A potent immunomodulatory compound, (S,R)-3-Phenyl-4,5-dihydro-5-isoxasole acetic acid, prevents spontaneous and accelerated forms of autoimmune diabetes in NOD mice and inhibits the immunoinflammatory diabetes induced by multiple low doses of streptozoto

Article abstract of DOI:10.1124/jpet.106.109272

(S,R)-3-Phenyl-4,5-dihydro-5-isoxasole acetic acid (VGX-1027) is an isoxazole compound that exhibits various immunomodulatory properties. The capacity of VGX-1027 to prevent interleukin (IL)-1β plus interferon-γ-induced pancreatic islet death in vitro pro

Full text of DOI:10.1124/jpet.106.109272

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
JPET F aT sh ti s Fa rot i rc wl e ah ar sd n.o Pt bue eb nl ic so ph y ee dit eo dna nDd ef ocr me ma tt eb d e. Tr h5e ,f i2n 0a l 0v 6e r sai os n Dm Oa yI d: 1i f f0e .r 1f r1o 2m 4t /h j ips ev et r. s1i o0 n 6. .109272
JPET #109272
1
A Potent Immunomodulatory Compound, VGX-1027, Prevents Spontaneous and
Accelerated Forms of Autoimmune Diabetes in NOD Mice and Inhibits the
Immunoinflammatory Diabetes Induced by Multiple Low Doses of Streptozotocin in
CBA/H Mice
Stanislava Stosic-Grujicic, Ivana Cvetkovic, Katia Mangano, Massimo Fresta, Danijela
Maksimovic-Ivanic, Ljubica Harhaji, Dusan Popadic, Miljana Momcilovic, Djordje
Miljkovic, Joseph Kim, Yousef Al Abed and Ferdinando Nicoletti
Institute for Biological Research "Sinisa Stankovic", Belgrade, Serbia (SSG, IC, DMI,
LH, MM, DM)
Department of Biomedical Sciences, University of Catania, Italy (KM, FN)
Department of Pharmacobiological Sciences, University Magna Graecia of Catanzaro,
Italy (MF)
Institute of Microbiology and Immunology, School of Medicine, Belgrade University
Belgrade, Serbia (DP)
VGX Pharmaceuticals, Blue Bell, PA, USA (JK)
Laboratory of Medicinal Chemistry, North Shore Long island Jewish Health System, NY,
USA (YAA)
Copyright 2006 by the American Society for Pharmacology and Experimental Therapeutics.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
2
Running title: VGX-1027 is antidiabetogenic
Author for correspondence:
Prof. Ferdinando Nicoletti, Department of Biomedical Sciences. Via Androne, 83, 95124,
Catania, Italy. E-mail: ferdinic@unict.it; phone: 0039-347-3369125; fax: 0039-095-
3
20267
Number of text pages: 30
Number of tables: 1
Number of figures: 9
Number of references: 40
Number of words in the Abstract: 185
Number of words in the Introduction: 686
Number of words in the Discussion: 1359
List of non-standard abbreviations:
VGX-1027: (S,R)-3-phenyl -4,5-dihydro-5-isoxasole acetic acid
CY: cyclophosphamide
NO: nitric oxide
MLD-STZ: multiple low doses of streptozotocin
PC: peritoneal cells
SMNC: spleen mononuclear cells
T1D: type 1 diabetes

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
3
Abstract
VGX-1027 ((S,R)-3-phenyl -4,5-dihydro-5-isoxasole acetic acid) is an isoxazole
compound which exhibits various immunomodulatory properties. The capacity of VGX-
1
027 to prevent IL-1β+IFN-γ-induced pancreatic islets death in vitro, prompted us to
evaluate its effects on the development of autoimmune diabetes in preclinical models of
human type 1 diabetes mellitus (T1D). Administration of VGX-1027 to NOD mice with
spontaneous or accelerated forms of diabetes induced either by injection of
Cyclophosphamide or by transfer of spleen cells from acutely diabetic syngeneic donors,
markedly reduced the cumulative incidence of diabetes and insulitis. In addition, VGX-
1
027 given either i.p. or per os to CBA/H mice made diabetic with multiple low doses of
streptozotocin (MLD-STZ), successfully counteracted the development of destructive
insulitis and hyperglycemia. The animals receiving VGX-1027 exhibited reduced
production of proinflammatory mediators TNF- , IL-1 , MIF and iNOS-mediated NO
generation in both pancreatic islets and peripheral compartments. These results indicate
that VGX-1027 probably exerts its antidiabetogenic effects by limiting cytokine-
mediated immunoinflammatory events leading to inflammation and destruction of
pancreatic islets. VGX-1027 seems worthy of being considered as a candidate drug in the
development of new therapeutic strategies for the prevention and early treatment of T1D.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
4
Introduction
Type 1 diabetes mellitus (T1D) is a T cell-mediated autoimmune disease that results from
selective destruction of the insulin-producing β cells in the pancreatic islets of
Langerhans. Both in human T1D and in rodent models of the disease such as the mouse
made diabetic with multiple low doses of streptozotocin (MLD-STZ-induced diabetes),
the NOD mouse and the diabetes-prone BB rat, the pancreatic β cells are selectively
destroyed by infiltrating mononuclear cells (reviewed by Kolb, 1997).
One common mechanism by which infiltrating macrophages and autoreactive T
lymphocytes promote β cell death may be related to their capacity of producing
proinflammatory mediators, such as tumor necrosis factor (TNF)- , interferon (IFN)- ,
interleukin (IL)-1β, IL-18 and nitric oxide (NO) which have all been implicated as
critical players in the initiation and propagation of the disease process (reviewed by
Rabinovitch and Suarez-Pinzon, 1998).
Although negating the action of proinflammatory and type 1 cytokines such as IL-
1
, IFN- , IL-12, IL-18 and TNF- may prevent autoimmune diabetogenesis in most of
these models (Rabinovitch and Suarez-Pinzon, 1998), the efficacy of similar approaches
in influencing the natural course of human T1D is not known because of the lack of
clinical studies with specific inhibitors of these cytokines in the human disease
counterpart.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
5
However, in spite of the promising data obtained in these rodent models, there are
several drawbacks limiting the possible application of specific cytokine inhibitors in
human T1D including their potential immunogenicity, the short half-life in vivo, the non-
oral route of administration, the side effects, and the high costs of treatment. For this
reason, studies aimed at discovering orally available small chemical compounds that
target the synthesis and/or the action of endogenous proinflammatory cytokines are
warranted. For example, a variety of small synthetic inhibitors of TNF- have been
designed to specifically inhibit the action of the cytokine at various levels, including
transcriptional and post-transcriptional inhibition of TNF- production, secretion, down-
regulation of TNF- -receptor mediated signal transduction, and inhibition of TNF-
bioactivity (Holstad and Sandler, 2001; Namazi et al., 2004; Abdul-Hai et al., 2005).
Since in vitro and in vivo data have proven a synergistic action of type 1
proinflammatory cytokines in determining β cell destruction (Rabinovitch and Suarez-
Pinzon, 1998), it seems likely that compounds capable of simultaneously inhibiting most
of them would prove more effective than specific neutralization of a single cytokine with
the specific inhibitor. Along this line of research, in vitro and in vivo studies were carried
out aimed at screening test compounds capable of inhibiting both the synthesis and the in
vivo activity of TNF-α. In a first line of screening, a series of test compounds were
analyzed for their capacity of inhibiting TNF-α synthesis from murine splenic
mononuclear cells (SMNC) and peritoneal macrophages (PM) stimulated with either
Concanavalin A (ConA) or lipopolysaccharide (LPS), respectively. Those test
compounds achieving a significant inhibition of TNF-α in these assays were additionally

