correspond to CD21intCD23hi follicular B cells, CD21hiCD23lo marginal zone B cells, and CD19þCD5þCD43þ B1a B cells in both Cd19Creþ/ePtenloxp/loxp and Cd19Creþ/e mice (Miles et al., 2017) (see Supplementary Figure S1 online). Although we previously confirmed that B cell populations in the blood, peripheral lymph node, and spleen consist mainly of follicular B cells in Cd19Creþ/ePtenloxp/loxp and Cd19Creþ/e mice (Matsushita et al., 2016), TIL-Bs mainly consist of B1a B cells in both mice (Figure 3a). As expected, the number of tumorinfiltrating IL-10þ B cells (Bregs) was significantly higher in Cd19Creþ/ePtenloxp/loxp mice than in Cd19Creþ/e mice (Figure 3b, upper right bar graphs). Additionally, most tumorinfiltrating Bregs consisted of B1a Bregs but not marginal zone Bregs in not only Cd19Creþ/ePtenloxp/loxp but also Cd19Creþ/e mice (Figure 3b, lower panels show only the IL10þ B cells). These results suggest that tumor-infiltrating B1a Bregs negatively regulate tumor immunity against B16F10 melanoma.

1538

Journal of Investigative Dermatology (2019), Volume 139

Adoptive transfer of B1a B cells exacerbated B16F10 melanoma growth and shortened animal survival

To investigate whether B1a Bregs inhibit tumor immunity against B16F10 melanoma, we evaluated the effect of adoptive transfer of B1a B cells into wild-type (WT) mice. We purified B1a B cells and non-B1a B cells from the splenocytes of Cd19Creþ/ePtenloxp/loxp mice and confirmed that purified B1a B cells accounted for >30% of IL-10þ B cells (Bregs), while purified non-B1a B cells accounted for <2% of Bregs (Figure 4a). We mixed purified B1a B cells or non-B1a B cells with B16F10 melanoma cells and injected the mixtures into WT mice. Adoptive transfer of B1a B cells, but not non-B1a B cells, significantly increased tumor growth and shortened animal survival (Figure 4b, 4c). These results indicate that B1a Bregs attenuate tumor immunity against B16F10 melanoma. Adoptive transfer of B1a B cells from WT mice, but not IL-10e/e mice, exacerbated B16F10 melanoma growth

Recent reports have shown that B cellemediated immune suppression occurs not only through IL-10, but also IL-35 and

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity

a

non-B1a B cells 105

B1a B cells 105

1.4%

CD1d

35.6% 104

104

103

103

102

102

0

59.4% 0

102

0

2.5% 103

104

4.1%

105

0

102

94.2% 103

104

105

CD5

IL-10

10

105

1.2%

4

10

103

102

102

0

0

0

102

103

104

105

32.4%

4

103

% IL-10+ B cells

105

0

102

103

104

40

***

20 non-B1a B cells B1a B cells

0

105

CD19

c

b B16F10 cells B16F10 cells + B1a B cells B16F10 cells + non-B1a B cells

*** †††

80

30

* ††

20

% Survival

Tumor volume (cm³)

40

B16F10 cells B16F10 cells + B1a B cells B16F10 cells + non-B1a B cells

100

10

60 40 20 0

0 0

4

7

11

14

17

21

Time after Tumor Inoculation (Days)

0

20

40

* †

60

Time after Tumor Inoculation (Days)

Figure 4. B16F10 melanoma growth was exacerbated by adoptive transfer of B1a regulatory B cells. (a) Isolated splenic B1a B cells and non-B1a B cells from B cellespecific PTEN-deficient mice. B cellespecific PTEN-deficient mice were stimulated in vitro with lipopolysaccharide, phorbol 12-myristate 13-acetate, and ionomycin for 5 hours in the presence of brefeldin A, followed by staining for CD1d, CD5, CD19, and IL-10. Error bars indicate the mean  standard error of the mean (n ¼ 5 for each group). ***P < 0.001 (unpaired two-sided t test). (b) Data represent mean tumor volumes  standard error of the mean (n ¼ 6 for each group). *P < 0.05, ***P < 0.001 for B16F10 melanoma versus B16F10 melanoma and B1a B cell, and ††P < 0.01, †††P < 0.001 for B16F10 melanoma and B1a B cell versus B16F10 melanoma and non-B1a B cell (two-way analysis of variance followed by Bonferroni post-test). (c) Kaplan-Meier survival curves of mice following inoculation (n ¼ 6 for each group). *P < 0.05 for B16F10 melanoma versus B16F10 melanoma and B1a B cell, and †P < 0.05 for B16F10 melanoma and B1a B cell versus B16F10 melanoma and non-B1a B cell (Mantel-Cox log-rank test).

