Diabetes Volume 71, January 2022 157

Multicomponent Plasmid Protects Mice From

Spontaneous Autoimmune Diabetes
Philippe P. Pagni,1 Jay Chaplin,1 Michael Wijaranakula,1 Johnna D. Wesley,1 Jaimie Granger,1
Justen Cracraft,1 Conor O’Brien,1 Nikole Perdue,1 Vijetha Kumar,1 Shangjin Li,1
Sowbarnika Sachithanantham Ratliff,2 Allie Roach,1 Ayesha Misquith,3 Chung-leung Chan,3
Ken Coppieters,4 and Matthias von Herrath2,5
Diabetes 2022;71:157–169 | https://doi.org/10.2337/db21-0327

Type 1 diabetes is an autoimmune disease in which being tested for safety and tolerability in a phase 1 trial

PHARMACOLOGY AND THERAPEUTICS

insulin-secreting b-cells are destroyed, leading to a life- (clinical trial reg. no. NCT04279613, ClinicalTrials.gov).
long dependency on exogenous insulin. There are no
approved disease-modifying therapies available, and
future immunotherapies would need to avoid general- Although the etiology of type 1 diabetes remains incom-
ized immune suppression. We developed a novel plas- pletely understood (1), islet antigen–specific T cells, espe-
mid expressing preproinsulin2 and a combination of cially CD81 T cells (2), are hypothesized to drive the loss
immunomodulatory cytokines (transforming growth of b-cells that characterizes type 1 diabetes. Antigen-pre-
factor-b1, interleukin [IL]-10, and IL-2) capable of near- senting cells pick up b-cell–derived antigens in the pan-
complete prevention of autoimmune diabetes in nonob- creas and present to autoreactive T cells in the draining
ese diabetic mice. Efficacy depended on preproinsulin2, lymph nodes (3). This promotes T cell infiltration of the
suggesting antigen-specific tolerization, and on the pancreas and T cell–mediated b-cell death. Most T cell–tar-
cytokine combination encoded. Diabetes suppression geted therapies tested in clinical trials, including anti-CD3
was achieved following either intramuscular or subcu-
monoclonal antibodies that delete or anergize all T cells
taneous injections. Intramuscular plasmid treatment
(4–6), may at least temporarily cause broad immunosup-
promoted increased peripheral levels of endogenous IL-
pression. When given at diagnosis, all treatments evaluated
10 and modulated myeloid cell types without inducing
global immunosuppression. In preparation for first-in- to date have yielded only transient benefits, and most are
human studies, the plasmid was modified to allow for associated with side effects such as reactivation of latent
selection without the use of antibiotic resistance; this viruses (7,8), making them unsuitable for chronic use. In
modification had no impact on efficacy. This preclinical contrast, immunotherapy given as early as possible may
study demonstrates that this multicomponent, plasmid- prevent clinical type 1 diabetes onset in individuals at risk.
based antigen-specific immunotherapy holds potential The promise of this approach was recently corroborated in
for inducing self-tolerance in persons at risk for develop- a clinical proof of concept phase 2 trial with the anti-CD3
ing type 1 diabetes. Importantly, the study also informs antibody teplizumab, which delayed clinical onset by >2
on relevant cytokine and immune cell biomarkers that years (9). Arguably, however, the ideal therapy for preven-
may facilitate clinical trials. This therapy is currently tion of type 1 diabetes should target only autoreactive

1
Type 1 Diabetes & Kidney Disease, Global Drug Discovery, Novo Nordisk This article contains supplementary material online at https://doi.org/10.2337/
Research Center Seattle, Inc., Seattle, WA figshare.15142284.
2
La Jolla Institute for Immunology, La Jolla, CA
3
Discovery Biologics, Global Research Technologies, Novo Nordisk Research P.P.P., J.C., M.W., and J.D.W. contributed equally to this study. K.C. and
Center Seattle, Inc., Seattle, WA M.v.H. contributed equally as senior authors.
4
Project and Alliance Management, Global Drug Discovery, Novo Nordisk A/S, © 2021 by the American Diabetes Association. Readers may use this article
Måløv, Denmark as long as the work is properly cited, the use is educational and not for
5
Global Chief Medical Office, Novo Nordisk A/S, Søborg, Denmark profit, and the work is not altered. More information is available at https://
Corresponding author: Matthias von Herrath, [email protected] www.diabetesjournals.org/journals/pages/license.
Received 22 April 2021 and accepted 8 August 2021
158 Plasmid Immunotherapy in Type 1 Diabetes Diabetes Volume 71, January 2022

