Expression of Pancreatic Islet MHC Class I,Insulin,and ICA 512 Tyrosine Phosphatase in Low-dose Streptozotocin-induced Diabetes in Mice

Abstract

Activated immune cells contribute to the development of diabetes mellitus in multiple low-dose streptozotocin-treated mice. However, a role in the process for MHC Class I restricted T-cells remains a matter of debate. In this study, we examined by confocal microscopy the pancreatic expression of MHC Class I protein, insulin, and ICA 512 protein tyrosine phosphatase in C57BL/Ks mice given 40 mg/kg bw streptozotocin IP on 5 consecutive days. All animals were hyperglycemic from Day 7 and onwards. A loss of ICA 512 from the central portions of the islets was noted on Day 3. On Day 7, an increase in MHC Class I expression, confined primarily to immune cells in the exocrine pancreas and the periinsular areas, was detected. Later, several MHC class I/glucagon and some MHC class I/insulin double-positive cells were found. The insulitis was maximal on Day 14 and declined thereafter. The induction of MHC Class I expression in endocrine cells, occuring only after the cellular infiltration and when the animals were diabetic, indicates that the immune component of the disease does not depend on MHC Class I-restricted cytotoxic T-cells but rather comprises a non-antigen-specific process.

Keywords

diabetes ICA 512 MHC class I insulin pancreatic islets streptozotocin

Streptozotocin given at a high dose causes massive β-cell destruction and permanent hyperglycemia in various experimental animals (Weiss 1982). When given in multiple low doses to susceptible mice it causes delayed-onset diabetes through the combined actions of direct β-cell damage and immunological injury (Like and Rossini 1976; Kolb-Bachofen et al. 1988). A pathogenetic role of immune cells in this model of insulin-dependent Type 1 diabetes is demonstrated by the ability of immunosuppression to prevent the development of disease (Kolb 1987; Hatamori et al. 1990). Macrophages are of prime importance (Oschilewski et al. 1986; Ihm et al. 1990; Andrade et al. 1993), and locally secreted lymphokines also appear to contribute (Campbell et al. 1988; Lukic et al. 1998), but a role for antigen-specific cytotoxic lymphocytes (McEvoy et al. 1984; Oschilewski et al. 1986) has remained controversial.

In human autoimmune diabetes, exogenous factors, e.g., viruses, toxic substances, are believed to trigger an immunological attack on β-cells (Mehta and Palmer 1996). Islets of newly diagnosed patients contain INF-α (Huang et al. 1995) and demonstrate hyperexpression of MHC Class I (Foulis 1996). Antigen-specific cytotoxic T-cells are considered responsible for the destruction of β-cells, and the presence of CD8-positive T-cells in islets from autopsy specimens (Hanninen et al. 1992; Itoh et al. 1993) supports such a view. A role of antigen-specific lymphocytes is underlined by the reports of transfer of disease after bone marrow transplantation (Lampeter et al. 1993; Vialettes et al. 1993) and the recurrence of disease in transplants obtained from identical twins or HLA-identical siblings (Sutherland et al. 1989). It is not known if the phenotype of the cells transferring the disease is that of CD4 or CD8 T-cells or B-lymphocytes.

In this study we examined the islet expression of MHC Class I, insulin, and ICA 512/IA-2 protein tyrosine phosphatase, key proteins of a putative antigen-specific process in the development of multiple low-dose streptozotocin-induced diabetes mellitus.

Materials and Methods

Animal Treatment

Inbred 12-16-week-old C57BL/Ks male mice (Biomedical Centre; Uppsala, Sweden), originally obtained from the Jackson Laboratory (Bar Harbor, ME) were used. The animals had free access to tapwater and pelleted food (R34; AnaLyzen, Lindköping, Sweden). The mice were treated either with IP injections of saline (0.2 ml) for 5 days or with IP injections of streptozotocin (40 mg/kg bw) dissolved in saline for 5 consecutive days. Some animals were sacrified by cervical dislocation before any injections (Day 0).

Blood glucose determinations (ExacTech blood glucose meter; Baxter Travenol, Deerfield, IL) were performed on blood samples taken from the tail tip on Day 0 before any injection, and on Days 3, 7, 10, 14, 21, and 28 after the first injection. On these days groups of mice (n=5) were sacrificed for morphological examination of the pancreatic glands.