TGF-b production (Gorosito Serra´n et al., 2015; Vadasz et al., 2013) and PD-L1 expression (Khan et al., 2015; Zacca et al., 2018). Therefore, we evaluated the effect of adoptive transfer of B1a B cells from WT or IL-10e/e mice into WT mice to investigate whether IL-10 produced by B1a B cells is responsible for the tumor-promoting effect in this context. Splenic B1a B cells from WT or IL-10e/e mice were sorted (Figure 5a) and mixed with B16F10 melanoma cells. The mixtures were injected subcutaneously into WT mouse flank, and tumor growth was monitored. Adoptive transfer of B1a B cells from WT mice, but not IL-10e/e mice, significantly increased tumor growth (Figure 5b). These results indicated

that IL-10 production by B1a B cells is important for the attenuation of tumor immunity against B16F10 melanoma. T helper type 1 cytokine production in tumor-infiltrating CD8D T cells decreased in B cellespecific PTEN-deficient mice

We assessed the cytokine production of TILs in Cd19Creþ/e Ptenloxp/loxp and Cd19Creþ/e mice on day 11 after B16F10 melanoma inoculation. TILs were stained for the cell surface markers CD3, CD4, CD8, and NK1.1 as well as intracellular IFN-g, TNF-a, and granzyme B expression and analyzed by flow cytometry. The phenotypes of the TILs included www.jidonline.org

1539

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity

a

TNF-aesecreting CD4þ T cells and granzyme Besecreting NK1.1þCD3e natural killer cells were similar in both Cd19Creþ/ePtenloxp/loxp and Cd19Creþ/e mice (Figure 6b, 6c). Cytokine production by tumor-infiltrating CD8þ T cells was correlated with B16F10 melanoma growth and animal survival in Cd19Creþ/ePtenloxp/loxp and Cd19Creþ/e mice. This suggests that an important factor contributes to the attenuation of tumor immunity against B16F10 melanoma in Cd19Creþ/ePtenloxp/loxp mice and supports that tumorinfiltrating Bregs inhibit cytokine production by CD8þ T cells in the melanoma tumor microenvironment.

Wild type IL-10 -/(before sorting) 105

105

CD1d

4.35%

5.93%

104

104

103

103

102

102

0

90.4% 0

102

0

3.49% 103

104

105

87.7% 0

102

5.37% 103

104

105

CD5 (after sorting) 10

5

105

CD1d

0.35%

0.17%

104

104

103

103

102

102

0

8.51% 0

2

10

0

91.1% 10

3

10

4

10

5

15.8% 0

102

84.0% 103

104

105

CD5

b B16F10 cells B16F10 cells + B1a B cells (WT)

Tumor volume (cm³)

60

*** †††

B16F10 cells + B1a B cells (IL-10 -/-) 40

† 20

0 0

4 7 11 14 17 Time after Tumor Inoculation (Days)

21

Figure 5. IL-10 production by B1a B cells is responsible for the melanomapromoting effect. (a) Isolated splenic CD19þ B cells from WT and IL-10e/e mice were stained for CD1d and CD5 and sorted into B1a (CD1dintCD5þ) B cell population. (b) Sorted B1a B cells (1.0  106) were mixed with B16F10 melanoma cells (2.0  105) and injected subcutaneously into WT mice. Data represent mean tumor volumes  standard error of the mean (n ¼ 6 for each group). ***P < 0.001 for B16F10 melanoma versus B16F10 melanoma and B1a B cell from WT mice, and †P < 0.05, †††P < 0.001 for B16F10 melanoma and B1a B cell from WT mice versus B16F10 melanoma and B1a B cell from IL-10e/e mice (two-way analysis of variance followed by Bonferroni post-test). WT, wild-type.

CD3þCD4þ T cells, CD3þCD8þ T cells, and NK1.1þCD3e natural killer cells. The percentages of IFN-ge and TNFaesecreting CD8þ T cells were significantly lower in Cd19Creþ/ePtenloxp/loxp mice than in Cd19Creþ/e mice (Figure 6a), whereas the percentages of IFN-ge and 1540

Journal of Investigative Dermatology (2019), Volume 139

DISCUSSION Previous papers reported that B cells have both positive and negative roles for anti-melanoma immunity (Chiaruttini et al., 2017). B cell depletion using an anti-CD20 mAb enhanced melanoma growth in a mouse model, which correlated with impaired activation of effector-memory T cells and cytotoxic cytokine-secreting T cells (DiLillo et al., 2010). This result suggested that B cells enhance anti-melanoma immunity. Furthermore, we have reported that B cells promote antimelanoma responses in B-cell linkeredeficient mice by promoting T-cell infiltration into melanoma tumors and T helper type 1 cytokine production within the tumor microenvironment (Kobayashi et al., 2014). Conversely, Zhang et al. (2016a) demonstrated that CD5þ B cells, but not CD5e B cells, inhibit anti-melanoma immunity. In the current study, we addressed the negative regulatory role of B cells, particularly tumor-infiltrating Bregs, in melanoma immunity. The number of tumor-infiltrating Bregs was significantly higher in B cellespecific PTEN-deficient mice, in which antimelanoma immunity was attenuated compared with in control mice, and most tumor-infiltrating Bregs consisted of B1a Bregs in both mice (Figure 3b and see Supplementary Figure S1). Adoptive transfer of B1a B cells into WT mice increased melanoma growth, while adoptive transfer of nonB1a B cells had no effect on melanoma growth (Figure 4b). In addition, adoptive transfer of B1a B cells from WT mice but not IL-10e/e mice, exacerbated B16F10 melanoma growth (Figure 5b). These results suggest that tumor-infiltrating B1a Bregs negatively regulate anti-melanoma immunity via IL-10 production, while non-B1a B cells do not possess immunosuppressive ability. These conflicting results may be due to the activation status of B cells in different model systems. T cell responses appear to be inhibited by resting B cells, but facilitated by activated B cells (Nielsen et al., 2012; Rodrı´guez-Pinto, 2005; Watt et al., 2007). In addition, previous studies using mice with B cell depletion induced by anti-CD20 mAb and B-cell linkeredeficient mice indicate that B cells as a whole can contribute to tumor rejection. By contrast, some B cell subsets, such as tumor-infiltrating CD5þ B cells and IL-10eproducing B1a Bregs, clearly showed that B cells acquire tolerant or pro-tumorigenic characteristics with melanoma progression, suggesting that the positive or negative effects of B cells on anti-melanoma immunity derive from the different subsets and functions. Therefore, this study indicates that the negative regulatory functions of B1a Bregs in anti-melanoma immunity are dominant in this context; however, this does not eliminate the anti-melanoma functions of B cells. Further