diabetogenic T cells to avoid systemic immune modulation standards at Novo Nordisk Research Center Seattle, Inc.,
resulting in immune exhaustion or deficiency. and La Jolla Institute for Immunology.
Antigen-specific immunotherapy (ASIT) has long been
proposed as key to type 1 diabetes prevention, primarily pVAX1 Standard Plasmid Constructs
based on successful studies in animals. ASIT typically The pVAX1 vector backbone was used to generate plasmid
works by immunization with islet-associated proteins or constructs (Thermo Fisher Scientific). pVAX1 contains a
peptides to suppress autoreactive T cells through deletion human cytomegalovirus (CMV) immediate-early (IE) enh-
or anergy. Unfortunately, to date, clinical trials have con- ancer promoter, kanamycin antibiotic resistance gene
vincingly demonstrated safety but not efficacy (10,11). In (NptII), and pBR322 origin of replication. The following
fact, ASIT has failed to alter the disease course after gene products were synthesized using Eurofins Genomics
hyperglycemia onset in preclinical and clinical studies or Integrated DNA Technologies: proinsulin2 (PI), pre-
(12). Likewise, oral administration of peptides or proteins proinsulin2 (PPI), viral 2A peptide sequences, modified
induces minimal, if any, efficacy before or after disease CD74 invariant chain, viral internal ribosome entry site
onset (13,14). Basic tolerogenic plasmids encoding only (IRES), and mouse transforming growth factor (TGF)-b1,
self-antigens are a promising modality for ASIT and have IL-10, and IL-2. PstI and XhoI restriction sites were used
been shown to be safe in recently diagnosed type 1 diabe- for insertion of the multicistronic gene product into
tes, whereas efficacy remains to be demonstrated (15,16). pVAX1 using restriction enzymes. Q5 High-Fidelity DNA
Efficacy may be enhanced by coexpression of or treatment Polymerase (New England Biolabs [NEB]) was used for
with immune modifiers to generate a micromilieu for cloning gene products and Gibson Assembly Master Mix
autoantigen presentation that favors tolerization (17). (NEB) was used to ligate the gene products into the final
For example, controlled secretion of type 1 diabetes auto- plasmids according to manufacturer-suggested protocols.
antigens, GAD65 or proinsulin, in combination with inter- Plasmids were sequence verified using BigDye Terminator
leukin (IL)-10 by genetically modified lactococcal bacteria v3.1 Cycling Sequencing Kit (Thermo Fisher Scientific).
has been shown to preserve functional b-cell mass in EndoFree Giga prep kits were used to purify plasmids fol-
recent-onset NOD mice but only with coadministration lowing the manufacturer’s protocol with minor modifica-
with anti-CD3 therapy (18,19). tions for maximum yield (QIAGEN): 1 mL bacteria was
We hypothesized that plasmid-mediated coexpression incubated for 8 h in LB Broth (Teknova) to seed 500 mL
of antigen and cytokines would create a microenviron- Super Broth (Teknova) in 1-L baffled flasks.
ment where antigen presentation to autoreactive T cells
leads to limited activation and, consequently, limited Antibiotic Resistance–Free Plasmid Construct
in vivo disease progression. We demonstrate that continu- The pVAX1 vector backbone (Thermo Fisher Scientific)
ous administration of this plasmid effectively prevents was used as a template for the antibiotic resistance–free
type 1 diabetes development in NOD mice without broad vector backbone. The plasmid was constructed in three
immune suppression. Collectively, these data provide a segments by cloning sequence regions using Q5 High-
strong foundation for translation of this novel approach Fidelity DNA Polymerase. The kanamycin antibiotic resis-
to clinical testing in individuals at risk for developing tance open reading frame was removed and substituted
type 1 diabetes. with a modified Escherichia coli infA gene. Additionally,
the pBR322 origin of replication was converted to pUC to
RESEARCH DESIGN AND METHODS increase copy expression. The three segments were ligated
Mice with Gibson Assembly Master Mix following the manufac-
Female NOD/ShiLtJ mice 5–8 weeks of age were from turer’s protocol. The resulting vector was sequence veri-
The Jackson Laboratory. Blood glucose was monitored fied, tested for temperature selection, and designated
weekly via tail vein bleeding with a CONTOUR NEXT pNN for differentiation from pVAX1. EndoFree Giga prep
glucometer (Bayer). Mice were considered diabetic when kits were used to purify plasmids as described above.
blood glucose was $250 mg/dL on two consecutive days
and were euthanized when blood glucose was $600 mg/ Modified E. coli for the Antibiotic Resistance–Free
dL on two consecutive days. For the cyclophosphamide Construct
model, 24 seven-week-old NOD mice were prepared for High-efficiency 5-a DH5a E. coli cells were used for plasmid
each group; 1 animal was removed from the plasmid growth (NEB). For modification of the cells for antibiotic-
group prior to any administrations due to multiple seiz- free selection, the upstream noncoding region of transla-
ures. Cyclophosphamide (Cytoxan) induction was per- tion initiation factor 1 (IF-1/infA) was replaced through
formed with a single injection of 200 mg/kg i.p. in 200 integration of a temperature-responsive transcriptional reg-
mL PBS. All mice were handled according to Novo Nordisk ulator segment of prfA from Listeria monocytogenes into the
guidelines, all local Institutional Animal Care and Use E. coli genome. After the sequence and growth selection
Committee protocols, and Ethical Review Council were confirmed, infA::prfA E. coli cells were cultured and
diabetesjournals.org/diabetes Pagni and Associates 159

generated to be electrocompetent. Competent cells were mice from each group by submandibular venipuncture in
stored at 80 C and used for pNN trans-formation. K2EDTA BD Microtainer Tubes, without anesthesia, at 13
weeks of age. A maximum volume of 150 mL was col-
pVAX1 Plasmid Constructs Quality Control lected. Leukocytes were resuspended for subsequent red
We conducted endotoxin testing by diluting plasmid in blood cell lysis and flow cytometry staining (Supple-
Endosafe LAL reagent water and loading onto an LAL test mentary Table 1). Cells were stained first with LIVE/
cartridge to be read on the Charles River Laboratories DEAD Aqua dye (Molecular Probes) for 30 min, incubated
Endosafe nexgen-MCS system. The quality control thresh- with Fc block for at least 5 min, and then stained with
olds for in vivo studies were set at 0.0025 EU/mg for 40 surface markers including anti-mouse CD45, CD3, CD19,
mg dose, 0.005 EU/mg for 20 mg dose, and 0.001 EU/mg CD11b, CD11c, CD49b, NKp46, Ly6C, Ly6G, F4/80, and
for 100 mg dose. Restriction enzyme digestion was per- CD210 biotin for 30 min. Lastly, cells were stained with
formed using XhoI and PstI-HF enzymes and CutSmart streptavidin-APC-Cy7 for CD210 detection for 30 min.
Buffer (NEB) according to the manufacturer’s protocol. PBS plus 2% FBS was used as wash buffer between the
The expected fragment size was confirmed by running staining steps, and FMOs (fluorescence minus one con-
electrophoresis on a 1% agarose gel. Plasmid sequencing trol) were made for each individual marker to ensure gat-
was performed with use of a set of 18 primers by Sanger ing strategy. Data were collected on a BD LSRFortessa
sequencing with services from GENEWIZ. The .ABI files Flow Cytometer and analyzed using FlowJo software (ver-
obtained from GENEWIZ were aligned to the plasmid ref- sion 10; BD).
erence sequence with use of Geneious software to confirm
the identity of the plasmid.
Insulin Autoantibody Titer Measurements
In Vivo Plasmid Administration Plasma samples were shipped to Barbara Davis Center for
All equimolar doses were diluted using sterile PBS to a Childhood Diabetes at University of Colorado Health Sci-
final volume of 50 or 100 mL for injection. Six to 11- ences Center, and insulin autoantibody (IAA) titers were
week-old mice were treated with 40 mg empty pVAX1 evaluated by radioassay as previously described (20).
plasmid, 50 mg pVAX1-PI, 67.5 mg pVAX1-PI/IL-10, 100
mg pVAX1-PPI/TGF-b/IL-10/IL-2, 100 mg pNN-PPI/TGF- OVA Challenge In Vivo
b/IL-10/IL-2, or PBS as a negative control. Mice were pretreated with 100 mg pVAX1-PPI/TGF-b/IL-
10/IL-2 once weekly starting at 6 weeks of age for 5
Whole Blood Collection and Processing weeks prior to treatment with 50 mg each OVA and alumi-
Blood collection was performed via the submandibular num hydroxide (alum) (Sigma-Aldrich) in 100 mL total
vein for interim time points and via cardiac puncture for volume intraperitoneally at 12 weeks of age. Mice contin-
terminal sample collection. For serum, blood was allowed ued weekly plasmid treatment for 4 weeks and were bled
to clot, centrifuged (4 C, 700g, 5 min), and carefully col- on day 30 after OVA/alum challenge. Serum was collected
lected in prelabeled tubes for storage at 80 C until anti- for anti-OVA IgG measurement using a mouse anti-OVA
body titers were measured. For plasma, blood samples IgG antibody assay kit (Chondrex). All samples were
were collected in K2EDTA BD Microtainer Tubes, inverted stored at 80 C prior to analysis.
10 times, and kept on a rocking platform until processing.
Plasma were separated (4 C, 2,000g, 10 min), carefully
Statistical Analyses
collected in prelabeled tubes, and stored at 80 C until
With use of GraphPad Prism, version 7 (GraphPad Soft-
multicytokine measurements. Leukocytes were resus-
ware, La Jolla, CA), group comparisons were performed
pended for subsequent red blood cell lysis and flow
with the Log-rank (Mantel-Cox) test for type 1 diabetes
cytometry staining immediately.
incidence curves, Kruskal-Wallis tests with Dunn posttest
comparisons for cytokine and flow cytometry data, and
Cytokine Analyses
Cytokine levels were measured with the mouse V-PLEX Mann-Whitney U test for OVA challenge studies.
Proinflammatory Panel 1 assay (Meso Scale Discovery)
following the manufacturer’s instructions and evaluated Data and Resource Availability
with the MESO QuickPlex SQ 120reader (Meso Scale All data generated or analyzed during this study are
Discovery). included in the published article (and Supplementary
Material). Raw underlying data can be requested from the
Flow Cytometry corresponding author upon reasonable request and for
Immunophenotype via flow cytometry was used to assess noncommercial purposes.
frequencies of innate immune cell populations and their The full plasmid sequences used in the current study
expression of IL-10 receptor, also known as CD210, in are available from the corresponding author upon reason-
peripheral blood. Blood was collected from a subset of able request and for noncommercial purposes.
160 Plasmid Immunotherapy in Type 1 Diabetes Diabetes Volume 71, January 2022