Morphological Examinations

After sacrifice, the pancreata were removed and one half of each gland was fixed in 10% formalin solution and the other half rapidly frozen in liquid nitrogen for confocal microscopy. For light microscopy, the fixed glands were embedded in paraffin and sections 5-μm thick were cut and stained using the PAP technique to demonstrate β-cells (guinea pig anti-porcine insulin, 1:100; DAKO, Carpinteria, CA). The sections were counterstained with Mayer's hematoxylin for observation of mononuclear cell infiltration. Frozen tissues were cut in 5-μm sections and kept at −70C before immunostaining.

The immunofluorescence stainings were performed on cryosections for MHC Class I (purified mouse anti-mouse H-2K^d monoclonal antibody, 1:80; Pharmingen, San Diego, CA) and ICA 512 (rabbit antiserum (89–59) against the cytoplasmic domain of the human ICA 512 protein, diluted 1:800; a generous gift from Bayer, Elkhart, IN).

To investigate the expression of MHC Class I in islet cells, double immunofluorescence stainings were used separately for MHC Class I/insulin, MHC Class I/glucagon (rabbit anti-porcine glucagon serum, 1:200; DAKO), and MHC Class I/ICA 512. Sections were fixed in acetone for 7 min and incubated with mixtures of primary antibodies at 4C overnight, then incubated with the biotin-avidin-Texas Red color system for MHC Class I and FITC-conjugated secondary antibodies for other reagents. Confocal microscopy (Zeiss LSM 410 invert Laser Scan Microscope; Carl Zeiss Jena, Jena, Germany) was used with excitations of 488 nm and 543-nm wavelength laser.

The average intensity of ICA 512 staining per islet area was evaluated by confocal microscopy at different time points after saline or streptozotocin injections. This was performed by measurements of >2000 μm² islet area selected from >20 islets/pancreas. For each pancreas three to five sections (>50-μm apart) were examined. To standardize the parameters for the digitized images, brightness and contrast were set for normal islets and then used for all other experimental islets during that session on the confocal microscope.

To control the specificities of the immunostaining procedures, the primary antibodies were replaced by normal serum from the appropriate species or by PBS.

Results

Blood Glucose and Insulitis

The mice treated with multiple low doses of streptozotocin (STZ) had normal blood glucose levels on Day 3, but from Day 7 the animals displayed hyperglycemia (Figure 1). Infiltration into the pancreas of lymphocyte-like cells was first seen on Day 7, at which time many mononuclear cells were noted in the exocrine tissue (Figure 2b). On Day 10 and onwards, the cells were homed to the peripheries of the islets and the number of infiltrating cells in the exocrine tissue diminished. The islet infiltration was most pronounced on Day 14 (Figure 2d). Occasionally, islets with no lymphocytes were also found at this time point (not shown).

Figure 1

Blood glucose concentrations in male C57BL/Ks mice after treatment with five daily IP injections of STZ (40 mg/kg bw) (closed circles) or saline (0.2 ml) (open circles). Values are means ± SEM for the number of mice given in parentheses) (Day 0, 28 mice; STZ Day 3, 17 mice, Day 7, 18 mice, Day 10, 21 mice, Day 14, 10 mice, Day 21, 11 mice, Day 28, 5 mice; Saline: Day 7, 6 mice, Day 10, 15 mice, Day 14, 11 mice, Day 21, 8 mice, Day 28, 2 mice). ^∗∗∗, p<0.001 vs the saline-treated mice, unpaired Student's t-test.

MHC Class I

In the normal pancreas (Day 0), MHC Class I expression was low and was found only in a few dendritic-like cells in the exocrine tissue and in the islets (Figure 3a). In one of the five normal pancreata examined, MHC Class I-positive dendrite-like and capillary cells were abundant (Figure 3b), an unexplained observation. In none of the five normal tissues was MHC Class I staining of endocrine cells detected. With STZ treatment, an increase in pancreatic MHC Class I protein was observed on Day 7 (Figure 4b). At this time point, the staining was associated with lymphocytelike cells in the exocrine pancreas and in the peri-insular regions (Figures 4b, 5c, and 5d). The MHC Class I expression then became more intense and was concentrated to the islets, also involving endocrine cells (Figures 4c-4f and 5e-5g). On Day 14, both insulin- (Figure 5f) and glucagon-containing cells (not shown) stained positive for MHC Class I, after which the staining in the islet areas gradually weakened. On Day 28, some MHC Class I staining associated with lymphocyte-like and dendritic cells (Figures 4f and 5h) remained.