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity

Isotype control

104

104

104

103

103

103

102

102

102

0

10

0

0 0 102

103

104

105

68.3%

5

0 102

10

103

104

105

40.6%

5

0 102

10

104

104

103

103

103

102

102

102

0 0 102

103

104

105

103

104

0 0

102

103

104

105

0 102

103

104

Percentages

* 80

Cd19Cre +/Cd19Cre +/Pten loxP/loxP

40 0

105

0.10%

5

104

0

0.10%

105

% IFN-γ+ cells

IFN-γ

60.7%

105

% TNF-α+ cells

85.3%

105

TNF-α

Cd19Cre +/Pten loxP/loxP

Cd19Cre +/-

80

% IFN-γ+ cells

a

40

105

40

**

Cd19Cre +/Cd19Cre +/Pten loxP/loxP

0

CD8

IFN-γ

105

30.5%

105

30.9%

Isotype control 105

104

104

104

103

103

103

102

102

102

0

0 0

102

103

104

0 102 10

103

104

105

59.5%

5

0 102 10

104

104

103

103

103

102

102

102

0

0 102

103

104

105

103

104

0 0 102

103

104

105

0 102

103

104

20

Cd19Cre +/Cd19Cre +/Pten loxP/loxP

0

105

0.20%

5

104

0

0.20%

0

105

63.7%

105

TNF-α

Cd19Cre +/Pten loxP/loxP

Cd19Cre+/-

% TNF-α+ cells

b

105

80 Cd19Cre +/-

40

Cd19Cre +/Pten loxP/loxP

0

Granzyme B

Cd19Cre +/Pten loxP/loxP

Cd19Cre+/-

Isotype control

105

105

105

104

104

104

103

16.4%

103

102

102

0

0 0 102

103

104

105

18.4%

103

0.10%

102 0 0

102

103

104

105

0

102

103

104

105

% Granzyme B+ cells

CD4

c

80 Cd19Cre +/-

40

Cd19Cre +/Pten loxP/loxP

0

NK1.1 Figure 6. T helper 1 type cytokine production of tumor-infiltrating CD8D T cells decreased in Cd19CreD/ePtenloxp/loxp mice. The subcutaneous B16F10 melanoma tumors were harvested on day 11 after inoculation. (a, b) Tumor-infiltrating lymphocytes were stimulated in vitro with phorbol 12-myristate 13acetate and ionomycin for 5 hours in the presence of brefeldin A, followed by staining for the cell surface markers CD3, CD4, and CD8, as well as intracellular IFN-g and TNF-a. (c) Tumor-infiltrating lymphocyte were stained in vitro for the cell surface markers CD3, NK1.1, and intracellular granzyme B. Error bars indicate the mean  standard error of the mean (n ¼ 5 for each group). Statistical significance was determined by unpaired two-sided t test. Significant differences are indicated: *P < 0.05, **P < 0.01. Similar results were obtained from two independent experiments. Cd19Creþ/ePtenloxp/loxp, B cellespecific PTEN-deficient mice.

studies are required to identify the different B cell populations that contribute to melanoma rejection. Conventional B2 cells recirculate between the lymphoid tissues and blood, while B1 B cells recirculate between the peritoneal cavity and blood (Ansel et al., 2002). Leukocyte migration into tissues is mediated by a multistep adhesion cascade requiring chemoattractant and adhesion receptors on the leukocyte surface. However, B1 B cells can migrate into inflamed skin despite their lack of responsiveness to ligands for the skin-associated chemokine receptors CCR4 and CCR10, which target T cells in the skin and CCR6, which

attracts Langerhans cells in the epidermis. Additionally, CXCR3 and CXCR4 are not likely to be responsible for the differential ability of peritoneal B1 cells versus splenic B2 cells to migrate into skin, as these B cell subsets showed similar expression (Geherin et al., 2016). Geherin et al. (2016) demonstrated that B1 B cells that migrated from the peritoneal cavity and entered the blood nearly uniformly expressed activated a4b1 integrin, an adhesion molecule, allowing for their migration into the skin. Importantly, blocking a4-integrin completely abrogated B1 B-cell migration into the inflamed skin, while this blockade did not affect www.jidonline.org