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
6
evaluated for their efficacy of counteracting TNF-α-dependent LPS-induced lethality in
mice. We found that, among the compounds tested, (S,R)-3-phenyl -4,5-dihydro-5-
isoxasole acetic acid- (VGX-1027) exhibited the highest potency in inhibiting the
synthesis of TNF- from macrophages and/or T cells and afforded the highest degree of
protection against the lethal effects of LPS in the mice (SSG, YAL and FN, unpublished
observations).
These immunopharmacological characteristics and the above mentioned
pathogenic importance of TNF-α in autoimmune diabetogenesis, make VGX-1027 a
suitable candidate for testing in preclinical models of human T1D and prompted us to
evaluate its effects both in the NOD mouse and the mouse made diabetic with MLD-STZ.
VGX-1027 exhibited powerful antidiabetogenic effects upon in vitro, ex vivo and in vivo
conditions. In fact, it prevented IL-1β+IFN-γ-induced pancreatic islets death in vitro and
it powerfully suppressed clinical and histological signs of the disease in both spontaneous
and accelerated (Lukic et al., 1998; Like and Rossini, 1976; Kolb H, 1987; Shoda et al.,
2
005) forms of autoimmune diabetes in NOD mice and in the mouse made diabetic with
MLD-STZ . Simultaneously, ex vivo studies carried out in mice with MLD-STZ-induced
diabetes indicated that VGX-1027 dampened the immunoinflammatory diabetogenic
processess at multiple levels, including the production of proinflammatory and cytotoxic
mediators from both macrophages and pancreatic β cells.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
7
Materials and Methods
Reagents and drugs
3
Streptozotocin
(STZ,
S-0130),
H-thymidine,
sulfanilamide,
naphthylethylenediamine dihydrochloride, and cyclophosphamide (CY) were purchased
from Sigma (St. Louis, MO). RPMI-1640 medium was supplemented with 1mM Hepes
buffer, 5-10% fetal calf serum (FCS), as indicated, 1% sodium pyruvate, 2mM l-
-
5
glutamine (all from Flow Laboratories, Irvine, GB), Penicillin/Streptomycin and 5x10
M 2-mercaptoethanol (Sigma). Recombinant mouse cytokines IFN- , TNF- , IL-1 were
from Sigma. VGX-1027 (Fig. 1) was synthesized as described previously (Eichenger et
al., 1997).
The concentrations of VGX-1027 used in vitro were selected on the basis of preliminary
experiments that evaluated the ability of the compound when added in vitro in the range
between 1 and 100 µg/ml to inhibit TNF-α production from either Con A-stimulated
murine splenic mononuclear cells or LPS-stimulated murine peritoneal macrophage. The
concentration of 10 µg/ml was found to be the most effective concentration at which
VGX-1027 inhibited the synthesis of TNF-α from these cells in the absence of toxicity
on cell vitality that was assessed by crystal violet staining (FN, IC and SSG, unpublished
observation).
The in vivo doses of VGX-1027 were chosen on the basis of preliminary studies carried
out in the TNF-α -dependent (Mohler et al. 1993) mouse model of LPS-induced

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
8
lethality that indicated the doses of 20 mg/kg bw i.p. and 100 mg/kg bw p.o. as the most
effective doses out of those tested (5, 10 and 20 mg/kg bw i.p. and 60, 100 and 120
mg/kg/bw p.o.) in significantly reducing the cumulative incidence of lethality as
compared to vehicle-treated controls (FN data not shown).
Mice
Inbred CBA/H male mice that are genetically susceptible to development of
immunoinflammatory diabetes following MLD-STZ were originally obtained from the
Jackson Laboratory (Bar Harbor, ME, USA) and have then been bred at the Institute for
Biological Research (Belgrade, Serbia) for more than 20 years. Adult male mice, at 6-8
weeks of age, in a body weight ranging from 25 to 30 g were used in all experiments, and
each experimental group consisted of 7-8 mice.
Female NOD mice were obtained from Charles River (Calco, Italy) and were kept
at the animal house of the Department of Biomedical Science of the University of Catania
(Italy).
Both strains of mice were kept under standard laboratory conditions (non specific
pathogen free) with free access to food and water. The handling of animals and the study
protocol were in accordance with international guidelines and approved by the local
Institutional Animal Care and Use Committee.
Phamacokinetic analysis of VGX-1027 in mice

