Diabetologia

, 54:1777

Transgenic expression of murine chemokine decoy receptor D6 by islets reveals the role of inflammatory CC chemokines in the development of autoimmune diabetes in NOD mice

Authors

  • G.-J. Lin
    • Department of Biology and AnatomyNational Defense Medical Center
    • Department of Microbiology and ImmunologyNational Defense Medical Center
  • S.-H. Huang
    • Department of General SurgeryTri-Service General Hospital
  • Y.-W. Chen
    • Department of Oral and Maxillofacial SurgeryTri-Service General Hospital
  • D.-Y. Hueng
    • Department of Neurological SurgeryTri-Service General Hospital
    • Keelung Civilian Administration DivisionTri-Service General Hospital
  • W.-T. Chia
    • Hsin Chu General Hospital, Department of Health
  • M.-W. Chien
    • Graduate Institute of Life SciencesNational Defense Medical Center
  • B. L. Yen
    • Institute of Cellular and System MedicineNational Health Research Institutes
    • Department of Microbiology and ImmunologyNational Defense Medical Center
    • Graduate Institute of Life SciencesNational Defense Medical Center
Article

DOI: 10.1007/s00125-011-2166-9

Cite this article as:
Lin, G., Huang, S., Chen, Y. et al. Diabetologia (2011) 54: 1777. doi:10.1007/s00125-011-2166-9

Abstract

Aims/hypothesis

Autoimmune diabetes results from a progressive destruction of insulin-producing beta cells in the pancreatic islets by chemokine-attracted lymphocytes. Because islet cells in NOD mice produce chemokines during the development of autoimmune diabetes, we investigated the role of inflammatory CC chemokines in disease progression in these mice.

Methods

We generated a transgenic NOD mouse model that overproduces the inflammatory CC chemokine decoy receptor D6 in pancreatic islets.

Results

The frequency of diabetes and insulitis scores of transgenic mice were decreased significantly, compared with non-transgenic control littermates. Transgenic expression of D6 (also known as Ccbp2) did not affect systemic lymphocyte development or alter: (1) the T cell subsets such as T helper (Th)1, Th2 and T regulatory cells; or (2) antigen-presenting cells such as dendritic cells or macrophages. The percentages and numbers of T and B lymphocytes were decreased significantly in the pancreas. Activation status, autoantigen-specific proliferation and diabetogenicity of lymphocytes were also markedly reduced.

Conclusions/interpretation

Inflammatory CC chemokines play a critical role in the development of autoimmune diabetes. Transgenic expression of D6 in pancreatic islets of NOD mice reduced this pathogenic process by suppressing activation of autoreactive lymphocytes and by reducing migration of lymphocytes to the pancreas.

Keywords

ChemokineD6NOD miceType 1 diabetes

Abbreviations

CCL

CC chemokine ligand

CCR

CC chemokine receptor

FOXP3

Forkhead box P3

GFP

Green fluorescent protein

hThy1

Human Thy-1 cell surface antigen

mD6

Murine D6

mThy1.1

Mouse thymus cell antigen 1, theta

PIL

Pancreas-infiltrating lymphocytes

PLN

Pancreatic lymph nodes

pINS-mD6

Insulin promoter-driven murine D6 construct

SCID

Severe combined immunodeficient

Th

T helper

Treg

T regulatory

Introduction

Type 1 diabetes mellitus results from the destruction of insulin-producing beta cells in the islets of the pancreas and has been identified as a T cell-mediated autoimmune disease [1]. The development of type 1 diabetes is usually diagnosed in young patients, so this disease is also referred to as juvenile- or childhood-onset diabetes. The NOD mouse spontaneously develops T cell-dependent beta cell destruction that resembles human type 1 diabetes and serves as an animal model of this autoimmune disease [2].

Chemokines are small heparin-binding proteins that induce the migration of circulating leucocytes to sites of inflammation or injury. They include the CC, CXC, CX3C and XC chemokine families [3]. The CC chemokine ligand (CCL) family is the largest group, members of which can be defined as homeostatic chemokines or inflammatory chemokines, depending on their biological functions. Inflammatory CCLs include CCL1 to CCL5; their main function is to attract mononuclear cells to sites of chronic inflammation [4].

The pathogenesis of type 1 diabetes involves the infiltration of immune cells into the pancreas, a process dependent on the attraction of these cells by chemokines. Islet cells in NOD mice have been reported to produce high amounts of CCL2 and CCL5 compared with BALB/c mice [5, 6]. Moreover, CC chemokine receptor (CCR)4-related chemokines such as CCL17 and CCL22 are involved in the pathogenesis of autoimmune diabetes in NOD mice [7]. CCL5 is produced on the islets of NOD mice and neutralisation of CCL5 by an anti-CCR5 antibody decreases the severity of insulitis and prevents islet destruction [8]. On the other hand, Grewal et al. demonstrated that overproduction of CCL2 on the beta cells of C57BL/6 × C3H mice leads to insulitis [9]. Transgenic expression of CCL2 by beta cells led to insulitis and islet destruction, and promoted the development of diabetes in C57BL/6 × DBA mice [10]. Furthermore, beta cell-specific production of a pan-chemokine blocking protein, M3, which was originally encoded in the genome of murine γ–herpesvirus 68, inhibited the migration of lymphocytes to islets and completely prevented the onset of diabetes in NOD mice [11]. Collectively, these results indicate that chemokines participate in the pathogenesis of autoimmune diabetes and that modulation of these chemokines affects disease progression and severity. However, the physiological roles of individual chemokine families in autoimmune diabetes have not been fully elucidated.