1541

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity

the migration of B2 cells. Expression of activated a4b1 integrin is unusual and has been described for a limited number of effector cells, such as effector T cells, natural killer cells, and metastatic tumor cells (Kato et al., 2012; Rose et al., 2000; Shulman et al., 2011). These findings suggest that a4b1 integrin is selectively required for B1 B cell migration into the inflamed skin and may explain why most tumor-infiltrating B cells consisted of B1a B cells. B1 B cells consist of not only B1a (CD5þ) B cells but also B1b (CD5eCD11bþ) B cells (Hardy and Hayakawa, 2015; Kantor, 1991). Although B1b B cells could produce IL-10, B1a B cells are major source of IL-10 (Geherin et al., 2016). Furthermore, the TIL-Bs consist mainly of CD5þ B1a B cells (Figure 3). Although B1a and B1b B cells are B220lo (Hardy and Hayakawa, 2015), the TIL-Bs in the current study are B220hi. Considering that the TIL-B population contains CD19þCD5þCD43þ B1a cells (Miles et al., 2017) in not only Cd19Creþ/ePtenloxp/loxp, but also Cd19Creþ/e mice (Figure 3 and see Supplementary Figure S1), these results suggest the possibility that the phenotype of TIL-Bs is influenced by the tumor microenvironment rather than the lack of PTEN in B cells. In this study, T helper 1 type cytokine production by tumorinfiltrating CD8þ T cells, but not CD4þ T cells, decreased in B cellespecific PTEN-deficient mice compared to that in Cd19Creþ/e mice (Figure 6a). Inoue et al. (2006) demonstrated that IL-10 inhibits IFN-g production by CD8þ T cells in co-culture with irradiated murine leukemia cells (Inoue et al., 2006). For CD4þ T cells, Del Prete et al. (1993) reported that IL-10 inhibits cytokine (IFN-g, IL-4, and IL-5) synthesis by CD4þ T cells. Although naı¨ve CD4þ T cells are targeted by IL-10, activated and memory T cells appear to be rather insensitive to IL-10 because of downregulation of the IL-10 receptor upon CD4þ T-cell activation (Liu et al., 1994). These findings suggest that IL-10 produced by tumor-infiltrating B1a Bregs inhibits T helper type 1 cytokine production by tumor-infiltrating CD8þ T cells, but not CD4þ T cells. Regarding the antigen specificity of Bregs, it was reported that antigen-specific signals were essential for IL-10 production from B cells (Fillatreau et al., 2002). Zhang et al. (2016b) demonstrated that Bregs act in an antigen-specific manner to inhibit immune responses using TgVH3B4 mice, in which VH was derived from an actin-reactive natural antibody, 3B4 (Li et al., 2007; Zhang et al., 2016b). In TgVH3B4 mice, higher expansion of Bregs was confirmed in actin immunization than in ovalbumin immunization. It has also been demonstrated that IL-10 production by CD5þ splenic B cells from melanoma-bearing mice was significantly higher than that by CD5e B cells in the presence of irradiated melanoma cells in vitro (Zhang et al., 2016a). These findings suggest that CD5þ B1a Bregs produce IL-10 in a melanoma antigenspecific manner. Although melanoma antigens recognized by Bregs were not identified, investigations into the drivers of immune responses to melanoma elucidated several antigenic epitopes derived from human melanocyte lineage-specific proteins (MART-1/Melan-A, gp100, gp75, and tyrosinase) recognized by T cells (Bakker et al., 1994; Cole et al., 1994; Kawakami et al., 1994; Topalian et al., 1996). Fotaki et al. (2018) found that infection-enhanced adenovirus encoding 1542

Journal of Investigative Dermatology (2019), Volume 139

gp100 and allogenic dendritic cells stimulated the proliferation of endogenous gp100-TCRþ T cells, resulting in delayed melanoma growth and prolonged survival (Fotaki et al., 2018). Therefore, the identification of antigens recognized by Bregs may provide insight into the mechanisms by which Breg responses are triggered and inhibit anti-melanoma immunity. In conclusion, our data demonstrate that tumor-infiltrating B1a Bregs suppress melanoma tumor immunity by producing IL-10 and inhibiting IFN-g and TNF-a production by tumorinfiltrating CD8þ T cells. Therefore, a protocol that selectively inhibits tumor-infiltrating B1a Bregs or their IL-10 production may represent a novel immunotherapy for melanomas. Further studies of the selective inhibition of these are required to develop this immunotherapy. MATERIALS AND METHODS Mouse breeding

C57BL/6 mice, Cd19Creþ/e mice (Rickert et al., 1997), Ptenloxp/loxp mice (Groszer et al., 2001), and IL-10e/e mice (Kuhn et al., 1993) were obtained from the Jackson Laboratory (Bar Harbor, ME). All mice were bred on the C57BL/6 background. For experiments, all female mice used were 9e12 weeks of age and housed in a specific pathogenefree barrier facility. All animal studies were approved by the Committee on Animal Experimentation of the Kanazawa University Graduate School of Medical Sciences.

Cell lines B16F10 melanoma cells were purchased from the American Type Culture Collection (Manassas, VA). Cells were cultured at 37 C in 5% CO2 in DMEM supplemented with fetal bovine serum and penicillin/streptomycin. Cells were passaged twice per week with trypsin. All cell culture reagents were obtained from Sigma-Aldrich (St. Louis, MO). Only early passages of the cell line were used in the experiments.