RESULTS In vitro transfection analyses of the plasmids con-

Treatment With a Plasmid Coexpressing Proinsulin2 firmed that PI and IL-10 proteins were produced as
and IL-10 Reduces Diabetes Incidence in the NOD expected (data not shown). The three plasmids were
Mouse Model administered via intramuscular (i.m.) injection into NOD
We initially generated three plasmids to study the impact mice beginning at 9 weeks of age. Mice were treated once
of antigen and response modifier expression on diabetes a week for 8 weeks (Fig. 1B) and monitored for disease
prevention: the plasmid DNA backbone without antigen development for 21 weeks. The pVAX1-empty group was
or cytokine (pVAX1-empty), a plasmid encoding mouse not protected from type 1 diabetes development, and
proinsulin2 (PI) antigen only (pVAX1-PI), and one plas- only modest protection was observed for the pVAX1-PI
mid co-encoding mouse PI and the immune-suppressing group, but the diabetes incidence was not statistically sig-
cytokine IL-10, using an IRES to generate a single bicis- nificant different from that of the pVAX1-empty group.
tronic mRNA (pVAX1-PI/IL-10) (21) (Fig. 1A). IL-10 was However, marked protection from type 1 diabetes was
chosen for the initial design, as it is a well-characterized observed for the pVAX1-PI/IL-10 group; the incidence
anti-inflammatory cytokine involved in counteracting was statistically significantly lower than for the pVAX1-
type 1 diabetes pathogenesis (22,23). empty group. This suggested that coexpression of a
response modifier such as IL-10 could increase antigen-
specific plasmid efficacy in vivo.
Additionally, an analogous treatment regimen was
tested in the accelerated cyclophosphamide-induced/syn-
chronized NOD diabetes model. In this model, administra-
tion of the alkylating agent cyclophosphamide, a chemo-
therapeutic for T-cell leukemia, was injected intraperitone-
ally to promote synchronized hyperglycemia onset 2.5
weeks after administration. The plasmid dosing scheme
was the same as described for the spontaneous NOD
model, except that two doses were administered prior to,
and three following, cyclophosphamide injection. The
results confirmed the findings in the less aggressive spon-
taneous-onset NOD model, with significant protection
from hyperglycemia in the pVAX1-PI/IL-10 plasmid treat-
ment group only (Fig. 1C).

Addition of TGF-b1, IL-2, and the Invariant Chain

Leader Peptide to the Plasmid
For further enhancement of the plasmid-mediated
immune modulation and efficacy, PI was replaced with
preproinsulin2 (PPI) with a Sec61-impairing mutation
to prevent cleavage and secretion and for enhancement
of potential antigen diversity (24). We also hypothe-
sized that concomitant addition of TGF-b1 and IL-2 to
the plasmid as response modifiers could promote the
induction and stabilization of T-cell tolerance and
enhance the prevention of type 1 diabetes in vivo. For
coexpression of TGF-b1, IL-10, and IL-2, along with
PPI, in a single mRNA derived from the vector, a viral
IRES and 2A peptides were added to the sequence to
promote protein separation. We further added the inv-
ariant chain (CD74) leader peptide to direct the PPI to
Figure 1—Antigen- and IL-10–encoding plasmid improves type 1 the endosome for MHC class I and II processing to
diabetes prevention efficacy compared with antigen alone counter-
part. A: Plasmid map of first tolerogenic plasmid generation:
improve tolerogenic antigen presentation to both CD41
pBR22 origin of replication, CMV IE promoter, IL-10, IRES, proinsu- and CD81 T cells (Fig. 2A and B) (25).
lin2 (PI), and kanamycin resistance noted. B: Diabetic incidence
graph of treated mice. NOD/ShiLtJ treated with pVAX1 (n = 29), Multicytokine Plasmid Improved Type 1 Diabetes
pVAX1-PI (n = 30), or pVAX1-PI/IL-10 (n = 26). All groups were
Prevention Compared With pVAX1-PI/IL-10
treated for 8 weeks starting at 9 weeks of age. **P < 0.01; ns, non-
significant. C: Diabetes incidence in the cyclophosphamide- In vitro transfection analyses of the plasmids confirmed
induced diabetes model in comparisons of mice treated with that PPI, TGF-b1, IL-10, and IL-2 proteins were produced
pVAX1-PI (n = 23) or left untreated (n = 24). *P < 0.05. as expected (data not shown). For determination of
diabetesjournals.org/diabetes Pagni and Associates 161