D6 is an inflammatory CC chemokine decoy receptor that exhibits a 7-transmembrane-domain similar to that of conventional chemokine receptors [12, 13], yet lacks the ability for intracellular signal transduction. It is mainly produced in the placenta and on the endothelial cells of lymphatic afferent vessels in the skin, gut and lung [14]. D6 binds to a broad range of inflammatory CC chemokines, including most ligands for CCR1 to CCR5 [15], and targets them for intracellular degradation [1620]. Strikingly, D6-deficient mice displayed delayed clearance of cutaneous inflammatory CC chemokines and developed a pathological condition in skin that resembled human psoriasis, indicating that D6 is involved in resolution of the cutaneous inflammatory response [21]. Another study has also demonstrated an increased inflammatory response in the skin of D6-deficient mice [22]. That study also demonstrated that the amount of CCL2 and the numbers of lymphocytes in skin-draining lymph nodes were increased in D6-deficient mice, supporting a critical role of D6 in the control of lymphocyte migration in lymph nodes [22].

To investigate the potential role of inflammatory CC chemokines in the pathogenesis of autoimmune diabetes, we generated a transgenic NOD mouse that expresses D6 (also known as Ccbp2) specifically on pancreatic beta cells. By analysing the process of diabetes and the status of immune responses, we demonstrated that inflammatory CC chemokines played a critical role in the development of autoimmune diabetes in this animal model, Moreover, transgenic expression of D6 in pancreatic islets was able to effectively reduce this pathogenic process by suppressing the activation of autoreactive lymphocytes and reducing migration of lymphocytes to the pancreas.

Methods

Animals

Inbred NOD/Sytwu mice (Kd, Db, Ld, I-Ag7) and NOD/severe combined immunodeficient (SCID) mice were originally purchased from the Jackson Laboratory (Bar Harbor, ME, USA) and subsequently bred at the animal centre of the National Defense Medical Center in Taipei, Taiwan, under specific pathogen-free conditions. T1/T2 double transgenic NOD mice were generated by back-crossing T1/T2 double transgenic BALB/c mice to NOD mice for more than 12 generations, as described [23]. Experiments were conducted in accordance with institutional guidelines and were approved by National Defense Medical Center’s Institutional Animal Care and Use Committee.

Real-time RT–PCR

Total RNA was prepared from splenocytes using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and then reverse-transcribed by SuperScript III Reverse Transcriptase (Invitrogen). Real-time RT-PCR was performed by iQ SYBR Green PCR supermix (Bio-Rad, Hercules, CA, USA) in an iCycler (Bio-Rad). Fold changes were measured using the ΔΔCt method. The sequences of primers used in this experiment are shown in electronic supplementary material (ESM) Table 1.

Generation of murine D6 transgenic NOD mice

The D6 gene of the NOD mouse was cloned from genomic DNA of mouse tail prepared by phenol–chloroform extraction and amplified by PCR. The PCR products were cloned into pINS plasmid for the generation of transgenic mice [24]. Murine D6 (mD6) DNA was inserted into the pINS plasmid to generate an insulin promoter-driven murine D6 construct (pINS-mD6). The linearised DNA fragment carrying pINS and D6 DNA was purified and microinjected into the pronuclei of NOD embryos at the one-cell stage, then implanted into pseudopregnant (BALB/c × FVB) F1 females. Southern blotting was performed to confirm the presence of the pINS-mD6 transgene. All transgenic mice used in our study were heterozygous for the D6 transgene.

Western blot

Protein samples were separated on 10% (wt/vol.) SDS-PAGE gel and then transferred to a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). The presence of D6 and β-actin on the membrane was detected with anti-D6 antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) or anti-β-actin antibody (Sigma-Aldrich, St Louis, MO, USA), respectively. The signals were detected by an imaging system (LAS-3000; Fujifilm, Tokyo, Japan).

Assessment of insulitis and diabetes

Pancreatic tissues were obtained from 14-week-old transgenic or non-transgenic mice and the severity of insulitis was scored on haematoxylin–eosin-stained sections and classified as described [25]. Urine glucose concentrations (glycosuria) were measured weekly or every other day using Chemstrips (Boehringer Mannheim, Indianapolis, IN, USA). Mice with urine glucose concentration >27.75 mmol/l on two consecutive tests were defined as diabetic.