In vivo tumor growth

B16F10 melanoma cells (1.0  106) in 100 ml sterile phosphate buffered saline were injected subcutaneously into the shaved right flank of anesthetized mice. Tumor growth was monitored on days 4, 7, 11, 14, 15, 17, 18, and 21 after tumor injection. The tumor volume was calculated using the equation: V ¼ 4p(L1  L22) / 3, where V ¼ volume (mm3), L1 ¼ longest diameter (mm), and L2 ¼ shortest diameter (mm).

Histological examination and immunohistochemistry See Supplementary Materials and Methods online.

Single-cell suspensions of tumor-infiltrating lymphocytes for flow cytometry Subcutaneous tumors were minced on day 11 after inoculation and then digested in complete medium (RPMI 1640 media containing 10% fetal bovine serum, 2 mg/ml crude collagenase (SigmaAldrich), 1.5 mg/ml hyaluronidase (Sigma-Aldrich), and 0.03 mg/ml DNase I (Sigma-Aldrich) at 37 C for 90 minutes to prepare tumor cell suspensions for flow cytometry. Digested cells were passed through a 70-mm cell strainer (BD Biosciences, San Jose, CA) to generate single-cell suspensions. The cell suspension was centrifuged at 460g for 5 minutes. The pellet was resuspended in 70% Percoll solution (GE Healthcare, Little Chalfont, UK) and then overlaid with a 37% Percoll solution (GE Healthcare), followed by centrifugation at 500g for 20 minutes at 4 C. The cells were

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity aspirated from the Percoll interface and passed through a 70-mm cell strainer (BD Biosciences). Cells were harvested by centrifugation and washed with phosphate buffered saline.

Flow cytometry See Supplementary Materials and Methods.

Intracellular cytokine staining See Supplementary Materials and Methods.

Isolation and adoptive transfer of B cells See Supplementary Materials and Methods.

Statistics Data are presented as mean  standard error of the mean. Unpaired two-sided Student t test was used to compare two groups, and P < 0.05 was considered significant. Three groups were compared by one-way analysis of variance, followed by Tukey’s multiple comparison test. Comparisons among tumor volumes and subsets of TILs were performed by two-way analysis of variance, followed by Bonferroni post-test. Survival was analyzed with the Mantel-Cox logrank test to compare Kaplan-Meier survival curves. Data were analyzed with GraphPad Prism software, version 5 (GraphPad, La Jolla, CA). DATA AVAILABILITY STATEMENT This study does not include large data sets, such as gene expression arrays, single nucleotide polymorphism arrays, proteomic data sets, high throughput sequencing, or genome-wide association study data.

ORCIDs Tadahiro Kobayashi: https://orcid.org/0000-0003-4367-2336 Kyosuke Oishi: https://orcid.org/0000-0002-9511-0100 Ai Okamura: https://orcid.org/0000-0001-8106-7984 Shintaro Maeda: https://orcid.org/0000-0001-7824-7060 Akito Komuro: https://orcid.org/0000-0002-5402-1515 Yasuhito Hamaguchi: https://orcid.org/0000-0001-5305-7770 Manabu Fujimoto: https://orcid.org/0000-0002-3062-4872 Kazuhiko Takehara: https://orcid.org/0000-0001-7790-0734 Takashi Matsushita: https://orcid.org/0000-0002-1617-086X

CONFLICT OF INTEREST The authors state no conflict of interest.

ACKNOWLEDGMENTS This work was supported by Japan Society for the Promotion of Science KAKENHI (grant no. JP16K18449). The authors thank Masako Matsubara and Yuko Yamada for technical assistance.

AUTHOR CONTRIBUTIONS TK contributed to conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, validation, visualization, writingeoriginal draft preparation, and writingereview and editing. TM contributed to conceptualization, methodology, project administration, supervision, and writingereview and editing. KO, SM, AO, YH, and MF contributed to resources. AK and KT contributed conceptualization and supervision.

SUPPLEMENTARY MATERIAL Supplementary material is linked to the online version of the paper at www. jidonline.org, and at https://doi.org/10.1016/j.jid.2019.02.016.

REFERENCES Ansel KM, Harris RB, Cyster JG. CXCL13 is required for B1 cell homing, natural antibody production, and body cavity immunity. Immunity 2002;16:67e76. Bakker AB, Schreurs MW, de Boer AJ, Kawakami Y, Rosenberg SA, Adema GJ, et al. Melanocyte lineage-specific antigen gp100 is recognized by melanomaderived tumor-infiltrating lymphocytes. J Exp Med 1994;179:1005e9. Bouaziz JD, Yanaba K, Tedder TF. Regulatory B cells as inhibitors of immune responses and inflammation. Immunol Rev 2008;224:201e14.