ICA 512, Insulin, and Glucagon

ICA 512 was found as an even staining of all endocrine cells before STZ treatment (not shown). The islet ICA 512 content, as quantified by confocal microscopy, was reduced by 36% on Day 3 (control group 124 ± 12; STZ 80 ± 13; p<0.05, unpaired Student's t-test) and by 67% on Day 7 (STZ 41 ± 6; p<0.001). In particular, the ICA 512 disappeared from the central portions of the islets. Endocrine cells positive for both ICA 512 and MHC Class I were not detected on Day 7 but were present from Day 10 onwards. On Day 21, several strongly ICA 512 and MHC Class I-positive cells were observed (Figure 6).

Figure 3

MHC Class I (H2K^d) immunofluorescence staining of normal mouse islets. Positive staining of few dendrite-like cells but not of endocrine islet cells (a). Abundant positive cells in one of the five pancreata examined. Both dendrite-like cells and capillary cells are stained (b).

Figure 2

Infiltration of lymphocytes into pancreatic tissue of mice treated with multiple low doses of STZ. Tissues were obtained on Day 7 (b,c) and Day 14 (d); normal pancreas (a). Note the diffuse infiltration in exocrine tissue on Day 7 (b) and the accumulation of infiltrating cells around the islet area on Day 14 (d). Immunostaining for insulin and counterstaining with Mayer's hematoxylin.

Insulin reactivity was reduced on Day 7 but was detectable in most of the islets (Figures 5c and 5d). Occasionally, cells staining for both insulin and MHC Class I were found (Figure 5d). On Days 10 and 14, few β-cells could be detected in the islets, and most of them showed double staining for insulin and MHC Class I (Figures 5e and 5f).

Figure 4

MHC Class I immunofluorescence staining in islets of mice treated with multiple low doses of STZ. The islets were examined on Day 3 (a), Day 7 (b), Day 10 (c), Day 14 (d), Day 21 (e), and Day 28 (f).

Figure 5

Insulin (green) and MHC Class I (red) double staining of frozen sections of pancreata obtained on Day 0 (a), Day 3 (b), Day 7 (c,d), Day 10 (e), Day 14 (f), Day 21 (g), and Day 28 (h). Double-positive cells are identified by the yellow color.

Glucagon/ICA 512-containing cells were localized to the islet periphery before treatment. At later time points, several strongly double-positive cells were found localized throughout the islets (not shown). Glucagon-staining cells that were also positive for MHC Class I were found on Days 10-21. The number of α-cells increased during the course of the disease (Li et al. in press), a phenomenon that reflected lack of intraislet insulin (Unger and Orci 1990) and which was similar to that noted in convertase knockout mice (Furuta et al. 1997) and in glut-2 knockout mice (Guillam et al. 1997).

Figure 6

ICA 512 (green) and MHC Class I (red) double staining of frozen sections of pancreata obtained on Day 7 (a-c), Day 14 (d-f), and Day 21 (g-i). Double-positive cells are identified by the yellow color.

Discussion

In this study we observed that the multiple low-dose STZ treatment caused β-cell loss and hyperglycemia, with the occurrence of lymphocyte-like cells in the exocrine pancreas and the peri-insular areas before the appearance of MHC Class I in endocrine cells. Because cytotoxic T-cells recognize antigens in the context of MHC Class I protein complexes on the target cell surfaces (Zinkernagel and Doherty 1997) in a dose-dependent fashion (Cai et al. 1998), the sequence of the events argues against a pathogenic role of MHC Class I restricted T-cells in promoting the development of diabetes in this experimental model.