T cell proliferation

Splenocytes were isolated from the spleen of 8-week-old non-transgenic or mD6 transgenic mice. T cell proliferation was performed as described [26, 27]. Splenocytes were treated with TRIS-buffered ammonium chloride to lyse the erythrocytes. After washing, splenocytes were resuspended at 5 × 106 cells/ml in RPMI 1640 medium. Cells (5 × 105 cells/well) were stimulated with an immobilised anti-CD3 antibody (BD Biosciences, Pharmingen, San Diego, CA, USA), concanavalin A or islet antigens. The islet antigens were prepared by two cycles of freeze and thaw on islet cells isolated from young male NOD mice (less than 7 weeks old) and dissociated by cell dissociation buffer (Invitrogen). Lysed islet cells (5 × 104) were added to each well. The incorporation of methyl-[3H]thymidine was detected at 72 h with a liquid scintillation counter (TopCount; Packard Instruments, Meriden, CT, USA).

Flow cytometry

Lymphocytes were taken from the spleen, pancreatic lymph nodes (PLN) or pancreas of mD6 transgenic or non-transgenic mice. Isolated cells were stained with antibodies to mouse CD4, CD8, CD19, CD25, CD11b, CD11c, CD80, CD86 and forkhead box P3 (FOXP3) (BD Biosciences), and to human CD90 (hThy1, clone 5E10) and mouse CD90.1 (mThy1.1, clone OX-7) (BD Biosciences). Flow cytometry was performed using a FACSCalibur (BD Biosciences).

Adoptive transfer

Splenocytes of female mD6 transgenic or non-transgenic NOD donors were treated with TRIS-buffered ammonium chloride for erythrocyte depletion and 2 × 107 cells were injected into female NOD/SCID mice via the retro-orbital plexus. In lymphocyte transfer-induced diabetes experiments, 6-week-old female mD6 transgenic mice or their non-transgenic control littermates were selected as recipients and injected with 2 × 107 splenocytes taken from 14-week-old female NOD donors using the procedure described above. Diabetes was assessed as described earlier.

Isolation and analysis of pancreas-infiltrating lymphocytes

Pancreas-infiltrating lymphocytes (PIL) were isolated as follows. Collagenase XI (Sigma-Aldrich) was injected into the common bile duct of 14-week-old mD6 transgenic or non-transgenic mice and the pancreases were isolated. After washing with RPMI-1640 medium, islets and lymphocytes in the pancreas were separated by a density gradient using Histopaque 1077–1 (Sigma-Aldrich). The cells separating into the density gradient interface were collected and incubated for 1 to 2 min with 1 ml cell dissociation buffer (GIBCO Invitrogen) at room temperature. After passing them through a cell strainer, suspended single cells were counted and stained with antibodies for flow cytometry.

Statistics

Data are presented as the mean ± SD or mean ± SEM. The significance of any difference in the frequency of diabetes was determined by Kaplan–Meier survival analysis. For the remaining experiments, p values were calculated using two-tailed Student’s t tests. Differences were considered significant at p < 0.05.

Results

Chemokine expression profiles in the islets of NOD mice and the generation of mD6 transgenic NOD mice

To assess the expression kinetics of chemokines in the islets of NOD mice, we used real-time RT-PCR to measure the expression of chemokines in isolated islets from NOD mice at different ages and compare it with that in islets from 12-week-old Balb/c mice. Our results indicate that NOD islets express higher levels, but diverse kinetics, of chemokines, including several members of inflammatory CC chemokines, compared with Balb/c islets (ESM Fig. 1). To investigate the potential of an organ-specific murine D6 transgene in the prevention of autoimmune diabetes, we used a pINS-mD6 construct to create a transgenic NOD mouse model expressing the transgene in a beta cell-specific manner. The pINS-mD6 construct contains the insulin promoter, the first and part of the second non-coding exons of insulin gene, an intron from rabbit β-globin, mouse D6 DNA and a poly-A signal from rabbit β-globin (Fig. 1a). Transgenic positive mD6 NOD mice can be distinguished easily from their non-transgenic littermates by PCR genotyping with P1 and P2 primers (Fig. 1b). Southern blot analysis of tail genomic DNA indicated that the founder mouse carried five to ten copies of the transgene (Fig. 1c). To determine the level of transcription of the D6 transgene specifically, we performed RT-PCR using the P1 and P2 primers. Except for thymus and pancreas, none of the organs tested (lung, heart, liver, spleen and intestine) from transgenic mice transcribed RNA of transgenic mD6 (Fig. 1d). A trace of transgenic D6 expression in the thymus could be caused by the effect of the autoimmune regulator (Aire) [28]. These results demonstrated that mD6 transgenic NOD mice specifically expressed transgenic D6 in the pancreas, indicating the high stringency of the insulin promoter. Western blot analysis also confirmed overproduction of D6 at the protein level in islets from mD6 transgenic NOD mice (Fig. 1e).
https://static-content.springer.com/image/art%3A10.1007%2Fs00125-011-2166-9/MediaObjects/125_2011_2166_Fig1_HTML.gif
Fig. 1

