| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Endocrinology (F.D., M.P., S.D.) and Experimental Medicine (L.L.) University of Rome "La Sapienza," Rome; the Department of Internal Medicine, University of Messina (M.P., D.C.), Messina; and the Department of Experimental and Clinical Medicine, University of Reggio Calabria (U.D.M.), Catanzaro, Italy; and Laboratoires de Recherche Louis Jeantet, University of Geneva (M.N.-A., P.A.H.), Geneva, Switzerland
Address all correspondence and requests for reprints to: Dr. Francesco Dotta, c/o Diabetes, Endocrinology Metabolism Foundation, Largo Marchiafava 1, 00161, Rome, Italy.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Insulin-dependent diabetes mellitus (IDDM) is a chronic autoimmune disease caused by the specific destruction of pancreatic islet ß-cells by the immune system in genetically predisposed individuals (2). In this disease, autoantibodies against islet antigens can be detected in the circulation well before the appearance of clinical symptoms, representing important tools for the early detection of subjects with increased risk of developing IDDM (3). Among diabetes-associated autoantigens identified to date, insulin, glutamic acid decarboxylase, the IA-2/ICA512 islet tyrosine phosphatase, carboxypeptidase H, a 69-kDa islet cell protein (ICA69) and the GM2-1 islet ganglioside have been shown to be the targets of autoantibodies associated with diabetes development. Cytoplasmic islet cell autoantibodies (ICA) are a heterogeneous population of autoantibodies (4, 5) reacting on pancreatic frozen sections. Two subsets of ICA have been identified to date: 1) restricted ICA reacting only with ß-cells of human and rat pancreatic islets and directed against glutamic acid decarboxylase (6); and 2) nonrestricted ICA (NR-ICA), reacting with whole human, rat, and mouse pancreatic islets. One of the target antigens of NR-ICA has been suggested (7) to be the GM2-1 monosialo-ganglioside, as this molecule can inhibit the binding of some NR-ICA-positive sera on pancreatic frozen sections. GM2-1 is an islet-specific ganglioside within the pancreas of man (8), rat (9), and mouse (10). This glycolipid is hyperexpressed (10) in islets of an animal model of ß-cell autoimmunity, the nonobese diabetic (NOD) mouse.
In addition, we have recently sequenced (11) the GM2-1 ganglioside, finding it to have a novel glycolipid structure (N-acetyl-neuramic acid-galactose-galactosamine-galactosamine-glucose-ceramide) and to be the target of IgG autoantibodies that are strongly correlated with progression to diabetes in ICA+ relatives of type 1 diabetic subjects (12). Interestingly, the expression of GM2-1, similarly to that of cytoplasmic ICA autoantigens, has been found (13) to be metabolically regulatable and to be absent in suppressed islet cells that do not contain secretory granules.
After this observation and for the potential importance of this ganglioside antigen in diabetes autoimmunity, we wished to establish 1) whether GM2-1 is ß-cell specific and 2) its intracellular localization.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Preparation of subcellular fractions
Subcellular fractionation was adapted from the method of Ullrich
and Wollheim (15) on cultured cells from a well granulated ß-cell
line (INS) derived from a rat insulinoma (16). Approximately 30–50
million cells were washed 3 times with ice-cold homogenization medium
(HM; 250 mmol/liter sucrose, 5 mmol/liter HEPES, 0.5 mmol/liter EGTA,
and 0.5 mmol/liter phenylmethylsulfonylfluoride, adjusted to pH 7.4
with KOH) and scraped into 3–5 ml HM. The suspension of scraped cells
was pressurized under nitrogen (350 psi) in a Parr bomb for 15 min at 4
C. After centrifugation of the homogenate at 700 x g for 10
min, the resulting postnuclear supernatant was diluted to 5 ml with HM
and mixed with 1 ml 90% isoosmotic Percoll solution (in HM) to give 6
ml of a 15% Percoll solution. A self-generating gradient was formed by
spinning the sample at 45,000 x g for 25 min at 4 C in
a fixed angle rotor (Heraeus Sepatech HFA 22.50, Hanau, Germany) in a
Heraeus Suprafuge 22. Six fractions (1 ml) were collected and numbered
1–6 from top to bottom. Opaque bands were visible in fraction 1
(corresponding to plasma membranes) and fraction 6 (corresponding to
secretory granules). Each fraction was washed twice with 4 vol HM and
centrifuged at 150,000 x g at 4 C for 30 min. The
final pellets were stored at -80 C before ganglioside extraction or
marker assays. The plasma membrane marker
Na+K+-adenosine triphosphatase
(Na+,K+-ATPase; 
1-subunit) was
determined by Western blotting using a mouse monoclonal antibody MCK1
obtained from Dr. Sweadner (Boston, MA) (17), diluted 1:500 in
Tris-buffered saline (TBS: 20 mmol/liter Tris and 137 mmol/liter NaCl,
adjusted to pH 7.6 with HCl) with 5% low fat milk and 0.1% Tween.