Chiaruttini G, Mele S, Opzoomer J, Crescioli S, Ilieva KM, Lacy KE, et al. B cells and the humoral response in melanoma: the overlooked players of the tumor microenvironment. Oncoimmunology 2017;6(4):e1294296. Cole DJ, Weil DP, Shamamian P, Rivoltini L, Kawakami Y, Topalian S, et al. Identification of MART-1-specific T-cell receptors: T cells utilizing distinct T-cell receptor variable and joining regions recognize the same tumor epitope. Cancer Res 1994;54:5265e8. Del Prete G, De Carli M, Almerigogna F, Giudizi MG, Biagiotti R, Romagnani S. Human IL-10 is produced by both type 1 helper (Th1) and type 2 helper (Th2) T cell clones and inhibits their antigen-specific proliferation and cytokine production. J Immunol 1993;150:353e60. DiLillo DJ, Yanaba K, Tedder TF. B cells are required for optimal CD4þ and CD8þ T cell tumor immunity: therapeutic B cell depletion enhances B16 melanoma growth in mice. J Immunol 2010;184:4006e16. Fillatreau S, Sweenie CH, McGeachy MJ, Gray D, Anderton SM. B cells regulate autoimmunity by provision of IL-10. Nat Immunol 2002;3: 944e50. Fotaki G, Jin C, Kerzeli IK, Ramachandran M, Martikainen MM, KarlssonParra A, et al. Cancer vaccine based on a combination of an infectionenhanced adenoviral vector and pro-inflammatory allogeneic DCs leads to sustained antigen-specific immune responses in three melanoma models. Oncoimmunology 2018;7(3):e1397250. Geherin SA, Go´mez D, Glabman RA, Ruthel G, Hamann A, Debes GF. IL10þ Innate-like B cells are part of the skin immune system and require a4b1 integrin to migrate between the peritoneum and inflamed skin. J Immunol 2016;196:2514e25. Gorosito Serra´n M, Fiocca Vernengo F, Beccaria CG, Acosta Rodriguez EV, Montes CL, Gruppi A. The regulatory role of B cells in autoimmunity, infections and cancer: perspectives beyond IL10 production. FEBS Lett 2015;589:3362e9. Groszer M, Erickson R, Scripture-Adams DD, Lesche R, Trumpp A, Zack JA, et al. Negative regulation of neural stem/progenitor cell proliferation by the Pten tumor suppressor gene in vivo. Science 2001;294(5549):2186e9. Hardy RR, Hayakawa K. Perspectives on fetal derived CD5þ B1 B cells. Eur J Immunol 2015;45:2978e84. Hodi FS, O’Day SJ, McDermott DF, Weber RW, Sosman JA, Haanen JB, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med 2010;363:711e23. Inoue S, Leitner WW, Golding B, Scott D. Inhibitory effects of B cells on antitumor immunity. Cancer Res 2006;66:7741e7. Iwata Y, Matsushita T, Horikawa M, Dilillo DJ, Yanaba K, Venturi GM, et al. Characterization of a rare IL-10-competent B-cell subset in humans that parallels mouse regulatory B10 cells. Blood 2011;117:530e41. Kantor A. A new nomenclature for B cells. Immunol Today 1991;12:388. Kato H, Liao Z, Mitsios JV, Wang HY, Deryugina EI, Varner JA, et al. The primacy of b1 integrin activation in the metastatic cascade. PLoS One 2012;7(10):e46576. Kawakami Y, Eliyahu S, Sakaguchi K, Robbins PF, Rivoltini L, Yannelli JR, et al. Identification of the immunodominant peptides of the MART-1 human melanoma antigen recognized by the majority of HLA-A2-restricted tumor infiltrating lymphocytes. J Exp Med 1994;180:347e52. Khan AR, Hams E, Floudas A, Sparwasser T, Weaver CT, Fallon PG. PD-L1hi B cells are critical regulators of humoral immunity. Nat Commun 2015;6:5997. Kobayashi T, Hamaguchi Y, Hasegawa M, Fujimoto M, Takehara K, Matsushita T. B cells promote tumor immunity against B16F10 melanoma. Am J Pathol 2014;184:3120e9. Kuhn R, Lohler J, Rennick D, Rajewsky K, Muller W. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 1993;75:263e74. Lampropoulou V, Calderon-Gomez E, Roch T, Neves P, Shen P, Stervbo U, et al. Suppressive functions of activated B cells in autoimmune diseases reveal the dual roles of Toll-like receptors in immunity. Immunol Rev 2010;233:146e61. Li W, Fu M, An JG, Xing Y, Zhang P, Zhang X, et al. Host defence against C. albicans infections in IgH transgenic mice with V(H) derived from a natural anti-keratin antibody. Cell Microbiol 2007;9:306e15. Lipsky PE. Systemic lupus erythematosus: an autoimmune disease of B cell hyperactivity. Nat Immunol 2001;2:764e6. Liu Y, Wei SH, Ho AS, de Waal Malefyt R, Moore KW. Expression cloning and characterization of a human IL-10 receptor. J Immunol 1994;152:1821e9.

www.jidonline.org

1543

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity Lykken JM, Candando KM, Tedder TF. Regulatory B10 cell development and function. Int Immunol 2015;27:471e7. Matsushita T, Kobayashi T, Mizumaki K, Kano M, Sawada T, Tennichi M, et al. BAFF inhibition attenuates fibrosis in scleroderma by modulating the regulatory and effector B cell balance. Sci Adv 2018;4(7):eaas9944.

Shen P, Fillatreau S. Antibody-independent functions of B cells: a focus on cytokines. Nat Rev Immunol 2015;15:441e51.