Macrophages are of key importance for the occurrence of diabetes in multiple low-dose STZ-treated mice. The toxic destruction of β-cells triggers recruitment of macrophages, single-cell insulitis (Kolb-Bachofen et al. 1988), and production of monokines such as IL-1 and TNF-α, which have a cytopathic action on the islet cells (Mandrup-Poulsen et al. 1990). Administration of silica, a macrophage toxin, prevents the development of disease (Oschilewski et al. 1986; Ihm et al. 1990). Macrophages stimulate T-helper cells to release IFN-γ, the cytokine most likely responsible for the induction of MHC Class I expression in the endocrine cells (Cockfield et al. 1989). Administration of IFN-γ has been reported to induce islet cell MHC antigens and to enhance STZ-induced diabetes in the CBA mouse (Campbell et al. 1988). The authors noted that the MHC Class I (H-2K) protein expression was increased on islet cells from mice given multiple low doses of STZ alone. The increase in islet MHC Class I was detected on Day 11, well in agreement with our findings. Lack of MHC Class I staining of normal islet endocrine cells, as we observed, was also described in the study by Cockfield and co-workers (1989), who found that multiple low-dose STZ triggered expression of MHC Class I in renal tubule cells, Kupffer cells, some hepatocytes, and occasional cardiac myocytes, in pancreatic ductal epithelia, and in cells within the pancreatic islets. The latter were believed to represent either marrow-derived cells or endocrine islet cells. Double staining and confocal microscopy to positively identify hormone-secreting cells were not employed in these two earlier studies.

The ICA 512 protein tyrosine phosphatase, an intrinsic membrane protein of secretory granules of peptide-secreting endocrine cells (Solimena et al. 1996), was reduced by a third on Day 3, suggesting that β-cell granules are highly susceptible to toxic damage. Autoantibodies against ICA 512 often appear in association with the onset of human autoimmune diabetes (Genovese et al. 1992; Naserke et al. 1998). This might be due to destruction by toxic agents of β-cells and release of granules.

The presence of some MHC Class I/insulin double-positive cells on Day 14, when the lymphocytic infiltration of the islets was most pronounced, argues against the existence of insulin-specific cytotoxic T-cells in the infiltrate. Likewise, no ICA 512-specific cytotoxic T-cells appear to be induced, because many ICA 512/MHC Class I-positive cells (including ICA 512/glucagon-positive cells) survive. The selective β-cell damage rather may be due to a direct toxic effect of STZ and to the effects of cytokines released from macrophages and activated T-helper cells. The disease appears to represent a form of immunological, non-antigen-specific diabetes mellitus. Absence of antigen-specific response is also supported by the survival of syngenic pancreatic islets transplanted to multiple low-dose STZ diabetic mice (Sandler and Andersson 1981). Moreover, studies demonstrating the presence of autoantibodies in this animal model are lacking.

In human autoimmune diabetes, insulin, ICA 512, and glutamate decarboxylase are important autoantigens and, according to common belief, cytotoxic T-cells reacting with peptides of these proteins in the context of MHC Class I are pathogenic. However, there is no direct evidence for the occurrence of such cytotoxic T-cells in the human disease (Wegmann 1996). Antigen specificity in human Type 1 diabetes might be confined to the T-helper cells and B-cell-driven autoantibody production, as reflected by the linkage to the MHC Class II rather than the MHC Class I genetic region. It may be that memory CD4 cells are of importance for the reappearance of disease after bone marrow transplantation and syngenic pancreatic grafts (Sutherland et al. 1989; Lampeter et al. 1993; Vialettes et al. 1993). Hypothetically, the disease in some patients with Type 1 diabetes without autoantibodies might have a pathogenesis with similarities to that of the STZ-treated mouse. Recent studies indicate that endogenous insulin production remains at a higher level in autoantibody-negative compared to autoantibody-positive individuals with new-onset disease (Bonfanti et al. 1998; Sabbah et al. 1999). In such individuals, regimens aimed at increasing the β-cell defense against toxic damage might be of particular value to preserve residual β-cell mass. We have noted that potassium channel openers, which inhibit insulin secretion, increase the β-cell resistance to STZ damage in vitro (Kullin et al. 1998) and, in a clinical trial with supplementary treatment with diazoxide to induce β-cell rest, beneficial effects were noted (Björk et al. 1996). The multiple low-dose streptozotocin-induced diabetes mellitus model offers a tool for further exploration of regimens to enhance β-cell resistance and repair.

Footnotes

Acknowledgements

Supported by grants from the Swedish Medical Research Council, the Novo Nordic Fund, the Swedish Diabetes Association, the Söderberg Foundation, the Family Ernfors Fund, and the Juvenile Diabetes Foundation-Wallenberg Fund.