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
9
Single dose of VGX-1027 (20 mg/Kg bw) in vehicle (Na
2
HPO 500mM) were
4
administered via i.p. Blood samples (approximately 0.25- 0.3 ml) were collected prior to
dose administration and 0.30, 1, 1.30, 2, 3, 4, 6 and 8 h after dosing via cava vein
puncture and processed for plasma. Each determination consists of plasma samples
pooled from 2 mice. There were 10 mice for each time point considered.
Sample preparation
Before the extraction procedure, clear and homogeneous plasma samples obtained
for pharmacokinetic evaluation were thawed in a water bath at 30°C and then centrifuged
at 2500×g for 10 min. Five hundred µl plasma samples pooled from different normal
(VGX-1027-untreated) mice containing VGX-1027 were mixed with a 0.1 M pH 7
phosphate buffer (500 µl). Then, samples were purified by solid-phase extraction. The
extraction procedure was carried out by using conditioned Oasis SPE cartridges HLB (1
ml, 30 mg) from Waters (Milford, MA), that were connected to a LiChrolut extraction
unit (Merck). The stationary phase of cartridges were activated by elution under vacuum
with 2 ml of a methanol/water (HPLC grade) mixture. Samples were charged on
extraction cartridges, washed under vacuum with a 5 % (v/v) methanol aqueous solution
(1 ml) and hence eluted twice with 1 ml of methanol acidified (pH 2.4) with
trifluoroacetic acid. The elutate was collected in a glass tube and evaporated under a
nitrogen stream at 40 °C by using a sample concentrator Teche DRI-BLOCK-3D. The
residue was dissolved with acetonitrile (200 l), centrifuged by a Mini Spin Eppendorf,
filtered by Anotop 10 LC (0.2 l/10mm) Whatman filters and then injected onto the
chromatographic apparatus.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
10
Control samples were prepared by adding 100 µl of an acetonitrile VGX-1027
solution at scalar concentrations. Control samples were frozen at -20 °C and then
submitted to the procedure above reported to validate both the analytical method and the
extraction procedure.
Extraction efficiency
The mean recovery of VGX-1027 from spiked mouse plasma samples (control
samples) was evaluated to test the efficiency and reproducibility of the extraction
procedure. The determination was carried out in replicate (n=3) in all samples. The
extraction was conducted as described above. The responses of these standards taken by
means of the extraction procedures have been compared with those of standard solution
at the same concentration injected directly into the liquid chromatographic apparatus. The
peak-areas were compared to that of standard aqueous samples without extraction.
The extraction efficiency was expressed as the percentage of the amount of VGX-
1
027 that was able to extract from control plasma samples spiked with known amount of
the drug. The extraction efficiency of the method, as a function of the drug amount was
reported in the Table.
HPLC experiments
The chromatographic system is a Jasco PU 1580 intelligent HPLC pump (Tokyo,
Japan) equipped with a JASCO MD 1510 multiwavelenght detector (Tokyo, Japan) set at
VGX-1027 (max) 260 nm. A 100 µl loop was used for analytical determinations. The
HPLC apparatus was connected to a computer running the Jasco Borwin software
program ver. 1.5 both for acquisition and data analysis. The chromatographic separation

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
11
was carried out by an Agilent Nucleosil C18 (250×4.6 mm, 5 m) reversed-phase column
with an Agilent column guard. The analytical column was thermostated to 25 °C with a
block heater Gastorr GF 103 (Jones Chromatography, CO). The mobile phase was an
acetonitrile/pH 2.01 water (acidified with trifluoroacetic acid) mixture (35:65 v/v). The
mobile phase was delivered at a flow rate of 1.0 ml/min.
To determine the amount of VGX-1027 in mouse plasma samples a
calibration curve was carried out by preparing control samples spiked with known
amounts of drug added in plasma samples and plotted as a function of the
2
chromatographic peak area (Figure 2). The calibration curve presented a r value of
0
.99984. The curve was constructed from five replicate measurements of five different
drug concentrations over an interval of 0.1-50 µg/ml.
The specificity of the analytical method was assessed by comparison of
chromatograms for the presence of interfering peaks and changes in retention times.
Chromatograms of extracted plasma samples, derived from non treated animals, and
spiked with low and high concentration of VGX-1027 were examined.
Clinical evaluation of diabetes
NOD mice were examined for diabetes development by twice weekly
determination of glycosuria followed, when positive, by measurement of glycaemia. The
NOD mice were defined as diabetics when fasting blood glucose levels exceeded 11.8
mmol/l for two consecutive days. The development of the disease in MLD-STZ-induced
diabetes was evaluated by measuring blood glucose level and body weight loss on a