Matsushita T, Le Huu D, Kobayashi T, Hamaguchi Y, Hasegawa M, Naka K, et al. A novel splenic B1 regulatory cell subset suppresses allergic disease through phosphatidylinositol 3-kinase-Akt pathway activation. J Allergy Clin Immunol 2016;138:1170e11782 e9.

Shulman Z, Cohen SJ, Roediger B, Kalchenko V, Jain R, Grabovsky V, et al. Transendothelial migration of lymphocytes mediated by intraendothelial vesicle stores rather than by extracellular chemokine depots. Nat Immunol 2011;13:67e76.

Matsushita T, Yanaba K, Bouaziz JD, Fujimoto M, Tedder TF. Regulatory B cells inhibit EAE initiation in mice while other B cells promote disease progression. J Clin Invest 2008;118:3420e30.

Suzuki A, Kaisho T, Ohishi M, Tsukio-Yamaguchi M, Tsubata T, Koni PA, et al. Critical roles of Pten in B cell homeostasis and immunoglobulin class switch recombination. J Exp Med 2003;197:657e67.

Mauri C, Gray D, Mushtaq N, Londei M. Prevention of arthritis by interleukin 10-producing B cells. J Exp Med 2003;197:489e501.

Topalian SL, Gonzales MI, Parkhurst M, Li YF, Southwood S, Sette A, et al. Melanoma-specific CD4þ T cells recognize nonmutated HLA-DRrestricted tyrosinase epitopes. J Exp Med 1996;183:1965e71.

Miles K, Simpson J, Brown S, Cowan G, Gray D, Gray M. Immune tolerance to apoptotic self is mediated primarily by regulatory B1a cells. Front Immunol 2017;8:1952.

Vadasz Z, Haj T, Kessel A, Toubi E. B-regulatory cells in autoimmunity and immune mediated inflammation. FEBS Lett 2013;587:2074e8.

Mizoguchi E, Mizoguchi A, Preffer FI, Bhan AK. Regulatory role of mature B cells in a murine model of inflammatory bowel disease. Int Immunol 2000;12:597e605.

Watt V, Ronchese F, Ritchie D. Resting B cells suppress tumor immunity via an MHC class-II dependent mechanism. J Immunother 2007;30: 323e32.

Nelson BH. CD20þ B cells: the other tumor-infiltrating lymphocytes. J Immunol 2010;185:4977e82.

Yanaba K, Bouaziz JD, Haas KM, Poe JC, Fujimoto M, Tedder TF. A regulatory B cell subset with a unique CD1dhiCD5þ phenotype controls T celldependent inflammatory responses. Immunity 2008;28:639e50.

Nielsen JS, Sahota RA, Milne K, Kost SE, Nesslinger NJ, Watson PH, et al. CD20þ tumor-infiltrating lymphocytes have an atypical CD27- memory phenotype and together with CD8þ T cells promote favorable prognosis in ovarian cancer. Clin Cancer Res 2012;18:3281e92. Rickert RC, Roes J, Rajewsky K. B lymphocyte-specific, Cre-mediated mutagenesis in mice. Nucleic Acids Res 1997;25:1317e8. Robert C, Thomas L, Bondarenko I, O’Day S, Weber J, Garbe C, et al. Ipilimumab plus dacarbazine for previously untreated metastatic melanoma. N Engl J Med 2011;364:2517e26. Rodrı´guez-Pinto D. B cells as antigen presenting cells. Cell Immunol 2005;238:67e75.

1544

Rose DM, Cardarelli PM, Cobb RR, Ginsberg MH. Soluble VCAM-1 binding to alpha4 integrins is cell-type specific and activation dependent and is disrupted during apoptosis in T cells. Blood 2000;95:602e9.

Journal of Investigative Dermatology (2019), Volume 139

Zacca ER, Onofrio LI, Acosta CDV, Ferrero PV, Alonso SM, Ramello MC, et al. PD-L1(þ) Regulatory B cells are significantly decreased in rheumatoid arthritis patients and increase after successful treatment. Front Immunol 2018;9:2241. Zhang C, Xin H, Zhang W, Yazaki PJ, Zhang Z, Le K, et al. CD5 Binds to interleukin-6 and induces a feed-forward loop with the transcription factor STAT3 in B cells to promote cancer. Immunity 2016a;44:913e23. Zhang J, Wan M, Ren J, Gao J, Fu M, Wang G, et al. Positive selection of B10 cells is determined by BCR specificity and signaling strength. Cell Immunol 2016b;304e305:27e34.

T Kobayashi et al.

Suppression of Melanoma Tumor Immunity

SUPPLEMENTARY MATERIALS AND METHODS Histological examination and immunohistochemistry

For immunohistochemical staining of B220 and CD3, subcutaneous tumors harvested with surrounding skin on day 11 after inoculation, freshly dissected spleens, and freshly dissected inguinal lymph nodes were fixed in 3.5% paraformaldehyde and then paraffin-embedded. Sections (4 mm in thickness) were then incubated with rat mAbs specific for mouse B220 (RA3-6B2) (BD Biosciences) and CD3 (CD3-12) (AbD Serotic, Oxford, UK). Sections of subcutaneous tumors were developed with New Fuchsin (Daco Japan, Tokyo, Japan) to stain the targeted cells red for differentiation from the brown melanin pigment, followed by counterstaining with hematoxylin. Representative images (100 magnification) of tumor samples were acquired using a BX50 microscope (Olympus, Tokyo, Japan) equipped with a DP73 camera (Olympus). The numerical aperture of the objective lens was 0.40, and the acquisition software was cellSens Standard, version 1.15 (Olympus).