Generation of mD6 transgenic NOD mice. a Scheme of the transgene construct. The black areas represent exons and the grey areas represent introns. A forward primer (P1) targeting the second exon of the insulin gene and a backward primer (P2) targeting the coding region of mD6 were designed to specifically evaluate pINS-mD6 transcription. b Genomic DNA was extracted from the tails of mD6 transgenic (Tg) or non-transgenic mice. The extracted genomic DNA was then amplified by PCR using primers P1 and P2. D6 indicates the product of the transgene of mD6. The Il23 (also known as Il23a) gene was amplified as an internal control for genomic DNA PCR. c Southern blot analysis of the transgenic D6 signal in mD6 transgenic founders is denoted as D6. The standard C0 to C20 was prepared by different copy numbers of the linearised pINS-mD6 plasmid, which was used in the microinjection for generation of mD6 transgenic NOD mice. Southern blot analysis revealed that there were five to ten copies of the transgene in the genome of the pINS-mD6 transgenic founder mouse. d Tissue-specific expression by mD6 transgenic NOD mice. Total RNA purified from various organs of mD6 transgenic and non-transgenic mice was reverse -transcribed to cDNA using oligo-dT primers and further amplified by PCR using primers P1 and P2 for transgenic mD6 expression. Hgprt (also known as Hprt) was used as an internal control. e Test organs were prepared from mD6 transgenic and non-transgenic mice and the abundance of D6 protein was measured by western blotting. β-Actin was used as a protein loading control

To evaluate the neutralisation effects of transgenic D6 on islets, we detected the amount of two D6-binding chemokines (CCL2 and CCL5) and one D6-non-binding chemokine (CXCL10) as control. The amount of CCL2 and CCL5 was decreased in the islets of mD6 transgenic mice, compared with non-transgenic littermates. In contrast, the level of CXCL10 was similar in islets between mD6 transgenic and non-transgenic mice (ESM Fig. 2). These data suggest that overproduction of D6 by islets specifically decreased the amount of inflammatory CC chemokines such as CCL2 and CCL5 in the islets of NOD mice.

Expression of D6 in NOD islets reduced the severity of insulitis and development of diabetes

To investigate the protective effects of organ-specific and membrane-bound expression of transgenic D6 on autoimmune diabetes, we analysed the incidence of spontaneous diabetes and the severity of insulitis in mD6 transgenic mice. The mD6 transgenic NOD mice were significantly prevented from developing autoimmune diabetes compared with non-transgenic littermates (p < 0.01; Fig. 2a), demonstrating islet-specific protection of transgenic D6 in NOD mice. Furthermore, the severity of insulitis in mD6 transgenic mice was significantly reduced (Fig. 2b). To determine whether the D6 transgene prevents islet inflammation, we compared the insulitis scores of transgenic and control mice at 8, 14 and 20 weeks. The severity of insulitis was reduced significantly at all ages examined (Fig. 2c). These results indicate reduced infiltration of lymphocytes to pancreatic islets in mD6 transgenic mice.
https://static-content.springer.com/image/art%3A10.1007%2Fs00125-011-2166-9/MediaObjects/125_2011_2166_Fig2_HTML.gif
Fig. 2

Characterisation of the diabetogenic process in mD6 transgenic NOD mice. a Spontaneous diabetic frequency of female mD6 transgenic NOD mice. Spontaneous diabetes in female mD6 transgenic NOD mice (black circles, n = 32) or their control littermates (white circles, n = 30) was monitored by weekly measurement of glycosuria. Compared with non-transgenic control littermates, which started to develop diabetes at 12 weeks of age, mD6 transgenic mice first developed diabetes after 15 weeks, indicating a delay in disease onset. At 20 weeks of age, the incidence of diabetes in control mice increased to 40%, but the incidence in mD6 transgenic mice was only 9%. After 25 weeks, more than 66% of non-transgenic mice had developed diabetes, but the incidence in mD6 transgenic mice was still only around 22%. b Histopathology of the severity of insulitis in mD6 transgenic NOD mice. Pancreases prepared from mD6 transgenic or non-transgenic mice of different ages as labelled were fixed and embedded for histopathology using haematoxylin and eosin staining. Scale bars 100 μm. c The severity of insulitis was examined on haematoxylin and eosin-stained sections of pancreas from mD6 transgenic or non-transgenic NOD mice at the indicated ages. Black, destructive insulitis; dark grey, severe insulitis; medium grey, intra-insulitis; light grey, peri-insulitis; white grey, intact islets

Immune cell development in mD6 transgenic NOD mice

To investigate whether overexpression of D6 on pancreatic beta cells would alter the development of lymphocytes, we analysed the populations of lymphocytes taken from spleen, PLN or pancreas using flow cytometry. There were no significant differences between mD6 transgenic and non-transgenic mice in the populations of CD4+, CD8+ T or CD19+ B cells in spleen, PLN or pancreas (Fig. 3a–c).
https://static-content.springer.com/image/art%3A10.1007%2Fs00125-011-2166-9/MediaObjects/125_2011_2166_Fig3_HTML.gif
Fig. 3