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
12
weekly basis. For determination of blood glucose levels in MLD-STZ diabetes,
nonfasting mice were bled at the indicated time points from the retro-orbital venous
plexus with heparinized capillary tubes. Plasma glucose concentration was determined by
a glucose-oxidase method, using a blood glucose meter with electrode (Sensimac®,
IMACO GmbH, Lüdersdorf, Germany).
Spontaneous and accelerated diabetes in NOD mice and in vivo treatments
To evaluate the impact of VGX-1027 on the development of spontaneous type 1
diabetes in the NOD mice, euglycaemic female 12 week old NOD mice were treated i.p.
th
with either 20 mg/kg bw VGX-1027 or its vehicle six times a week from the 12 until the
th
2
5 week of age. An additional control group of mice was left untreated.
For induction of diabetes with CY, the drug was dissolved in water for injection
and injected intraperitoneally (i.p.) at the dose of 200 mg/kg bw into 12 to 14 week old
euglycemic NOD mice.
7
For adoptively transferred diabetes, 1x10 spleen cells from acutely diabetic mice
were injected intravenously (i.v.) to euglycemic 4-weeks old female NOD mice.
Both in CY-induced and adoptively transferred diabetes, the mice were treated i.p.
with either 20 mg/kg bw VGX-1027 or its vehicle daily, six times a week starting one
day after the diabetogenic challenge. Treatment was continued until the end of the
studies, 2 weeks after CY-injection or 40 days after cell transfer. At the end of the
experiment, these latter mice were sacrificed and pancreata specimens collected for
histological analyses.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
Histological evaluation of insulitis in NOD mice
13
Mononuclear cells infiltration of the pancreatic islets was graded in a blind
fashion as described elsewhere (Nicoletti et al 1996): 0, no infiltrate; 1, peri-ductular
infiltrate; 2, peri-islet infiltrate; 3, intra-islet infiltrate; 4, intraislet infiltrate associated
with β cell destruction. At least 15 islets were counted for each mouse. A mean score for
each pancreas was calculated by dividing the total score by the number of islets
examined. Insulitis scores (IS) are expressed as mean values ± SD.
MLD-STZ-induced diabetes in CBA/H mice and in vivo treatments
Immunoinflammatory diabetes was induced with MLD-STZ, as described earlier
(Lukic et al., 1991). Briefly, STZ was dissolved in citrate buffer pH 4.5 and injected i.p.
at doses of 40 mg/kg (low dose) daily for 5 consecutive days. Day 0 was defined as the
first injection of STZ.
In order to evaluate the effect of VGX-1027 on the disease development, the drug
was administered as a continuous 12-day treatment, upon either an “early” or “late”
prophylactic regime; while for the former, treatment of the mice started 1 day prior to
MLD-STZ, for the latter it started the day after the last STZ dose. VGX-1027 was
dissolved in 500 mM Na
2
HPO
4
(pH 8.5), further diluted in H O and administered to mice
2
either i.p. or perorally (p.o.), as indicated in the results. A daily i.p. dose of VGX-1027
was 10 mg/kg bw or 20 mg/kg bw, while 100 mg/kg bw was used for p.o. treatment. The
control mice were treated under similar experimental conditions with an equivalent
amount of vehicle.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
Histology and immunohistochemical analysis of pancreas in MLD-STZ diabetic mice
14
In selected experiments of MLD-STZ-induced diabetes the pancreata from
individual mice (n = 7-8 per group) were fixed in 10% formalin, embedded in paraffin,
sectioned, and routinely stained with hematoxylin and eosin for histological examination
by light microscopy. Multiple non-consecutive sections (4 or more) randomly selected
from each pancreas (a total of 20-40 islets/animal) were analyzed.
Immunohistochemical analyses were performed on frozen 9- m-thick pancreata
sections obtained from mice with MLD-STZ-induced diabetes as described elsewhere
(
Lukic et al., 1991), or 5- m-thick paraffin sections. To optimize immunohistological
staining of paraffin sections, the microwave antigen retrieval technique was used
Penkowa and Hidlago, 2000). Briefly, deparafinized sections were immersed in 0.01M
(
sodium citrate buffer (pH 6.0), boiled in a 750-W microwave oven for 10 min and cooled
to room temperature. Staining was performed using primary mAbs: guinea pig anti-swine
insulin Ab, strongly cross-reactive with insulin from several mammalian species,
including mouse (Tian et al., 2004) (A0564, Dako, Hamburg, Germany), and rabbit anti-
mouse iNOS (N 7782, Sigma). For detection, we then used the ExtrAvidin peroxidase
staining kit (Sigma) with 3,3’-diaminobenzidine as substrate. Sections were
counterstained with Mayer’s hematoxylin and mounted in Canada balsam.
Apoptosis was assessed with a basic terminal deoxynucleotidyl transferase
(TUNEL) assay using TACS™ TdT in situ apoptosis detection kit (R&D Systems,
Oxford, United Kingdom) according to manufacturer’s guidelines. The slides were then
counterstained in methyl green and examined on a Zeiss Axiolab microscope. At least 4
mice per condition and 20-40 islets per mouse were examined.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
15
Cell preparations and cultures
Pancreas, spleen and resident peritoneal cells (PC) were collected from individual
mice given MLD-STZ and treated with either VGX-1027 or its vehicle, on day 15 after
the first injection of STZ, as well as from normal untreated animals. The islets of
pancreas, splenic mononuclear cells (SMNC), and PC, were prepared as previously
described (Cvetkovic et al., 2005; Stosic-Grujicic et al. 2004). Cell culture supernatants,
used for ex vivo detection of cytokines and nitric oxide (NO), were obtained by culturing
6
the cells for 48 h in 24-well Limbro culture plates in 1 ml of a standard medium (1x10
6
5
PC, 5x10 SMNC, or 1x10 islets/well).
Culture of murine insulinoma cell lines
Mouse insulinoma MIN6 cells (with permission of Dr. J.-I. Miyazaki, Osaka
University, Osaka, Japan) (Miyazaki et al. 1990), and rat insulinoma RIN-m5F cell line,
were kindly donated by Dr. Karsten Buschard (Bartholin Instituttet, Copenhagen,
Denmark). Cells were cultured in tissue culture flasks (Sarsted, Numbrecht, Germany) in
RPMI-1640 medium containing 10% FCS, until reaching approx. 80% confluence, when
they were detached by standard trypsinization procedure. Cells were washed and seeded
for cytokine and NO production into 96-well flat-bottomed cell culture plates
4
(
6x10 /well) in the presence of various combinations of recombinant mouse cytokines
(IL-1 , IFN- and/or TNF- ), with or without VGX-1027. Culture supernatants were
collected after 48 h. In some experiments the assessment of cell viability after cytokine
treatment was performed. The cytotoxic action of cytokines was evaluated by fluorescein