optimal concentrations for multicolor immunofluorescence analysis using a FACSCanto II flow cytometer (BD Biosciences). Data were analyzed using FlowJo software, version 8.8.7 (Tree Star, Ashland, OR). Intracellular cytokine staining

To detect cytokine production, T cells were stimulated for 5 hours with phorbol 12-myristate 13-acetate (50 ng/ml; Sigma-Aldrich), ionomycin (1 mg/ml; Sigma-Aldrich), and brefeldin A (3 mM; BioLegend). B cells were stimulated for 5 hours with lipopolysaccharide (10 mg/ml; Escherichia coli serotype 0111: B4; Sigma-Aldrich), phorbol 12-myristate 13-acetate (50 ng/ml), ionomycin (1 mg/ml), and brefeldin A (3 mM). Natural killer cells were not stimulated for detection of granzyme B. After cell-surface staining, the cells were washed, fixed, and permeabilized using a Cytofix/Cytoperm Kit (BD Biosciences), followed by staining with antieIFN-g (XMG1.2), antieTNF-a (MP6-XT22), antieIL-10 (JES5-16E3), and antiegranzyme B (GB11) mAbs from BioLegend. Isolation and adoptive transfer of B cells

Flow cytometry

The following mAbs were used: fluorescein isothiocyanate-, phycoerythrin-, phycoerythrin-cyanine 7e, peridininchlorophyll-protein complex-cyanine 5.5e, allophycocyanin-, allophycocyanin-cyanine 7e, and Pacific Blueeconjugated mAbs to mouse B220 (RA3-6B2), CD19 (1D3), CD1d (1B1), and CD5 (53-7.3) from BD Biosciences (San Jose, CA); CD3ε (145-2C11), CD4 (RM4-5), CD90.2 (30-H12), CD21 (7E9), CD23 (B3B4), CD25 (PC61.5), and CD43 (1B11) from BioLegend (San Diego, CA); CD8a (53-6.7), and NK1.1 (PK136) from eBioscience (San Diego, CA); and LIVE/DEAD Fixable Aqua Dead Cell Stain from Invitrogen (Carlsbad, CA). Single-cell suspensions of splenocytes and TILs were stained at 4 C for 20 minutes using mAbs at predetermined

Splenic B cells from Cd19Creþ/ePtenloxp/loxp mice were isolated using the Pan B Cell Isolation Kit (Miltenyi Biotech, Bergisch Gladbach, Germany) with purities of approximately 95%. Subsequently, CD5þ B1a B cells and CD5e non-B1a B cells were isolated using CD5 (Ly-1) MicroBeads (Miltenyi Biotech). Splenic B1a cells from WT and IL-10e/e mice were purified using anti-CD19 mAb-coated microbeads (Miltenyi Biotech) and by means of cell sorting with the FACS Aria Fusion (BD Bioscience). Purified B1a B cells (1.0  106) and non-B1a B cells (1.0  106) were respectively mixed with B16F10 melanoma cells (2.0  105) in 100 ml sterile phosphate buffered saline. The mixtures were injected subcutaneously into the shaved right flank of anesthetized wild-type mice.

Cd19Cre +/105

Cd19Cre+/- Pten 105

105

3.83%

104

104

103

103

105

9.46%

0.16% 10

loxP/loxP

4

13.4%

104

60.1%

102

103

104

0

105

B220

104

104

103

103

95.3% Cell number

105

CD23

102

103

104

102

92.4%

0

102

CD21

103

104

105

0

102

103

104

105

0 102

0

103

104

CD1dint CD5+

isotype control

0

102

CD43

103

104

105

0

25.8% 0

102

103

104

105

CD5

105

105

104

104

103

103

92.2%

102

102

105

B220

0

0

102

105

CD5

105

102

13.2% 0

CD23

0

102

81.1% Cell number

0

103

CD1d

102

103

CD19

CD1d

CD19

85.7%

102

0

CD1dint CD5+

isotype control

0

0

102

CD21

103

104

105

0

102

103

104

105

0

102

103

104

105

CD43

Supplementary Figure S1. Confirmation of phenotypically defined tumor-infiltrating B cell subsets. Tumor-infiltrating lymphocytes from Cd19Creþ/ePtenloxp/loxp and Cd19Creþ/emice were stained for CD19, B220, CD5, CD1d, CD21, CD23, and CD43. Representative gating used for identification of tumor-infiltrating CD1dintCD5e B cells, CD1dhiCD5e cells, and CD1dintCD5þ B cells with CD21intCD23hi follicular B cell, CD21hiCD23lo marginal zone B cell, and CD5þCD43þ B1a B cell phenotypes, respectively, during flow cytometric analysis. Cell-surface expression of CD43 was determined in CD1dintCD5þ B cells (lines) or isotype control (shaded gray). Cell frequencies within the indicated gates are shown. All data are representative of two independent experiments. Cd19Creþ/ePtenloxp/loxp, B cellespecific PTEN-deficient mice.

www.jidonline.org 1544.e1