Immune cell development in mD6 transgenic NOD mice. a Composition of lymphocytes in the spleen (n = 5), (b) PLN (n = 4) and (c) PIL (n = 5) from 8-week-old mD6 transgenic mice (black bars) and non-transgenic controls (white bars) was analysed by flow cytometry. d The percentages of Th1 (CD4+hThy1+), Th2 (CD4+mThy1.1+) and Treg cells in the spleen (n = 4) and (e) PLN (n = 4) from 8-week-old mD6/T1/T2 triple transgenic (black bars) or T1/T2 double transgenic (white bars) NOD mice were analysed by flow cytometry. f The percentages of Th1 (CD4+IFN-γ+), Th2 (CD4+IL-4+) and Treg lymphocytes in pancreatic tissues of mD6 transgenic (black bars) or non-transgenic (white bars) mice were subjected to intracellular cytokine staining and analysed by flow cytometry (n = 4). g Composition of CD11b- or CD11c-positive cells in the spleen or (h) PLN of mD6 transgenic (black bars, n = 3) or non-transgenic (white bars, n = 3) mice was analysed by flow cytometry. i The abundance of CD80 or CD86 on CD11c+ dendritic cells in the spleen or PLN of mD6 transgenic (black bars, n = 3) or non-transgenic (white bars, n = 3) mice was determined by mean fluorescence intensity (MFI) in flow cytometry. Data are expressed as the mean ± SEM. There were no significant differences between mD6 transgenic and non-transgenic mice

It is known that an imbalance between T helper (Th)1 and Th2 responses [29], and a relative deficiency in CD4+CD25+FOXP3+ regulatory T (Treg) cells [30, 31] and pathological dendritic cells [3237] predispose NOD mice to developing autoimmune diabetes. To investigate whether the protective effect of transgenic D6 might work through regulation of lymphocyte development, we crossed mD6 transgenic mice with T1/T2 double transgenic mice to generate mD6/T1/T2 triple transgenic mice. T1/T2 double transgenic mice bear two transgenes: (1) human THY1 under control of the murine Ifn-γ (also known as Ifng) promoter; and (2) murine Thy1.1 (also known as Thy1) under control of the murine Il4 promoter. Using these mice, the kinetic development of Th1 and Th2 cells could be measured directly by detecting the presence of human Thy-1 cell surface antigen (hThy1) (a T1 marker) and mouse thymus cell antigen 1, theta (mThy1.1) (a T2 marker), respectively. The percentages of IFN-γ-producing (hThy1+), IL-4-producing (mThy1.1+) or Treg cells in spleen, PLN and pancreas (Fig. 3d–f) were similar between mD6/T1/T2 triple (black bar) and T1/T2 double (white bar) transgenic mice, suggesting that overexpression of transgenic mD6 by islets did not alter systemic or local lymphocyte development in NOD mice.

To investigate whether overexpression of D6 on beta cells would affect the development of antigen-presenting cells such as macrophages and dendritic cells, we analysed the distribution of CD11c+ and/or CD11b+ cells in spleen and PLN by flow cytometry. There were no significant differences between mD6 transgenic and non-transgenic mice in the populations of CD11c+CD11b+, CD11c+CD11b or CD11cCD11b+ cells (Fig. 3g, h), indicating that the development of macrophages and dendritic cells was not affected by overexpression of mD6. We also assessed the activation status of dendritic cells by determining the levels of the co-stimulatory molecules CD80 or CD86 on CD11c+ cells. The mean fluorescence intensity values of CD80 or CD86 were similar between mD6 transgenic mice and non-transgenic mice in spleen and PLN (Fig. 3i), indicating that overproduction of D6 by islets did not influence activation of dendritic cells in these tissues.

Production of D6 by islets reduced infiltration and activation of lymphocytes in islets

To investigate whether overproduction of D6 by islets would impede migration of lymphocytes to the pancreas, we used flow cytometry to quantify lymphocytes among the cells isolated from the pancreas of mD6 transgenic mice. The percentages and numbers of CD3+ T cells and CD19+ B cells were decreased significantly compared with the cells from non-transgenic mice (Fig. 4a, b). Thus, overproduction of the CC chemokine decoy receptor D6 by islets effectively impeded the migration of T and B cells to the pancreas.
https://static-content.springer.com/image/art%3A10.1007%2Fs00125-011-2166-9/MediaObjects/125_2011_2166_Fig4_HTML.gif
Fig. 4