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
diacetate (FDA) staining, and by crystal violet colorimetric assay (Kaludjerovic et al.,
005), after 24 h and 48 h of cell cultivation, respectively, at the same conditions as
16
2
described above. Nonfluorescent molecule FDA is hydrolyzed inside the viable cells to
fluorescein by the intracellular esterases, the activity of which is reduced in the cells
undergoing apoptosis. To discriminate live and dead cells, FDA dissolved in RPMI
medium was added in cell cultures (100 ng/ml) and after 20 min on 37°C, cells were
trypsinized, washed twice and analyzed on FACSCalibur flow cytometer using
CellQuestPro software (Becton Dickinson, Heidelberg, Germany). FDA positive cells
were considered viable. The staining with crystal violet has been used as a test for the
viability of adherent cells (Kaludjerovic et al., 2005). At the end of incubation, cell
cultures were washed with PBS to remove nonadherent dead cells, the remaining
adherent cells were fixed with methanol and stained with 1% crystal violet. After
thorough washing, the stain was extracted by the addition of 33% acetic acid, and the
absorbance of the dissolved dye, corresponding to the number of live adherent cells, was
measured at 570 nm in a microplate reader.
Nitric oxide production
Nitrite accumulation, an indicator of NO release in the supernatant was
determined using the Griess reaction as described (Cvetkovic et al., 2005). Briefly,
samples of cell-free culture supernatants were mixed with an equal volume of Griess
reagent (a 1:1 mixture of 0.1% naphthylethylenediamine dihydrochloride and 1%
sulfanilamide in 5% H
3
PO ). After 10 min, the OD was measured at 570 nm in a
4
microplate reader, and compared with a standard curve of NaNO
2
.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
17
Measuring of cytokines in culture supernatants
Cell-culture supernatant samples were analyzed in duplicate for murine TNF- ,
IL-1 , IFN- , and IL-10 by ELISA using anti-mouse paired antibodies (R&D System,
Minneapolis, MN, USA) according to manufacturer’s instructions.
Cell-based ELISA
The ex vivo expression of MIF in PC and islets of Langerhans was determined by
cell-based ELISA according to previously described protocol (Cvetkovic et al., 2005).
Rabbit anti-mouse MIF antibody (at 1:2000 dilution) was used as a primary Ab, while
detecting Ab was HRP-conjugated anti-rabbit IgG (U.S.Biochemical Corporation,
Cleveland, OH) at 1:2000 dilution. The data obtained by measuring the light absorbance
at 492 nm were corrected for differences in cell number by staining the cells with crystal
violet after the ELISA procedure.
Semiquantitative RT-PCR
Total RNA was isolated from pancreatic islets with Trizol reagent (Genosys,
Woodlands, TX) according to manufacturer’s instruction. RNA was reverse transcribed
using Moloney leukemia virus reverse transcriptase and random primers (Pharmacia,
Uppsala, Sweden). PCR amplification of cDNA with primers specific for TNF- and -
actin as a house-keeping gene, was carried out in a Mastercycler Gradient thermal cycler
o
(
Eppendorf, Hamburg, Germany) as follows: 30 s of denaturation at 95 C, 30 s of
o
o
annealing at 58 C, and 30 s of extension at 72 C. For each gene, preliminary experiments