Figure 2—Modified plasmid increases type 1 diabetes prevention efficacy. A: Plasmid maps of tolerogenic plasmids: pBR22 origin of rep-
lication, CMV IE promoter, IL-10, IRES, proinsulin2 (PI), CD74 invariant chain, preproinsulin2 (PPI), 2A polylinkers, TGF-b, IL-2, and kana-
mycin resistance noted. B: Overview of transcriptional and translational regulation of all elements encoded by pVAX1-PPI/TGF-b/IL-10. C:
Diabetes incidence of treated mice. NOD/ShiLtJ treated with pVAX1-PI/IL-10 (n = 24), pVAX1-PPI/TGF-b/IL-10 (n = 24), or pVAX1-PPI/
TGF-b/IL-10/IL-2 (n = 24) or left untreated (n = 24). All groups were treated for 21 weeks starting at 9 weeks of age. ***P < 0.001. D: Dia-
betes incidence of treated mice. NOD/ShiLtJ mice treated with i.m. pVAX1-Empty (n = 21), i.m. pVAX1-PPI/TGF-b/IL-10 (n = 41), i.m.
pVAX1-PPI/TGF-b/IL-10/IL-2 (n = 42), or s.c. pVAX1-PPI/TGF-b/IL-10/IL-2 (n = 41) or left untreated (n = 21). All groups were treated start-
ing at 6, and until 24, weeks of age. *P < 0.05; ***P < 0.001. ns, nonsignificant.

whether the modifications further improved type 1 diabe- origin using mass spectrometry–based antibody characteri-
tes prevention, NOD mice were treated with pVAX1-PI/ zation to discriminate between endogenous and the plas-
IL-10, pVAX1-PPI/TGF-b/IL-10, or pVAX1-PPI/TGF-b/IL- mid-derived IL-10 that carries a short 2A-derived tag (data
10/IL-2 weekly (i.m.) for 21 weeks starting at 9 weeks of not shown). Peripheral IL-2 levels remained very low and
age. A near-complete prevention of type 1 diabetes was similar across groups. Circulating TNF-a and IFN-g were
observed with the modified plasmids (pVAX1-PPI/TGF- elevated in mice treated with pVAX1-PPI/TGF-b/IL-10/IL-
b/IL-10/IL-2 and pVAX1-PPI/TGF-b/IL-10), and this was 2 in comparison with other treatment groups (Fig. 3).
statistically significantly different in comparison with the While these cytokines are conventionally considered proin-
untreated group (Fig. 2C). When treatment was adminis- flammatory (26), their protective role during the late
tered to mice at 6 weeks of age, pVAX1-PPI/TGF-b/IL-10/ prevention phases of autoimmune diabetes in mice
IL-2 significantly outperformed pVAX1-PPI/TGF-b/IL-10 has long been known (27,28).
(Fig. 2D). We concluded that the most robust effect was
obtained when IL-2 was included in the multicytokine Multicytokine Plasmid Therapy Alters Plasma Cytokine
plasmid. Concentrations and IL-10R Expression in a Dose-
Further, 7 weeks post–first injection, higher IL-10 levels Dependent Manner
were detectable with pVAX1-PPI/TGF-b/IL-10/IL-2 than For determination of whether plasmid therapy induced
with pVAX1-PI/IL-10 and pVAX1-PPI/TGF-b/IL-10 (Fig. 3). acute changes in immune cell phenotypes, 9-week-old
We confirmed that the IL-10 measured was of endogenous NOD mice were treated i.m. once weekly with 25 mg, 50
162 Plasmid Immunotherapy in Type 1 Diabetes Diabetes Volume 71, January 2022

Figure 3—Modified plasmid induces a peripheral endogenous cytokine signature. A–D: Mouse IL-10, IL-2, TNF-a, and interferon-g
(IFN-g) cytokine concentrations of NOD mice at 12 weeks of age after 8 weeks of treatment. n = 3/group. *P < 0.05; **P < 0.01.

mg, 100 mg, or 200 mg pVAX1-PPI/TGF-b/IL-10/IL-2; all clinical benefit in individuals with IAA $80 nU/mL (29).
groups were significantly and comparably protected from Preclinical data in NOD mice have also shown that preex-
type 1 diabetes development versus untreated controls isting IAA could predict the efficacy of oral insulin in cur-
(Supplementary Fig. 1). At 13 weeks of age, flow cytome- ing type 1 diabetes in combination with anti-CD3 (30).
try was performed on peripheral blood mononuclear cells Thus, we sought to determine whether the efficacy of the
from 12 randomly selected mice per group. In compari- plasmid therapy was dependent on the baseline IAA status
sons of untreated mice versus mice treated with the plas- of the mice. Mice were bled at 9 weeks of age before receiv-
mid, no statistically significant difference was found ing their first plasmid dose, and plasma was analyzed for
regarding innate immune cell frequencies (Supplementary IAA. Though IAA detection at 9 weeks of age in NOD mice
Fig. 2 and Supplementary Tables 1 and 2). However, after is relatively infrequent (20), the number of IAA1 mice in
treatment with pVAX1-PPI/TGF-b/IL-10/IL-2, CD210, our study was 12 of 70 across all treatment groups, com-
the IL-10 receptor, was dose-dependently upregulated on prising 4 of 24 untreated and 8 of 46 plasmid-treated
CD11b1 myeloid cells (Fig. 4A), myeloid dendritic cells
mice. To gain insights into whether IAA would impact
(Fig. 4B), monocytes, Ly6C monocytes, and Ly6C1 clas-
responsiveness to plasmid therapy, we pooled all mice
sical monocytes (Supplementary Table 1). Other cell types
from the 25-mg, 50-mg, 100-mg, and 200-mg pVAX1-PPI/
either showed no (e.g., NK cells) or little (eosinophils and
TGF-b/IL-10/IL-2 plasmid-treated groups based on their
neutrophils) difference in CD210 expression across treat-
IAA status (8 IAA1, 38 IAA ) and plotted versus the diabe-
ment groups. Additionally, while no increase was detected
for circulating IL-2 and plasmid-derived mature TGF-b1, tes outcome at 30 weeks of age. We found that among
there was a dose-dependent increase in IL-10 and TNF-a untreated mice, the proportion of mice developing type 1
(up to the 200-mg dose) and in IFN-g and IL-6 (up to the diabetes was statistically significantly greater among those
100-mg dose) (Fig. 4C and not shown). that were IAA1 (100% by 18 weeks of age) compared with
those that were IAA (30% by 18 weeks of age and 75% at
Multicytokine Plasmid Therapy Is Efficacious end point). In contrast, the proportion of plasmid-treated
Regardless of Individuals’ IAA Status at Baseline mice developing type 1 diabetes was similar in IAA
Post hoc analyses from the Diabetes Prevention Trial–Type (23.7%) and IAA1 (25%) animals (Fig. 5). The conclusion
1 (DPT-1) with use of oral insulin demonstrated a potential was similar based on pooled groups of mice treated
diabetesjournals.org/diabetes Pagni and Associates 163

Figure 4—Plasmid treatment dose dependently alters myeloid cell phenotypes and peripheral cytokine signatures. A: Flow plot represen-
tation for increasing IL-10R expression on CD11b1 myeloid cells with increasing treatment amounts of pVAX1-PPI/TGF-b/IL-10/IL-2. B:
Increasing IL-10R expression on myeloid dendritic cells (DCs) with increasing treatment amounts of pVAX1-PPI/TGF-b/IL-10/IL-2. n = 12/
group. C: Cytokine concentrations of NOD mice at 13 weeks of age after 4 weeks of treatment. *P < 0.05; **P < 0.01; ****P < 0.0001.
164 Plasmid Immunotherapy in Type 1 Diabetes Diabetes Volume 71, January 2022

subcutaneously (s.c.) with plasmid once and thrice weekly

(data not shown). These data seem to indicate that the effi-
cacy of plasmid-mediated type 1 diabetes prevention is not
altered by presence of insulin autoantibodies at treatment
initiation.