We thank Margareta Ericson and Astrid Nordin for excellent technical assistance.

References

Andrade

Conde

Sobrino

Bedoya

(1993) Activation of peritoneal macrophages during the prediabetic phase in low-dose streptozotocin-treated mice. FEBS Lett 327:32–34

Björk

Berne

Kämpe

Wibell

Oskarsson

Karlsson

(1996) Diazoxide treatment at onset preserves residual insulin secretion in adults with autoimmune diabetes. Diabetes 45:1427–1430

Bonfanti

Bazzigaluppi

Calori

Riva

Viscardi

Bognetti

Meschi

Bosi

Chiumell

Bonifacio

(1998) Parameters associated with residual insulin secretion during the first year of disease in children and adolescents with Type 1 diabetes mellitus. Diabet Med 15:844–850

Cai

Brunmark

Luxembourg

Garcia

Degano

Teyton

Wilson

Peterson

Sprent

Jackson

(1998) Probing the activation requirements for naive CD8⁺ T cells with Drosophila cell transfectants as antigen presenting cells. Immunol Rev 165:249–265

Campbell

Oxbrow

Koulmanda

Harrison

(1988) IFN-gamma induces islet cell MHC antigen and enhances autoimmune, streptozotocin-induced diabetes in the mouse. J Immunol 140:1111–1116

Cockfield

Ramassar

Urmson

Halloran

(1989) Multiple low dose streptozotocin induces systemic MHC expression in mice by triggering T cells to release IFN-gamma. J Immunol 142:1120–1128

Foulis

(1996) The pathology of the endocrine pancreas in type 1 (insulin-dependent) diabetes mellitus. APMIS 104:161–167

Furuta

Yano

Zhou

Rouille

Holst

Carroll

Ravazzola

Orci

Furuta

Steiner

(1997) Defective prohormone processing and altered pancreatic islet morphology in mice lacking active SPC2. Proc Natl Acad Sci USA 94:6646–6651

Genovese

Bonifacio

McNally

Dean

Wagner

Bosi

Gale

Bottazzo

(1992) Distinct cytoplasmic islet cell antibodies with different risks for type 1 (insulin-dependent) diabetes mellitus. Diabetologia 35:385–388

10.

Guillam

Hummler

Schaerer

Yeh

Birnbaum

Schmidt

Deriaz

Thorens

(1997) Early diabetes and abnormal postnatal pancreatic islet development in mice lacking Glut-2. Nature Genet 17:327–330

11.

Hanninen

Jalkanen

Salmi

Toikkanen

Nikolakaros

Simell

(1992) Macrophages, T cell receptor usage, and endothelial cell activation in the pancreas at the onset of insulin-dependent diabetes mellitus. J Clin Invest 90:1901–1910

12.

Hatamori

Yokono

Hayakawa

Taki

Ogawa

Nagata

Hari

Shii

Taniguchi

Baba

(1990) Anti-interleukin 2 receptor antibody attenuates low-dose streptozotocin-induced diabetes in mice. Diabetologia 33:266–271

13.

Huang

Yuan

Goddard

Foulis

James

Lernmark

Pujol-Borell

Rabinovitch

Somoza

Stewart

(1995) Interferon expression in the pancreases of patients with type I diabetes. Diabetes 44:658–664

14.

Ihm

Lee

Rhee

Min

(1990) Initial role of macrophage in the development of anti-beta-cell cellular autoimmunity in multiple low-dose streptozotocin-induced diabetes in mice. Diabetes Res Clin Pract 10:123–126

15.

Itoh

Hanafusa

Miyazaki

Miyagawa

Yamagata

Yamamoto

Waguri

Imagawa

Tamura

Inada

Kawata

Tarui

Kono

Matsuzawa

(1993) Mononuclear cell infiltration and its relation to the expression of major histocompatibility complex antigens and adhesion molecules in pancreas biopsy specimens from newly diagnosed insulin-dependent diabetes mellitus patients. J Clin Invest 92:2313–2322

16.

Kolb

(1987) Mouse models of insulin dependent diabetes: low-dose streptozotocin-induced diabetes and nonobese diabetic (NOD) mice. Diabetes Metab Rev 3:751–778

17.