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
18
were conducted to ascertain that amplification of cDNA was in the linear range under the
respective cycling conditions. For TNF- , the primers were: sense, 5’-
ACGCTCTTCTGTCTACTGAAC-3’; antisense, 5’-CTTGTCCCTTGAAGAGAACC-3’
and the PCR product was 304 bp. The primers for -actin were: sense, 5’-
TCCTTCTTGGGTATGG-3’; antisense, 5’-ACGCAGCTCAGTAACAG-3’ and the PCR
product was 358 bp. The PCR products were visualized by electrophoresis through 2.5%
agarose gel containing ethidium bromide, gels were photographed and results analyzed
by densitometry using KODAK 1D 3.6 software.
Statistical analysis
Results are shown as mean values ± SD. Unless otherwise specified, statistical
significance of differences between groups was evaluated using ANOVA, followed by
Student-Newman-Keuls test for multiple comparisons between treatment groups,,
2
unpaired Student’s t-test for means between two groups, or χ test as indicated. In the
spontaneous model of diabetes of NOD mice differences in the kinetic and cumulative of
incidence of the diabetes were assessed by Logrank (Mantel-Cox). A p<0.05 was
considered to be significant.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
Results
19
In vitro effects of VGX-1027 on the production of TNF-α and NO by islet β cells and on
their survival
The exposure of islet cells to IL-1 in combination with IFN- and/or TNF- is
known to induce severe functional suppression and death (Rabinovitch and Suarez-
Pinzon, 1998; Eizirik and Mandroup-Poulsen, 2001). Therefore, experiments were
carried out to ascertain whether VGX-1027 could modulate production of
proinflammatory mediators as well as survival of cytokine-induced pancreatic islet death
in vitro. Since the sensitivity of the islets to the action of cytokines is known to vary
among different species we carried out the experiments in freshly isolated pancreatic
islets, rat (RIN-m5F) and mouse (MIN6) insulinoma cell lines as cell models. Exposure
of MIN6 or RIN-m5F cells to 5 ng/ml of recombinant IL-1 + IFN- for 48 h resulted in
the production of TNF- (Fig. 3A). Similarly, overnight cytokine exposure of freshly
isolated islets, as well as RIN-m5F cells resulted in high levels of nitrite (Fig. 3B). VGX-
1
027 (10 g/ml) significantly inhibited both IL-1 /IFN- -induced TNF- and nitrite
accumulation (Fig. 3A and B). We also assessed the capacity of VGX-1027 to interfere
with cytotoxic effects of the cytokines by examining the death and survival of cells.
Results from FDA-staining (Fig. 3C) and crystal violet assay (Fig. 3D) of MIN6 cells
showed that significant increase of cell survival was observed in the presence of VGX-
1
027. In contrast, the protective effect of VGX-1027 was annulled by exposure to
exogenous TNF- . These in vitro results urged us to proceed testing of VGX-1027 in
various preclinical models of T1D.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
20
Pharmacokinetic analysis of VGX-1027 in mice
A high sensitive and selective analytical method was developed to allow the
determination of VGX-1027 in the microgram range in plasma samples. This procedure
consists of the use of solid-phase extraction and reversed-phase high-performance liquid
chromatography with ultraviolet detection.
Figure 4 (panel A) showed a retention time for VGX-1027 of 5.027±0.3 min. The
lower limit of detection was 0.1 µg/ml, while the lower limit of quantification was 0.5
µg/ml.
The high selectivity of the chromatographic method is proven by the absence of
interfering peak at the retention time of VGX-1027 (Figure 4, panel A and B).
The pharmacokinetic profile (Figure 5) of mice treated i.p. with VGX-1027 showed a
peak plasma concentration of 130 µg/ml (Cmax) that was achieved after 2 h (Tmax). In
figure 4,( panel C), it is possible to observe the presence of the VGX-1027 followed by
very small chromatographic peaks which suggests that VGX-1027 is not subjected to
extensive biotransformation in vivo. Of particular interest when observing the
pharmacokinetic profile of VGX-1027 was the observation that concentrations of the
drug well above those found to exert iummunopharmacological effects in vitro (10
micrograms/ml) were already achieved after 30 minutes from the i.p. injection of the drug
and that these levels were maintained until 5 hours after its administration (Figure 3). In
addition, referring to the linear part of the pharmacokinetic profile of VGX-1027 a t½ of
~90 min was calculated.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
21
VGX-1027 prevents development of spontaneous type 1 diabetes in NOD mice and also
counteracts accelerated diabetogenesis induced by CY-challenge or adoptive transfer of
diabetogenic spleen cells in NOD mice
Most of the female NOD mice treated with either the vehicle of VGX-1027
(13/16, 81.3%)or that were left untreated (12/16, 75%) developed diabetes by the end of
the study at the age of 25 weeks. In contrast, the prolonged treatment with VGX-1027
markedly and significantly reduced the cumulative incidence of the disease in these mice
with only 5 out of 16 mice (31.3%) becoming diabetics by this period of age (Fig. 6A, p<
0
.0001 vs both vehicle-treated controls and untreated NOD mice by Logrank [Mantel-
Cox] ). Importantly, the prolonged treatment with the drug appeared to be well tolerated
by the mice as judged from their behavior and the general appearance. No difference in
food consumption was either observed between the mice treated with VGX-1027 and the
mice treated with the vehicle or left untreated (data not shown). Accordingly, there were
no significant differences in body weight gain between the mice treated with VGX-1027,
its vehicle or that were left untreated that remained euglycaemic throughout the study
period (the diabetic mice were not considered in the analysis as they characteristically
lose body weight after disease onset [data not shown]).
Concurring with the antidiabetogenic effects of VGX-1027 observed in the
spontaneous model of diabetes, the drug turned out to markedly reduce the development
of the disease in accelerated models of autoimmune diabetogenesis. VGX-1027 was first
tested in the model of CY-induced diabetes. As expected, within 2 weeks after the
challenge, most of the control NOD mice that were treated with the vehicle of VGX-1027
(15/20, 75%) developed classical signs of diabetes with two or more consecutive days of

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
22
glycosuria and fasting hyperglycaemia. In contrast, the incidence of diabetes was
significantly lower in the mice that were treated with VGX-1027, with only 6/20 (30%)
having developed diabetes during the same period of time (p= 0.01 vs vehicle-treated
2
controls by χ ) (not shown in the figure) .
In another model of accelerated diabetes in NOD mice, starting from 23 days after
the transfer of spleen cells from acutely diabetic donors, the control mice treated with the
vehicle of VGX-1027 progressively started to develop diabetes with a cumulative
incidence of 81.2% (13/16) on day 40 post transfer (Fig. 6B). In agreement with the
preventive effects of VGX-1027 in CY-induced diabetes, the drug was also capable of
significantly reducing the development of adoptively-transferred diabetes in the NOD
2
mice as only 25% (4/16, p=0.005 vs vehicle-treated controls by χ ) of the VGX-1027-
treated mice developed diabetes within the same period of time. In the VGX-1027-
treated mice that developed disease there was no significant difference in the kinetics of
disease development between VGX-1027 and vehicle-treated diabetic mice (30.3±3.0 vs
3
1.6± 4.1 days, respectively).
Concurring with the clinical findings, histological analyses carried out in
pancreatic specimens obtained in these two groups of mice at the end of the study showed
that VGX-1027-treated mice had a significantly milder form of insulitis than the vehicle-
treated controls (insulitis score =1.94± 0.91 vs 2.91 ± 0.8 , respectively, p= 0.003 by one
way ANOVA).
VGX-1027 treatment reduces clinical signs of MLD-STZ-induced diabetes and
suppresses pathohistological changes of pancreas

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
23
In order to evaluate the ability of VGX-1027 to interfere with
immunoinflammatory diabetogenic pathways induced by MLD-STZ, the CBA/H mice
were treated i.p. daily, for 12 consecutive days with either 10 or 20 mg/kg bw of the
drug. The group of control mice that were challenged with MLD-STZ and received the
vehicle from day –1 to day 10 developed persistent hyperglycemia that started from
around 2 weeks after the first injection of the STZ. In contrast, mice treated under the
same experimental conditions with VGX-1027 exhibited a dose-dependent reduction in
blood glucose levels (Fig. 7A). This protection did not depend on continuous application
of the drug, since none of the mice developed hyperglycemia throughout the entire
follow-up period after treatment withdrawal. Moreover, VGX-1027 (100 mg/kg bw per
day) showed similar effect when given p.o. (Fig.7B).
In order to avoid direct interference of the drug with STZ or its initial damaging
effects towards cells, we next tested the effect of treatment with VGX-1027 at a later
stage of disease development upon a “late prophylactic” regime when, though the mice
are still normoglycaemic, immunopathogenic processes have already been initiated by
STZ (Karabatas et al., 2005). In this case, treatment with VGX-1027 was commenced
immediately after the last STZ injections (from day 5 to day 16). In accordance with the
results obtained when the drug was used upon “early prophylaxis”, down-regulation of
diabetes development was also achieved by the “late prophylactic” treatment with VGX-
1
027 (Fig. 7C). In addition, the slight lost in body weight observed in MLD-STZ-treated
control mice during disease development (Fig. 7D) was not observed in VGX-1027-
treated animals.