PPI Antigen Is Required for Maximal Efficacy, and

Plasmid Therapy Does Not Alter Recall Responses to
Irrelevant Antigen
For determination of whether antigen was required for Figure 5—Multicytokine plasmid therapy is efficacious regardless
maximal prevention, NOD mice were treated weekly, of individuals’ IAA status at baseline. NOD mice at 9 weeks of age
beginning at 11 weeks of age, with i.m. injections of a (wk9) were bled, and their plasma was collected for IAA measure-
plasmid encoding only cytokines and monitored for dis- ments before i.m. weekly dosing was started with 25 mg, 50 mg, 100
mg, or 200 mg pVAX1-PPI/TGF-b/IL-10/IL-2. A group of mice was
ease development. Mice treated with pVAX1-PPI/TGF- left untreated as negative controls. Blood glycemia was followed
b/IL-10/IL-2 were almost completely protected. In com- weekly, and type 1 diabetes diagnosis was determined with two
parison, the diabetes incidence was statistically signifi- consecutive glycemia values >250 mg/dL. At end point: categori-
cantly greater in mice treated with pVAX1-TGF-b/IL-10/ zation of untreated (square symbols) and plasmid-treated (circle
symbols) mice according to negative (empty symbols) or positive
IL-2 encoding immune modifiers but no antigen (Fig. 6A). (closed symbols) IAA status at baseline, plotted relative to glycemic
This demonstrated that the in vivo efficacy associated status (diabetic nonresponders vs. protected responders). Num-
with the multicytokine plasmid is driven, in part, by the bers of mice per group are included in the graph. Kaplan-Meier sur-
coexpression of an islet antigen. vival curves were analyzed with a Log-rank Mantel-Cox test. **P <
0.01. ns, nonsignificant.
Next, for determination of the effect of plasmid ther-
apy on the recall response to an irrelevant antigen, the
antigenic response to OVA in NOD mice was evaluated
with regard to whether our plasmid therapy induced efficacy using the s.c. injection method, we evaluated
broad immune suppression. NOD mice were treated with both once weekly and thrice weekly administration
pVAX1-PPI/TGF-b/IL-10/IL-2 or left untreated for 5 schedules beginning at 9 weeks of age. In an earlier
weeks prior to immunization with OVA in alum (Fig. 6B). study, NOD mice injected s.c. once weekly beginning at
Serum anti-OVA IgG antibodies were measured 30 days 6 weeks of age displayed a diabetes incidence that was
postchallenge. We did not observe any statistically signifi- statistically significantly higher than in the group
cant differences between plasmid-treated and untreated dosed once per week i.m. (14% vs. 0%, respectively;
groups, suggesting that plasmid therapy did not induce P < 0.05) (Fig. 2D). Subsequently, we observed that
general immune suppression in vivo (Fig. 6C). Mice that treatment thrice weekly with pVAX1-PPI/TGF-b/IL-10/
were OVA challenged and either left untreated or contin- IL-2 completely prevented type 1 diabetes (0%) (Fig.
ued being dosed with pVAX1-PPI/TGF-b/IL-10/IL-2 were 7B); in comparison, with once-weekly injections, statis-
followed for type 1 diabetes development until 21 weeks tically significantly more mice (29%) developed dia-
of age. Upon OVA challenge, full protection was seen in betes (P < 0.01).
plasmid-dosed mice compared with untreated counter-
parts, suggesting that plasmid efficacy is not hindered by A Plasmid Without Antibiotic Selection Marker Has
concomitant immunization with an irrelevant antigen Efficacy Comparable With That of the pVAX1 Plasmid
(not shown). Human chromosomal integration of the antibiotic resis-
tance genes, gene transfer to other microorganisms,
Frequent Administration of the Multimodular Plasmid and downstream production complications are concerns
Is Needed to Achieve Optimal Protection in NOD Mice in development of plasmid therapy for clinical use (32).
Though i.m. is a common route of administration for Therefore, to facilitate clinical translation of the multi-
plasmid therapy, the ability to dose through s.c. injec- cytokine plasmid, we developed a novel selection sys-
tion likely has a more favorable path to clinical transla- tem using the temperature selection properties of L.
tion due to the discomfort associated with i.m. monocytogenes. Specifically, prfA, a thermoregulation
injections. Therefore, we next compared s.c. injection virulence regulator for L. monocytogenes that selectively
to i.m. and evaluated the dosing frequency. The mini- promotes the growth of bacteria at temperatures >
mal dosing regimen required for maximal efficacy with 30 C (33), was inserted upstream of the IF-1 gene in E.
i.m. administration was once weekly (Fig. 7A). How- coli to control the expression of the essential gene
ever, our plasmid is adjuvant free, and we were uncer- (34,35). Additionally, we substituted the kanamycin
tain about whether transfection efficiency of the antibiotic resistance gene (nptII) in the pVAX1 vector
plasmid would be sufficient to mount an effective tol- with InfA, creating a new plasmid backbone (Fig. 8A
erogenic response with injection once weekly via s.c. and B). The engineered infA::prfA E. coli produced anti-
(31). To compensate for potential loss of transfection biotic resistance–free plasmid at both 30 C and 37 C. The
diabetesjournals.org/diabetes Pagni and Associates 165

Figure 6—Tolerogenic plasmid requires antigen and does not affect secondary antigen response. A: Diabetes incidence of 11-week-old
NOD/ShiLtJ mice treated with either pVAX1-TGF-b/IL-10/IL-2 (n = 32) or pVAX1-PPI/TGF-b/IL-10/IL-2 (n = 31). *P < 0.05. B: Outline of
ovalbumin 1 alum challenge. NOD mice were either pretreated with pVAX1-PPI/TGF-b/IL-10/IL-2 (n = 8) or left untreated (n = 8) for 5
weeks. At 11 weeks of age, mice were challenged with ovalbumin and alum. Following challenge, mice were left untreated or continued
weekly treatment with pVAX1-PPI/TGF-b/IL-10/IL-2 for an additional 4 weeks and were bled on day 30 postchallenge for anti-OVA IgG
antibodies. C: ELISA analysis for anti-OVA IgG antibodies. n.s., not significant.

antibiotic-free tolerogenic plasmid was then desig- thrice weekly dosing regimens demonstrated efficacy
nated pNN-PPI/TGF-b/IL-10/IL-2. Treatment of NOD comparable with that of the pVAX1-based plasmid.
mice with the new plasmid using once weekly and (Fig. 8C).