Detection was performed using a second antibody coupled to horseradish
peroxidase and the Amersham ECL method (Amersham, Aylesbury, UK).
Immunoreactive bands (97 kDa) were quantified by using a flatbed
scanner from MacIntosh and the Image 1.33 program (MacIntosh, Apple
Computers, Cupertino, CA). Immunoreactive insulin was determined using
a standard RIA protocol (18) and was used as a marker for secretory
granules.
Ganglioside extraction
Gangliosides were isolated according to the method of
Svennerholm et al. (19) with minor modifications. Briefly,
tissues were extracted twice in chloroform-methanol-water and subjected
to Folch partition by addition of water to give a final
chloroform-methanol-water ratio of 1:2:1.4 (vol/vol/vol). The upper
phase was evaporated to dryness in a rotary evaporator (Buchi RE 121,
Buchi, Flawil, Switzerland), and glycolipids were purified of salts and
low mol wt contaminants using Sep-Pak C18 cartridges (Water
Associates, Millford, MA) according to the method of Williams and
McCluer (20). Purified glycolipids were then applied to a
diethylaminoethyl-Sephadex A-25 (Pharmacia, Uppsala, Sweden) anion
exchange column in acetate form. Bound charged glycolipids were eluted
using methanol containing 1 M sodium acetate and desalted
by Sep-Pak C18 cartridge chromatography. Acidic glycolipids
were finally analyzed by high performance TLC (HPTLC).
Gangliosides in the present work were defined according to the nomenclature of Svennerholm (21), in which GM indicates monosialo-gangliosides, GD indicates disialo-gangliosides, and GT indicates trisialo-gangliosides.
HPTLC
Acidic glycolipids were analyzed by HPTLC using analytical
precoated Silica gel 60 HPTLC plates (Merck, Darmstadt, Germany). All
plates were first activated by heating to 100 C for 30 min. Samples
were spotted onto plates with a Hamilton syringe (Hamilton, Reno, NV)
in chloroform-methanol-0.25% aqueous KCl (5:4:1, vol/vol/vol).
Ganglioside standards GM3, GM1, GD1a, GD1b, GT1b (Sigma Chemical Co.,
St. Louis, MO), GM2, and GD3 (gift from Fidia Research Laboratories,
Abano Terme, Italy) purified from bovine brain were included in every
HPTLC plate. Plates were air-dried and stained with resorcinol spray
reagent. Resorcinol stains sialic acid-containing glycolipids
(gangliosides) blue-black (22).
Gangliosides on HPTLC plates were quantified by scanning densitometric analysis of resorcinol-stained bands (model GS300, Hoefer Scientific Instruments, San Francisco, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) of
gangliosides extracted from rat islet cells showed an identical pattern
of ganglioside expression in ß- and in non-ß-cells. In both
extracts, analyzed by TLC, three resorcinol-stained bands corresponding
to different gangliosides were visualized: a band with a migration
position between GM2 and GM1 standards (GM2-1), a band comigrating with
the GM3 standard, and another comigrating with the GD3 standard.