The infiltration of lymphocytes in the pancreas of mD6 transgenic NOD mice was reduced. a The percentages and (b) cell numbers of CD3+ and CD19+ cells in cells isolated from pancreatic tissues of 12-week-old mD6 transgenic (black bars) or non-transgenic (white bars) mice were analysed by flow cytometry. The proportions and cell numbers of CD3+ cells (n = 4) and CD19+ cells (n = 5) in mD6 transgenic mice were significantly less than in non-transgenic controls. Data are expressed as the mean ± SEM; *p < 0.05 and ***p < 0.001. c Characterisation of lymphocyte transfer-induced diabetes progression in mD6 transgenic NOD mice. Female 6-week-old mD6 transgenic and non-transgenic mice were injected intravenously with 2 × 107 splenocytes isolated from 14-week-old female NOD mice. The incidence of diabetes in mD6 transgenic (black circles, n = 12) and non-transgenic (white circles, n = 15) recipients was monitored by testing glycosuria every other day. The incidence was significantly lower in mD6 transgenic mice (p < 0.05). d GFP-positive splenocytes were transferred to mD6 transgenic NOD mice and their non-transgenic control littermates. The infiltration of transferred splenocytes in the islets at day 65 was assessed by immunofluorescence assay. Pancreas tissues paraffin-embedded sections were rehydrated and heated in 0.1 mol/l citrate buffer. Sections were probed with with Cy5-conjugated anti-GFP antibody (blue spots) and FITC-conjugated anti-CD45 antibody (green spots). Propidium iodide (PI) was used for nuclear counterstain (red spots). Images were captured on a Zeiss confocal LSM510 microscope (Zeiss, Thornwood, NY, USA). e Images showing the signals of CD45 or GFP were magnified. Double-positive cells are indicated by white arrows

To further investigate whether D6 overproduction in situ would protect mice from diabetes by reducing the infiltration of pathogenic lymphocytes to islets, we performed experiments on lymphocyte transfer-induced diabetes. About 40% of the non-transgenic mice developed diabetes 70 days after transfer; however, more than 90% the mD6 transgenic recipients remained diabetes-free (Fig. 4c). About 90% of non-transgenic mice became diabetic by 84 days after transfer, but only 41% of transgenic recipients were diabetic, suggesting that islet-specific production of D6 not only delayed, but also reduced the infiltration of pathogenic lymphocytes to islets. To examine whether the inhibition of transferred diabetes in mD6 transgenic recipients resulted mainly from delayed migration of transferred splenocytes, we performed the lymphocyte transfer-induced diabetes experiment by transferring splenocytes positive for green fluorescent protein (GFP) into mD6 transgenic or non-transgenic NOD mice. The GFP-positive splenocytes were isolated from GFP-transgenic NOD mice in which transgenic GFP was driven by the ubiquitin promoter. We assessed the infiltration of transferred splenocytes in islets at day 65 after transfer, using immunofluorescence assay. We found that most CD45-positive cells in islets are GFP-positive, indicating that the development of diabetes in recipients was mainly due to the transferred splenocytes (Fig. 4d, e). We also observed that infiltrated immune cells in mD6 transgenic recipients were much less numerous than in non-transgenic recipients, suggesting that overexpression of D6 on islets inhibited the migration of immune cells into islets.

The protective effect of the D6 transgene in mD6 transgenic mice is likely to result from a reduction of lymphocyte infiltration to the pancreas. To further evaluate whether this reduction of lymphocyte infiltration was caused by alteration of T cell activation status in mD6 transgenic mice, we used flow cytometry to analyse early activation of surface marker CD69 on CD4+ or CD8+ T cells in the spleen, PLN and pancreas. The percentage of CD69-positive cells in CD4+ lymphocytes was significantly reduced in the PLN and that of CD69-positive cells in CD8+ lymphocytes was significantly reduced in the PLN and PIL of mD6 transgenic mice (Fig. 5a, b), indicating that reduction of T cell activation in transgenic NOD mice is associated with mD6 overexpression.
https://static-content.springer.com/image/art%3A10.1007%2Fs00125-011-2166-9/MediaObjects/125_2011_2166_Fig5_HTML.gif
Fig. 5

Activation of CD4 and CD8 T cells in mD6 transgenic NOD mice. Abundance of the early activation marker, CD69, on lymphocytes from spleen, PLN and PIL of mD6 transgenic (black bars, n = 6) or non-transgenic (white bars, n = 6) mice was analysed by flow cytometry. a The percentage of CD4+CD69+ T cells in PLN was significantly lower in the mD6 transgenic mice. b The percentage of CD8+CD69+ T cells in the PLN and PIL of mD6 transgenic mice was significantly lower than in non-transgenic mice. Data are expressed as the mean ± SEM; *p < 0.05 and ***p < 0.001