Figure 7—Optimal plasmid efficacy requires weekly i.m. dosing or thrice weekly s.c. dosing. A: Weekly i.m. dosing is efficacious as
opposed to biweekly and once monthly. Diabetes incidence graph of NOD/ShiLtJ mice treated i.m. with pVAX1-PPI/TGF-b/IL-10/IL-2
once weekly (n = 29), once every other week (n = 28), or once monthly (n = 30). B: Subcutaneous dosing requires thrice weekly dosing for
optimal efficacy. Diabetes incidence of NOD/ShiLtJ mice treated s.c. with pVAX1-PPI/TGF-b/IL-10/IL-2 either once weekly (n = 24)
(inverted red triangles) or thrice weekly (n = 24) (red diamonds). All groups were treated for 21 weeks starting at 9 weeks of age. **P <
0.01; ***P < 0.001. wk, week.
166 Plasmid Immunotherapy in Type 1 Diabetes Diabetes Volume 71, January 2022

Figure 8—Antibiotic-free plasmid does not alter type 1 diabetes prevention efficacy. A: Plasmid maps comparing antibiotic-free plasmids.
CMV IE, multiple cloning site, infA (nptII replacement), and pUC origin of replication noted. B: Temperature selection properties of the plas-
mid. From left to right, the effects of bacterial strain type and plasmid complementation are shown in relation to permissive growth. C:
Nine-week-old NOD mice were dosed s.c. with 100 mg pNN-PPI/TGF-b/IL-10/IL-2 in lead formulation buffer either once or thrice a week
for 21 weeks. Vehicle control group dosed s.c. three times a week for 21 weeks with lead formulation buffer (calcium-containing formula-
tion) was also included. *P < 0.05; ****P < 0.0001. ns, not significant; wk, week.

DISCUSSION sured by the prevention of hyperglycemia in the NOD

Plasmid-based vaccination and immunotherapy to stimu- mouse model.
late or silence antigenic responses, respectively, have Our results demonstrate that a single-autoantigen-
advanced into clinic trials (16,36). While safety has con- encoding plasmid was capable of preventing diabetes
sistently been excellent, neither potent immunogenicity almost completely in a mouse model where b-cell destruc-
nor antigenic tolerization has been convincingly demon- tion is predominantly driven by reactivity to insulin (38),
strated. We therefore sought to improve upon previous as is also likely the case in many humans with type 1 dia-
work with tolerizing antigenic plasmid therapy by gener- betes (39–41). Additionally, the IAA status at baseline did
ating a novel multicomponent plasmid that encoded both not influence the therapeutic efficacy of the plasmid at end
anti-inflammatory cytokines and a disease-relevant auto- point. This suggests that a clinical trial testing such a plas-
antigen to promote a tolerance-inducing microenviron- mid in people at risk for developing type 1 diabetes might
ment around the expressing cells (17,37). Using this not require participant stratification or exclusion based on
plasmid, we demonstrated that such coexpression of IAA status. There was no evidence of broad immune sup-
immune modulators markedly improves efficacy as mea- pression, and we detected little to no systemic levels of
diabetesjournals.org/diabetes Pagni and Associates 167

plasmid-derived cytokines, supporting the hypothesis that cytokines, there is conflicting information as to whether
plasmid-encoded cytokines affect the immune system these cytokines may have a protective or harmful effect
locally to suppress the response to PPI rather than pro- by limiting autoreactivity or stimulating pathogenic CD81
mote global suppression. Furthermore, when one of the cells (47–50). It is possible that plasmid therapy directly
components was omitted or its expression level altered, promotes expression of IFN-g and TNF-a to limit prolif-
efficacy was reduced, indicating that our extensive com- eration and activation of pathogenic CD81 cells, but
pound screening has yielded a unique combination of plas- whether this is an underlying mechanism that facilitates
mid backbone elements, antigen, and cytokines that, tolerance induction remains unclear. Future studies,
together, results in potent disease suppression. including the ongoing phase 1 trial with the pNN plasmid
A strength in our approach was the stringent in vivo (TOPPLE T1D [clinical trial reg. no. NCT04279613,
screening procedure by well-powered NOD studies, neces- ClinicalTrials.gov]), are needed to clarify the relevance of
sitated by our intention to identify a plasmid candidate these unexpected findings via longitudinal measurements
for clinical use. It is generally acknowledged that the tim- of pro- and anti-inflammatory cytokines and chemokines.
ing of diabetes development or overall incidence and con- This would involve investigations to clarify the relevance
sistency of results in, and between, NOD studies can of the findings to the human setting. In this context,
differ significantly between mice from different vendors recent evidence from a trial of anti–TNF-a treatment
and between laboratories (42). To address this, we con- with golimumab, which appeared to improve b-cell func-
ducted our prevention studies at multiple sites, using tion in people with new-onset type 1 diabetes, indicates a
NOD mouse colonies not only from The Jackson Labora- pathogenic role of this proinflammatory cytokine in
tory but also from Taconic Biosciences (data not shown). human type 1 diabetes (51). The current study describes
This provided assurance that the observed efficacy was the molecular design, superior preclinical efficacy, and
robust, as results for both plasmids (pVAX1 and pNN) translational biomarker profile of an advanced plasmid--
were similar across sites and colonies. based immunotherapy. This therapy was tailored for
The in vivo mechanism of action of naked plasmid chronic use in humans for the prevention of type 1 diabe-
immunization for antigenic tolerance induction remains tes in individuals at risk, with IAA positivity, by virtue of
ill-defined (43). It has been suggested that 10–30% of its s.c. route, antibiotic resistance gene–free, and nonim-
cells located near the injection site express the encoded munosuppressant properties. The approach has cleared
protein but that only 1–2% of cells in the nearby muscle formal toxicology studies and has been produced at qual-
(i.e., quadriceps) are transfected (44). The use of adju- ity and scale for the ongoing, multiple ascending doses
vants and/or injection equipment such as gene guns or trial to investigate its safety, tolerability, and pharmacoki-
electroporation may increase transfection efficiency con- netic profile (TOPPLE T1D). In summary, the results dem-
siderably, but it is uncertain what the ideal transfection onstrate that this unique plasmid-based combination
efficiency should be to induce an adequate tolerogenic immunotherapy has potential to provide a path to pre-
response. A small fraction of transfected cells are likely vention of type 1 diabetes.
antigen-presenting cells that migrate to draining lymph
nodes (45), a hypothesis that was recently elegantly con-
firmed with use of fluorescent plasmid tracers upon intra- Acknowledgments. The authors thank Frederik Flindt Kreiner, Novo
Nordisk A/S, Søborg, Denmark, for medical writing and submission support;
dermal administration (46). These cells could produce
David Rodriguez and staff, Novo Nordisk Research Center Seattle, Inc.
regulatory T cell–inducing cytokines and present antigen
[NNRCSI], for vivarium management and animal care; Claire Gibson Bamman,
to circulating CD41 or CD81 T cells, potentially tolerizing Tamar Boursalian, and Jose Luis Vela, NNRCSI, for scientific discussions; and
them to the encoded antigen, preventing autoimmunity. David Purdy and Bong Kim, NNRCSI, for conducting mass spectrometry
These transfected cells could also explain the increased experiments.
CD210 expression on circulating myeloid dendritic cells. Duality of Interest. The study was funded by Novo Nordisk A/S. All
Increasing the understanding of the mechanism of action authors are or have been employees of Novo Nordisk A/S. No other potential
is further complicated by use of s.c. over i.m. administra- conflicts of interest relevant to this article were reported.
tion; s.c. injection likely results in transfection of different Author Contributions. P.P.P., J.Ch., J.D.W., K.C., and M.v.H. gener-
cell populations or reduced transfection efficiency, ated the idea for the study. P.P.P., J.Ch., M.W., J.D.W., and V.K. planned the
experiments. J.Ch., M.W., J.G., J.Cr., C.O., N.P, S.L., S.S.R., A.R., A.M., and
explaining the need for a higher injection frequency with
C.-l.C. conducted the experiments. P.P.P., J.Ch., M.W., J.D.W., N.P., V.K., and
s.c. to mimic the efficacy of once weekly i.m. injection.
S.L. analyzed the collected data. P.P.P., J.D.W., K.C., and M.v.H. wrote and
Studies are required to fully understand the transfection reviewed the manuscript. M.v.H. is the guarantor of this work and, as such,
process and to discover the cell populations that are mod- had full access to all the data in the study and takes responsibility for the
ulated post–plasmid administration. integrity of the data and the accuracy of the data analysis.
We unexpectedly observed increased plasma or serum
IFN-g and TNF-a after the first i.m. injection (4 and References
8 weeks, respectively). While it remains unclear how the 1. van Belle TL, Coppieters KT, von Herrath MG. Type 1 diabetes: etiology,
plasmid promotes expression of these proinflammatory immunology, and therapeutic strategies. Physiol Rev 2011;91:79–118
168 Plasmid Immunotherapy in Type 1 Diabetes Diabetes Volume 71, January 2022