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
24
We next evaluated the impact of VGX-1027 on histopathological events taking
place at the level of pancreatic β cells during development of diabetes in MLD-STZ-
induced diabetes. Representative examples of the light microscopical evaluation of the
islets are presented in Fig. 7E-J. We have previously reported the appearance of a limited
but significant insulitis and subsequent depletion of cells in mice by 10-20 days after
MLD-STZ administration (Lukic et al., 1991). Consistent with our earlier studies, as
early as 15 days after the injections of MLD-STZ, mild mononuclear cell infiltrates of
endocrine pancreas accompanied by initial necrotic changes could be observed (not
shown). From this period up to the end of the serial measurement of blood glucose levels
(
8 weeks), the destructive process progressed, and most of the islets lost clear margins
Fig. 7E). By contrast, the prophylactic administration of VGX-1027 abrogated the
(
development of these changes. The pancreatic sections from VGX-1027-treated mice did
not harbor a substantial infiltrate and the majority of pancreatic islets appeared normal,
with no signs of inflammation (Fig. 7F). Concordantly with the clinical status of these
mice, the distribution of insulin positive
cells (Fig. 7H), was normal long after
treatment cessation. Moreover, histological evidence for apoptosis was consistently found
in the control diabetic mice, as evidenced by TUNEL staining (Fig. 7I), whereas in the
islets of VGX-1027-treated mice no such evidences for the apoptosis were found
throughout the entire follow-up period (Fig 7J), indicating that the treatment with VGX-
1
027 preserved islets from autoimmune attack, thus enabling appropriate regulation of
plasma glucose levels. These data confirmed that the antidiabetogenic effect of VGX-
027 was maintained both at the clinical and histological level.
1

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
VGX-1027 treatment down-regulates the production of proinflammatory mediators
25
To elucidate the mechanisms by which VGX-1027 treatment prevents disease
onset, functional studies were performed on both pancreas and peripheral compartments
during early progression of the disease. Hence, pancreatic islets, splenocytes and
peritoneal cells were harvested on day 15 after MLD-STZ injection from mice that had
been treated i.p. daily for 12 consecutive with either VGX-1027 (20 mg/Kg) or its vehicle
and the levels of the diabetogenic mediators TNF-α , IL-1β, MIF and NO were examined
ex vivo.
As shown in Fig. 8, both pancreatic and peripheral immune cells established
from mice treated with 0.5 mg VGX-1027 i.p, released significantly reduced amounts of
TNF-
in comparison to control diabetic animals. The inhibition by VGX-1027
treatment in PC (Fig. 8A) and SMNC (Fig. 8B) was 46.6% and 93.4%, respectively, and
in the islets 91.4% (Fig. 8C). Consistent with this finding, a marked reduction in mRNA
expression was detected between pancreatic islets isolated from VGX-1027-treated
versus diabetic mice (inhibition of 61.3%), indicating that VGX-1027 effect occurs
through down-regulation of TNF-α gene expression (Fig. 8D).
Fig. 9 shows that the animals receiving VGX-1027 produced or released less MIF,
IL-1 and NO in response to MLD-STZ than vehicle treated control mice. The effect was
observed both in PC (Fig. 9A,C,E) (the inhibition 29.7%, 36.4% and 60.9%) and in the
islets of pancreas (Fig. 9B,D,F) (the inhibition 36.0%, 59.6% and 45.1%, respectively, in
comparison to control diabetic mice). In line with the results of nitrite measurement,
iNOS staining of pancreatic islets was significantly lower in mice that received VGX-
1
027 (Fig. 9H) in comparison to vehicle-treated diabetic control (Fig. 9G), suggesting

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
26
that the observed down-regulation of NO production was mediated through reduction of
iNOS expression.
We next studied the effects of VGX-1027 on the production of two prototypical
type 1 and type 2 cytokines such as IFN-γ and IL-10. As previously shown by us and
others (Lgssiar et al., 2004), MLD-STZ-treated mice demonstrated significant up
regulation of IFN- , and down regulation of IL-10 (not shown). However, neither of the
cytokines was significantly affected by VGX-1027 administration. IFN- production by
SMNC obtained from VGX-1027-treated and nontreated mice was 800±151 vs. 651±126
pg/ml, respectively, P=0.261, while IL-10 production by PC cultures was 351±46 vs.
3
74±24, P=0.485. Similarly, in the micromilieu of pancreatic islets there were no
noticeable effects of VGX-1027 on either Th1 or Th2 cytokine production (not shown).

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
Discussion
27
VGX-1027 is an orally bioactive small isoxazole compound that is being
developed for the treatment of immune-mediated pathologies. We have shown here for
the first time that VGX-1027 exhibited multiple and powerful antidiabetogenic effects
including prevention of β cell death and suppression of clinical and histological signs of
immunoinflammatory diabetes both in spontaneous and accelerated models of the disease
in the NOD mouse as well as in mice with MLD-STZ-induced diabetes.
Though pathogenic and immunotherapeutic differences occur between MLD-
STZ-induced diabetes and NOD mouse diabetes, it is believed that in both models the
damaged pancreatic islets ultimately become targets for cell-mediated (auto)immune
reactions mounted from T lymphocytes and macrophages which migrate into the
.-
pancreatic islets. β cell destruction could result from the toxic effect of free radicals (O
,
2
.
H
2
O
2
, NO ), cytokines (IL-1, TNF- , TNF- , IFN- ), and other inflammatory products
released by activated macrophages and T cells, as well as β cells themselves (Like and
Rossini, 1976; Kolb, 1987; Rabinovitch, 1998; Rabinovitch and Suarez-Pinzon, 1998;
Lukic et al., 1998; Wachlin et al., 2003; Gurgul et al., 2004; Shoda et al., 2005). In
addition, it has been shown that TNF- and IL-1β may increase vulnerability of
pancreatic cells to autoimmune destruction by inducing the expression of Fas on the
cell surface (Wachlin et al., 2003). Therefore, the capacity of VGX-1027 to counteract
the generation of TNF- and NO by cytokine-stimulated islet cells in vitro anticipates
that its antidiabetogenic property relies on the downmodulation of cytotoxic mediators
within the islets. Although the mechanism by which VGX-1027 regulates the level of
TNF- expression is not clear, it seems that VGX-1027-mediated rescue of islet cells is