Moreover, a quantitative analysis of ganglioside expression in ß- as
well as in non-ß-cells showed no difference between the two
populations; the GM2-1 ganglioside was present at a percentage of 40%
of the total ganglioside content in both cell populations, whereas GM3
and GD3 represented 50% and 10%, respectively.
|
). The GM2-1 band was present mainly in
the fraction enriched in secretory granules (in this fraction the
presence of immunoreactive insulin has also been demonstrated) and to a
lesser extent in plasma membranes (in this fraction the presence of
Na+K+-ATPase 1-subunit has been
revealed by Western blot analysis; Fig. 2).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-, 
-, and PP cells. It thus remains unclear from
the present study whether ganglioside distribution among these three
non-ß-cell types is equal. To obtain sufficient material for
analyzing ganglioside distribution in subcellular fractions, it was
necessary to use a transformed ß-cell line. Of the many rat lines
available, the INS line (16) is the best differentiated. In particular,
this line has been shown to be well granulated, containing up to 20%
the normal insulin content of a primary ß-cell. Insulin release from
INS cells is, furthermore, sensitive to glucose in the physiological
range. These cells thus represent a useful model for studying ß-cell
function. In keeping with their transformed state, however, these cells
are not identical with their primary counterpart. Indeed, we have shown
aberrant intracellular C peptide degradation and slow proinsulin
conversion in these cells compared with ß-cells (23, 24). There is,
however, no reason to suppose that intracellular ganglioside
distribution should be different in INS and primary ß-cells. In other
cell types, gangliosides are expressed predominantly in the plasma
membrane after synthesis in the Golgi complex (25). They are degraded
in lysosomes, and fragments arising from such degradation are known to
recycle. Interestingly, gangliosides thus share with granule proteins
at least some common intracellular vesicular compartments (26),
especially when one considers the cross-talk within and between the
secretory and endocytotic pathways (27, 28). Granulogenesis is thought
to involve specific targeting from the trans-Golgi network
to nascent, immature granules (26, 29, 30). How GM2-1 becomes targeted
to granules remains as mysterious as the mechanism responsible for
targeting proinsulin itself (26). The presence of GM2-1 ganglioside in granules is in keeping with its absence in metabolically suppressed islet cells, which are largely degranulated. To this end, it is important to note the results of those studies, which showed that metabolically suppressed islets in animal models of ß-cell autoimmunity such as the BB rat and the NOD mouse are less susceptible to autoimmune destruction. These suppressed islets are largely degranulated, do not bind to cytoplasmic ICA and do not express the GM2-1 ganglioside or, very likely, other secretory granule-associated target autoantigens of the autoimmune response in IDDM, such as insulin (31), carboxypeptidase H (32), 38-kDa protein (33), IA-2 islet tyrosine phosphatase and its related protein phogrin (34, 35), and the recently identified Glima-38 antigen (36).
These data suggest that the loss of immune tolerance toward the components of the insulin secretory granule may be central to the immunopathogenesis of ß-cell destruction; as a consequence, it is possible to speculate that metabolically suppressed islets, by not containing secretory granules, are devoid of the target molecules of the autoimmune response and, therefore, may become resistant to the autoimmune destructive process.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received August 18, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-adrenoreceptors in a plasma-membrane enriched fraction from the
insulin-secreting cell-line RINm5F. Acta Endocrinol (Copenh) 121:525–532This article has been cited by other articles:
 |
|
 |
|
  J. Vikman, X. Ma, G. H Hockerman, P. Rorsman, and L. Eliasson Antibody inhibition of synaptosomal protein of 25 kDa (SNAP-25) and syntaxin 1 reduces rapid exocytosis in insulin-secreting cells. J. Mol. Endocrinol., June 1, 2006; 36(3): 503 - 515. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Copani, D. Melchiorri, A. Caricasole, F. Martini, P. Sale, R. Carnevale, R. Gradini, M. A. Sortino, L. Lenti, R. De Maria, et al. beta -Amyloid-Induced Synthesis of the Ganglioside Gd3 Is a Requisite for Cell Cycle Reactivation and Apoptosis in Neurons J. Neurosci., May 15, 2002; 22(10): 3963 - 3968. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Pilon, X.-R. Peng, A. M. Spence, R. H.A. Plasterk, and H.-M. Dosch The Diabetes Autoantigen ICA69 and Its Caenorhabditis elegans Homologue, ric-19, Are Conserved Regulators of Neuroendocrine Secretion Mol. Biol. Cell, October 1, 2000; 11(10): 3277 - 3288. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