2. Bender C, Rodriguez-Calvo T, Amirian N, Coppieters KT, von Herrath MG. 24. Skowera A, Ellis RJ, Varela-Calvi~no R, et al. CTLs are targeted to kill beta
The healthy exocrine pancreas contains preproinsulin-specific CD8 T cells that cells in patients with type 1 diabetes through recognition of a glucose-
attack islets in type 1 diabetes. Sci Adv 2020;6:eabc5586 regulated preproinsulin epitope. J Clin Invest 2008;118:3390–3402
3. Lehuen A, Diana J, Zaccone P, Cooke A. Immune cell crosstalk in type 1 25. Szymczak AL, Workman CJ, Wang Y, et al. Correction of multi-gene
diabetes. Nat Rev Immunol 2010;10:501–513 deficiency in vivo using a single ‘self-cleaving’ 2A peptide-based retroviral
4. Herold KC, Gitelman SE, Ehlers MR, et al.; AbATE Study Team. vector. Nat Biotechnol 2004;22:589–594
Teplizumab (anti-CD3 mAb) treatment preserves C-peptide responses in 26. Koulmanda M, Bhasin M, Awdeh Z, et al. The role of TNF-a in mice with
patients with new-onset type 1 diabetes in a randomized controlled trial: type 1- and 2- diabetes. PLoS One 2012;7:e33254
metabolic and immunologic features at baseline identify a subgroup of 27. Jacob CO, Aiso S, Michie SA, McDevitt HO, Acha-Orbea H. Prevention of
responders. Diabetes 2013;62:3766–3774 diabetes in nonobese diabetic mice by tumor necrosis factor (TNF): similarities
5. Roep BO, Peakman M. Antigen targets of type 1 diabetes autoimmunity. between TNF-alpha and interleukin 1. Proc Natl Acad Sci U S A 1990;87:968–972
Cold Spring Harb Perspect Med 2012;2:a007781 28. Driver JP, Racine JJ, Ye C, et al. Interferon-g limits diabetogenic CD81
6. Michels AW, von Herrath M. 2011 update: antigen-specific therapy in T-cell effector responses in type 1 diabetes. Diabetes 2017;66:710–721
type 1 diabetes. Curr Opin Endocrinol Diabetes Obes 2011;18:235–240 29. Vehik K, Cuthbertson D, Ruhlig H, Schatz DA, Peakman M; DPT-1 and
7. Kroll JL, Beam C, Li S, et al.; Type 1 Diabetes TrialNet Anti CD-20 Study TrialNet Study Groups. Long-term outcome of individuals treated with oral
Group. Reactivation of latent viruses in individuals receiving rituximab for new insulin: Diabetes Prevention Trial–Type 1 (DPT-1) oral insulin trial. Diabetes
onset type 1 diabetes. J Clin Virol 2013;57:115–119 Care 2011;34:1585–1590
8. Keymeulen B, Candon S, Fafi-Kremer S, et al. Transient Epstein-Barr 30. Mamchak AA, Manenkova Y, Leconet W, et al. Preexisting autoantibodies
virus reactivation in CD3 monoclonal antibody-treated patients. Blood predict efficacy of oral insulin to cure autoimmune diabetes in combination
2010;115:1145–1155 with anti-CD3. Diabetes 2012;61:1490–1499
9. Sims EK, Bundy BN, Stier K, et al.; Type 1 Diabetes TrialNet Study Group. 31. B€ohm W, Mertens T, Schirmbeck R, Reimann J. Routes of plasmid DNA
Teplizumab improves and stabilizes beta cell function in antibody-positive vaccination that prime murine humoral and cellular immune responses.
high-risk individuals. Sci Transl Med 2021;13:eabc8980 Vaccine 1998;16:949–954
10. Alhadj Ali M, Liu YF, Arif S, et al. Metabolic and immune effects of 32. Bennett PM. Plasmid encoded antibiotic resistance: acquisition and
immunotherapy with proinsulin peptide in human new-onset type 1 diabetes. transfer of antibiotic resistance genes in bacteria. Br J Pharmacol 2008;
Sci Transl Med 2017;9:eaaf7779 153(Suppl. 1):S347–S357
11. Nikolic T, Zwaginga JJ, Uitbeijerse BS, et al. Safety and feasibility of 33. Johansson J, Mandin P, Renzoni A, Chiaruttini C, Springer M, Cossart P.
intra-dermal injection with tolerogenic dendritic cells pulsed with proinsulin An RNA thermosensor controls expression of virulence genes in Listeria
peptide-for type 1 diabetes. Lancet Diabetes Endocrinol 2020;8(6):470–472 monocytogenes. Cell 2002;110:551–561
12. Ludvigsson J, Krisky D, Casas R, et al. GAD65 antigen therapy in recently 34. Cummings HS, Hershey JW. Translation initiation factor IF1 is essential
diagnosed type 1 diabetes mellitus. N Engl J Med 2012;366:433–442 for cell viability in Escherichia coli. J Bacteriol 1994;176:198–205
13. Pham MN, Gibson C, Ryd
en AKE, et al. Oral insulin (human, murine, or 35. H€agg P, de Pohl JW, Abdulkarim F, Isaksson LA. A host/plasmid system
porcine) does not prevent diabetes in the non-obese diabetic mouse. Clin that is not dependent on antibiotics and antibiotic resistance genes for stable
Immunol 2016;164:28–33 plasmid maintenance in Escherichia coli. J Biotechnol 2004;111:17–30
14. Skyler JS, Krischer JP, Wolfsdorf J, et al. Effects of oral insulin in 36. Liu MA, Ulmer JB. Human clinical trials of plasmid DNA vaccines. Adv
relatives of patients with type 1 diabetes: the Diabetes Prevention Trial–Type Genet 2005;55:25–40
1. Diabetes Care 2005;28:1068–1076 37. Solvason N, Lou YP, Peters W, et al. Improved efficacy of a tolerizing DNA
15. Coon B, An L-L, Whitton JL, von Herrath MG. DNA immunization to vaccine for reversal of hyperglycemia through enhancement of gene expression
prevent autoimmune diabetes. J Clin Invest 1999;104:189–194 and localization to intracellular sites. J Immunol 2008;181:8298–8307
16. Roep BO, Solvason N, Gottlieb PA, et al. Plasmid-encoded proinsulin 38. Nakayama M, Abiru N, Moriyama H, et al. Prime role for an insulin
preserves C-peptide while specifically reducing proinsulin-specific CD81 T epitope in the development of type 1 diabetes in NOD mice. Nature 2005;
cells in type 1 diabetes. Sci Transl Med 2013;5:191ra182 435:220–223
17. Bot A, Smith D, Bot S, et al. Plasmid vaccination with insulin B chain 39. Coppieters KT, Dotta F, Amirian N, et al. Demonstration of islet-
prevents autoimmune diabetes in nonobese diabetic mice. J Immunol autoreactive CD8 T cells in insulitic lesions from recent onset and long-term
2001;167:2950–2955 type 1 diabetes patients. J Exp Med 2012;209:51–60
18. Takiishi T, Korf H, Van Belle TL, et al. Reversal of autoimmune diabetes 40. Leete P, Willcox A, Krogvold L, et al. Differential insulitic profiles
by restoration of antigen-specific tolerance using genetically modified determine the extent of b-cell destruction and the age at onset of type 1
Lactococcus lactis in mice. J Clin Invest 2012;122:1717–1725 diabetes. Diabetes 2016;65:1362–1369
19. Robert S, Gysemans C, Takiishi T, et al. Oral delivery of glutamic acid 41. Rodriguez-Calvo T, Ekwall O, Amirian N, Zapardiel-Gonzalo J, von Herrath
decarboxylase (GAD)-65 and IL10 by Lactococcus lactis reverses diabetes in MG. Increased immune cell infiltration of the exocrine pancreas: a possible
recent-onset NOD mice. Diabetes 2014;63:2876–2887 contribution to the pathogenesis of type 1 diabetes. Diabetes 2014;63:3880–3890
20. Yu L, Robles DT, Abiru N, et al. Early expression of antiinsulin 42. Gill RG, Pagni PP, Kupfer T, et al. A preclinical consortium approach for
autoantibodies of humans and the NOD mouse: evidence for early determination assessing the efficacy of combined anti-CD3 plus IL-1 blockade in reversing
of subsequent diabetes. Proc Natl Acad Sci U S A 2000;97:1701–1706 new-onset autoimmune diabetes in NOD mice. Diabetes 2016;65:1310–1316
21. Hellen CUT, Sarnow P. Internal ribosome entry sites in eukaryotic mRNA 43. Quintana FJ, Rotem A, Carmi P, Cohen IR. Vaccination with empty
molecules. Genes Dev 2001;15:1593–1612 plasmid DNA or CpG oligonucleotide inhibits diabetes in nonobese diabetic
22. Goudy KS, Burkhardt BR, Wasserfall C, et al. Systemic overexpression of mice: modulation of spontaneous 60-kDa heat shock protein autoimmunity. J
IL-10 induces CD41CD251 cell populations in vivo and ameliorates type 1 Immunol 2000;165:6148–6155
diabetes in nonobese diabetic mice in a dose-dependent fashion. J Immunol 44. Wolff JA, Malone RW, Williams P, et al. Direct gene transfer into mouse
2003;171:2270–2278 muscle in vivo. Science 1990;247:1465–1468
23. Saraiva M, O’Garra A. The regulation of IL-10 production by immune 45. Gurunathan S, Klinman DM, Seder RA. DNA vaccines: immunology,
cells. Nat Rev Immunol 2010;10:170–181 application, and optimization*. Annu Rev Immunol 2000;18:927–974
diabetesjournals.org/diabetes Pagni and Associates 169