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
28
secondary to reduced synthesis of the cytokine rather than on antagonizing its bioactivity
as the addition of exogenous TNF- reverted VGX-1027-mediated protection.
In agreement with these in vitro observations, VGX-1027 showed beneficial
effects in both spontaneous and accelerated forms of diabetes in NOD mice. Similarly, in
MLD-STZ-induced diabetes, VGX-1027 counteracted the diabetogenic effects of the
toxin even when administered upon “late prophylactic” regime after the last injection of
STZ. Ex vivo studies carried out in this model showed that VGX-1027 acted in a bimodal
manner, entailing both a direct protective effect on the cells as well as a profound
impact on key immunopathogenic events associated with disease development including
diminished production of proinflammatory and cytotoxic mediators by macrophages and
pancreatic cells.
An interesting immunopharmacological characteristic of VGX-1027 arising
from these data is its impact on the macrophage secretory capacity of the type 1
proinflammatory cytokines TNF- , IL-1 and MIF and as well as of the pleiotropic
soluble immunomodulatory mediator NO. Type 1 cytokines of the innate immune system
have been proposed as key players in initiating, coordinating and maintaining cell
destruction during autoimmune diabetogenesis (Beyan et al., 2003). NO has repeatedly
been reported to synergize with cytokines in inducing apoptosis of pancreatic cells
possibly via potentiation of JNK activity and inhibition of Akt (Beyan et al., 2003). In
agreement with these observations, we and others have shown that negating the action of
endogenous TNF- , IL-1 , MIF and NO either with specific inhibitors or by gene
deletion prevent development of autoimmune diabetes in NOD mice and/or mice given
MLD-STZ (Lukic et al., 1991; Nicoletti et al., 1994; Sandberg et al., 1994; Yang et al.,

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
29
1
994; Flodstrom et al., 1999; Holstad and Sandler, 2001; Suarez-Pinzon et al., 2001;
Drage et al., 2002; Wang et al., 2002; Thomas et al., 2004).
Note however that it is unlikely that prevention of diabetes by VGX-1027 is
solely due to interference with priming and activation of lymphocytes, because the
protective effect was also achieved by a “late prophylactic” treatment, after the initial
induction of the disease.
The progression of T1D in mice is marked by two general “checkpoints”; the first
is associated with “benign” insulitis with limited β cell destruction, when animals remain
diabetes-free, and the second corresponds to the shift to “aggressive” insulitis when β
cells are destroyed to promote overt diabetes (Andre et al., 1996). The transition from
benign to pernicious insulitis requires an islet cell response to TNF- (Pakala et al., 1999)
which is consistent with the recent observation that TNF- plays a central role in the
+
effector function of diabetogenic CD4 Th1 T cell clones (Cantor and Haskins, 2005).
The marked inhibitory effects of VGX-1027 on in vitro and in vivo production of TNF-
at the pancreatic level as well as the fact that VGX-1027-induced reduction of TNF-
synthesis correlated with reduced severity of pancreatic inflammation all suggest that
inhibition of TNF- may represent an important mechanism by which VGX-1027
influences the transition to aggressive insulitis and hence to clinical appearance of
diabetes. Nonetheless, since the pathogenic effect of endogenous TNF- in NOD mice is
age-dependent and primarily occurs in the early stages of the diabetogenic process (Yang
et al., 1994; Pakala et al., 1999; Cantor and Haskins, 2005; Green and Flavell, 2000;
Christen et al., 2001) it is possible that the antidiabetogenic action of VGX-1027 might

JPET Fast Forward. Published on December 5, 2006 as DOI: 10.1124/jpet.106.109272
This article has not been copyedited and formatted. The final version may differ from this version.
JPET #109272
30
have depended on its presently demonstrated capacity to simultaneously inhibit the
production of other diabetogenic mediators such as IL-1β, MIF and NO.
In immunoinflammatory diabetes Th1 response has been connected with a cell
destructive insulitis, whereas a Th2/3 type response was associated with the protection
from the disease (Kolb, 1997; Rabinovitch and Suarez-Pinzon, 1998). Pharmacological
compounds capable of deviating the predominant Th1 state to a Th2/Th3 state have been
considered suitable candidates for the prevention and early treatment of the disease.
However, in spite of its in vitro and in vivo antidiabetogenic efficiency as well as its
capacity to prevent pancreatic and/or macrophage production of diabetogenic type 1
proinflammatory cytokines and NO, VGX-1027 failed to influence the production of the
two prototypical Th1 and Th2 cytokines, IFN-γ and IL-10. Knowing the powerful
stimulatory effects of IFN-γ on macrophages, these data suggest that the VGX-1027
inhibits macrophage functions in an IFN- -independent manner.
Two observations of particular relevance from the clinical point of view are the
apparent low toxicity of the drug that was witnessed by the lack of characteristic body
weight loss occurring in MLD-STZ-induced diabetes and that, unlike control mice, was
not observed in VGX-1027-treated mice, and the efficacy of the drug to equally prevent
MLD-STZ diabetes regardless of whether it was given i.p. or p.o. The proper evaluation
of VGX-1027 in the treatment of newly diagnosed T1D patients will require preclinical
studies carried out under “therapeutic” dosing to animals with established disease. It is
nonetheless encouraging for the translation of these findings to the clinical setting, the