Kolb-Bachofen

Epstein

Kiesel

Kolb

(1988) Low-dose streptozocin-induced diabetes in mice. Electron microscopy reveals single-cell insulitis before diabetes onset. Diabetes 37:21–27

18.

Kullin

Björk

Sandler

Karlsson

(1998) Diazoxide modifies streptozotocin induced beta-cell damage. Diabetologia 41(suppl 1):A160

19.

Lampeter

Homberg

Quabeck

Schaefer

Wernet

Bertrams

Grosse-Wilde

Gries

Kolb

(1993) Transfer of insulin-dependent diabetes between HLA-identical siblings by bone marrow transplantation. Lancet 341:1243–1244

20.

Karlsson

Sandler

(in press) Alpha-cell expansion in Type 1 diabetes induced by low-dose streptozotocin in mice. J Endocrinol

21.

Rossini

(1976) Streptozotocin-induced pancreatic insulitis: new model of diabetes mellitus. Science 193:415–417

22.

Lukic

Stosic-Grujicic

Shahin

(1998) Effector mechanisms in low-dose streptozotocin-induced diabetes. Dev Immunol 6:119–128

23.

Mandrup-Poulsen

Helqvist

Wogensen

Mlvig

Pociot

Johannesen

Nerup

(1990) Cytokines and free radicals as effector molecule in the destruction of pancreatic beta cells. Curr Top Microbiol Immunol 164:169–193

24.

McEvoy

Andersson

Sandler

Hellerström

(1984) Multiple low-dose streptozotocin-induced diabetes in the mouse. Evidence for stimulation of a cytotoxic cellular immune response against an insulin-producing beta cell line. J Clin Invest 74:715–722

25.

Mehta

Palmer

(1996) The natural history of the IDDM process. In Palmer

Prediction, Prevention and Genetic Counseling in IDDM. Chichester, UK, John Wiley & Sons, 3–16

26.

Naserke

Ziegler

Lampasona

Bonifacio

(1998) Early development and spreading of autoantibodies to epitopes of IA-2 and their association with progression to type 1 diabetes. J Immunol 161:6963–6969

27.

Oschilewski

Schwab

Kiesel

Opitz

Stunkel

Kolb-Bachofen

Kolb

(1986) Administration of silica or monoclonal antibody to Thy-1 prevents low-dose streptozotocin-induced diabetes in mice. Immunol Lett 12:289–294

28.

Sabbah

Savola

Kulmala

Veijola

Vahsalo

Karjalainen

Akerbom

Knip

(1999) The Childhood Diabetes in Finland Study Group. Diabetes-associated autoantibodies in relation to clinical characteristics and natural course in children with newly diagnosed type 1 diabetes. J Clin Endocrinol Metab 84:1534–1539

29.

Sandler

Andersson

(1981) Islet implantation into diabetic mice with pancreatic islets. Acta Pathol Microbiol Scand [A] 89:107–112

30.

Solimena

Dirkx

Jr Hermel

Pleasic-Williams

Shapiro

Caron

Rabin

(1996) ICA 512, an autoantigen of type 1 diabetes, is an intrinsic membrane protein of neurosecretory granules. EMBO J 15:2102–2114

31.

Sutherland

DER

Goetz

Sibley

(1989) Recurrence of disease in pancreas transplants. Diabetes 38:85–87

32.

Unger

Orci

(1990) Glucagon secretion in diabetes mellitus. In Rifkin

Porte

Jr Diabetes Mellitus: Theory and Practice. 4th ed. New York, Elsevier Science, 118–120

33.

Vialettes

Maraninchi

San Marco

Birg

Stoppa

Mattei-Zvaco

Thivolet

Hermitte

Vague

Mercier

(1993) Autoimmune polyendocrine failure-type (insulin-dependent) diabetes mellitus and hypothyroidism-after allogeneic bone marrow transplantation in a patient with lymphoblastic leukemia. Diabetologia 36:541–546

34.

Wegmann

(1996) The immune response to islets in experimental diabetes and insulin-dependent diabetes. Curr Opin Immunol 8:860–864

35.

Weiss

(1982) Streptozotocin: a review of its pharmacology, efficacy, and toxicity. Cancer Treat Rep 66:427–438

36.

Zinkernagel

Doherty

(1997) The discovery of MHC restriction. Immunol Today 18:14–17