The diabetogenicity of splenocytes was reduced in mD6 transgenic NOD mice

It is possible that the protection observed in transgenic mice was caused by modulation of the development of autoreactive T cells by transgenic expression of D6. We therefore performed adoptive transfer experiments and characterised the pathogenicity of T lymphocytes in mD6 transgenic NOD mice. NOD/SCID recipients were injected intravenously with splenocytes from 12-week-old transgenic or control mice and progression of diabetes in these two groups was compared. The diabetes incidence was significantly lower in recipients transferred with splenocytes from mD6 transgenic mice (Fig. 6a). These results indicate a significant inability of splenocytes in mD6 transgenic mice to induce diabetes and suggest a critical role of transgenic D6 in regulating autoimmunity in NOD mice. To further characterise the proliferation potential of splenocytes in mD6 transgenic mice, we stimulated those cells with an anti-CD3 antibody, with concanavalin A or with islet antigens in vitro. Splenocytes isolated from mD6 transgenic and non-transgenic mice exhibited similar proliferative capacity under stimulation with anti-CD3 antibody or concanavalin A (Fig. 6b, c), indicating that the responses of lymphocytes to non-specific mitogenic stimulators were not altered in mD6 transgenic mice. However, when stimulated with islet antigens, the proliferation of splenocytes from mD6 transgenic mice was significantly lower than in non-transgenic splenocytes (Fig. 6d), suggesting that the reduction of splenocyte proliferation in mD6 transgenic mice was islet antigen-specific.
https://static-content.springer.com/image/art%3A10.1007%2Fs00125-011-2166-9/MediaObjects/125_2011_2166_Fig6_HTML.gif
Fig. 6

Characterisation of the pathogenicity of splenocytes in mD6 transgenic NOD mice. a The incidence of diabetes in NOD/SCID recipients with splenocytes adoptively transferred from 12-week-old mD6 transgenic (black circles, n = 15) or non-transgenic (white circles, n = 15) mice was assessed by testing glycosuria every other day. In recipients with splenocytes transferred from non-transgenic controls, diabetes first developed on day 30 after transfer. By contrast, in mice receiving splenocytes from mD6 transgenic mice, the first instance of disease was on day 48. On day 82, all recipients transferred with control splenocytes became diabetic, while disease incidence in recipients with D6 splenocytes was still only 46.6%. b Proliferative capacity of T cells in mD6 transgenic NOD mice. Splenocytes isolated from mD6 transgenic (black bars) or non-transgenic (white bars) mice were stimulated with plate-bound anti-CD3 antibody (n = 4) or (c) concanavalin A (Con A, n = 5) at the indicated dosages. d Proliferative response of splenocytes to islet antigens in mD6 transgenic NOD (black bar) or non-transgenic (white bar) mice. The stimulation index of splenocytes taken from mD6 transgenic mice was significantly reduced by stimulation with islet antigens compared with non-transgenic mice; n = 3; *p < 0.05. Data are expressed as the mean ± SEM

Discussion

Chemokines have been reported to be involved in the development of autoimmune diabetes. Martin et al. demonstrated that blockade of chemokines from their ligands by transgenic overproduction of the chemokine-blocking protein M3 in beta cells protected NOD mice from developing diabetes [11]. However, the M3 protein, originally encoded in the genome of murine γ-herpesvirus 68, blocks interactions between almost all chemokines and their corresponding receptors; therefore, the physiological role of individual chemokine family members in the pathogenesis of autoimmune diabetes in NOD mice remains to be elucidated. To investigate whether inflammatory CC chemokines might play a role in the development of autoimmune diabetes, we generated a transgenic NOD mouse model that expresses the inflammatory CC chemokine decoy receptor D6 on beta cells. This significantly reduced the infiltration of lymphocytes into the pancreas and decreased the incidence of diabetes in NOD mice, indicating that inflammatory CC chemokines play a critical role in the pathogenesis of autoimmune diabetes in NOD mice. Moreover, based on the biological function and natural existence of D6 in human beings, our results further imply a therapeutic potential for D6 manipulation in the prevention or treatment of human type 1 diabetes.

To further investigate the protective mechanism(s) mediated by transgenic overexpression of D6 by islets, we analysed the systemic development of lymphocytes by flow cytometry. Our results indicated that the populations of CD4+ T cells, CD8+ T cells and CD19+ B cells were not altered in the spleen, PLN or pancreas of transgenic mice compared with control littermates. The protective mechanism in M3 transgenic mice is caused, at least in part, by a reduction of CD11b+CD11c+ dendritic cells in the PLN and islets. However, the percentages of dendritic cell subsets in spleen or PLN between mD6 transgenic and non-transgenic mice were similar, suggesting that transgenic D6-mediated protection is not likely to act through direct modulation of dendritic cell development. Our results also show that levels of the maturation markers CD80 and CD86 on dendritic cells were not altered in mD6 transgenic mice. Instead, the infiltration of CD3+ lymphocytes and CD19+ B cells in the pancreas were significantly reduced in mD6 transgenic mice, suggesting that CC chemokines and CXC chemokines might modulate the pathogenesis of autoimmune diabetes differentially in NOD mice.

It has been reported that D6 reduces the local gradient of inflammatory CC chemokines at sites of inflammation [15, 16] and thus plays an important role in the resolution of pathological inflammatory conditions [38]. To further investigate whether overexpression of inflammatory CC chemokine decoy receptor D6 on beta cells would impede the migration of pathogenic lymphocytes to islets, we evaluated the lymphocyte transfer-mediated induction of diabetes. Our results demonstrated that the protective mechanism in mD6 transgenic mice is through blockage of the migration of pathogenic lymphocytes to islets.