46. Postigo-Fernandez J, Creusot RJ. A multi-epitope DNA vaccine enables a 49. Lee L-F, Xu B, Michie SA, et al. The role of TNF-a in the pathogenesis of
broad engagement of diabetogenic T cells for tolerance in Type 1 diabetes. J type 1 diabetes in the nonobese diabetic mouse: analysis of dendritic cell
Autoimmun 2019;98:13–23 maturation. Proc Natl Acad Sci U S A 2005;102:15995–16000
47. Driver JP, Racine JJ, Ye C, et al. Interferon-g limits diabetogenic 50. Satoh J, Seino H, Abo T, et al. Recombinant human tumor necrosis factor
CD8 1 T-cell effector responses in type 1 diabetes. Diabetes 2017;66: alpha suppresses autoimmune diabetes in nonobese diabetic mice. J Clin
710–721 Invest 1989;84:1345–1348
48. Wang B, Andr
e I, Gonzalez A, et al. Interferon-g impacts at multiple 51. Quattrin T, Haller MJ, Steck AK, et al.; T1GER Study Investigators.
points during the progression of autoimmune diabetes. Proc Natl Acad Sci Golimumab and beta-cell function in youth with new-onset type 1 diabetes. N
U S A 1997;94:13844–13849 Engl J Med 2020;383:2007–2017