In the M3 transgenic mouse model, in situ production of M3 by islets not only delays the migration of diabetogenic cells into the islets, but also decreases the frequency and/or impairs the effector function of pathogenic lymphocytes in peripheral lymphoid organs. Thus, NOD/SCID recipients with splenocytes transferred from M3 transgenic NOD mice were more resistant to the development of disease than mice receiving splenocytes from non-transgenic littermates [11]. In this study, we also observed a significant decrease in the incidence of diabetes among recipients with splenocytes transferred from mD6 transgenic mice. Interestingly, around 65% of recipients eventually developed diabetes, suggesting that the biological effect mediated by M3, a wide spectrum immune cell migration inhibitor, is stronger than that of the murine endogenous chemokine decoy receptor D6. However, this high selectivity of D6 in immune cell migration blockage might provide a safety advantage in future clinical applications.

By evaluating the activation status of lymphocytes, we found that the percentage of activated CD4+ T cells was reduced in PLN and that of activated CD8+ T cells was reduced in PLN and pancreas. This suggested that reduction of lymphocytes mediated by D6 overproduction in the pancreas might subsequently influence the activation status of lymphocytes in PLN. Furthermore, in adoptive transfer experiments, we observed a significant decrease in disease development, in terms of onset and incidence, in recipients with splenocytes transferred from mD6 transgenic mice, suggesting that D6 overexpression in situ somehow affects the generation and/or effector function of autoreactive lymphocytes in peripheral lymphoid organs. To further address whether the reduction in pathogenicity of splenocytes from mD6 transgenic donors was beta cell-specific, we performed in vitro proliferation assays by stimulating splenocytes with anti-CD3 antibody, concanavalin A or islet-specific antigens. The proliferation capacity of splenocytes was reduced under stimulation with islet antigens, suggesting that the D6-mediated protective effect in transgenic mice was islet antigen-specific. These results further imply a possible critical role of inflammatory CC chemokines in the generation and/or activation of diabetogenic lymphocytes in NOD mice.

In conclusion, our study demonstrated that the transgenic expression of inflammatory CC chemokines decoy receptor D6 by pancreatic islets prevented autoimmune diabetes by delaying the migration and decreasing the activation of lymphocytes in the islets of NOD mice. These results suggest that inflammatory CC chemokines might promote the migration of lymphocytes to the pancreas and influence the activation of autoreactive lymphocytes in NOD mice. This study also provides a theoretical basis for D6-based genetic manipulation in islets as a potential therapeutic strategy for the prevention and/or treatment of autoimmune diabetes.

Acknowledgements

This work was supported by the National Science Council, Taiwan, Republic of China (NSC-96-2628-B-016-002-MY3, NSC98-3112-B-016-002 and NSC99-3112-B-016-001 to H.-K. Sytwu) and by a research grant from Tri-Service General Hospital, Taiwan, Republic of China (TSGH-C98-12-S01), as well as, in part, by the C.Y. Foundation for Advancement of Education, Sciences and Medicine.

Duality of interest

The authors declare that there is no duality of interest associated with this manuscript.

Supplementary material

125_2011_2166_MOESM1_ESM.pdf (12 kb)
ESM Table 1Primers used in real-time RT-PCR (PDF 12 kb)
125_2011_2166_MOESM2_ESM.pdf (361 kb)
ESM Fig. 1The expression profiles of chemokine genes as indicated in the islets of Balb/c and NOD mice. Islets were isolated from 12-week-old Balb/c mice and NOD mice of different ages as labelled. Total RNA was extracted from islets by TRIZOL and reverse-transcribed with oligo-dT by SuperScript III Reverse Transcriptase. Expression of chemokines was measured by real-time RT-PCR, which was performed by using iQ SYBR Green PCR supermix in an iCycler. Data are expressed as the mean ± SEM. Ifn-γ, also known as Ifng; Il-4, also known as Il4 (PDF 360 kb)
125_2011_2166_MOESM3_ESM.pdf (209 kb)
ESM Fig. 2D6 overexpression by islets specifically decreased the amount of CCL2 and CCL5 in the islets of NOD mice. The amount of CCL2, CCL5 and CXCL10 was detected by immunofluorescent assay. Paraffin-embedded sections of pancreas tissue were rehydrated and heated in 0.1 mol/l citrate buffer. Sections were probed with FITC-conjugated anti-CCL2, -CCL5 or -CXCL10 antibodies, and a Cy5-conjugated anti-GLUT2 antibody was used to identify the presence of islets (blue). Propidium iodide (PI, 2 μg/ml) was used for nuclear counterstain (red spots). Images were captured on Zeiss confocal LSM510 microscope (PDF 208 kb)

Copyright information

© Springer-Verlag 2011