page 1 of 50 diabetes...1 baseline assessment of circulating micrornas near diagnosis of type 1...

1

Baseline assessment of circulating microRNAs near diagnosis of type 1 diabetes

predicts future stimulated insulin secretion

Isaac Snowhite1, Ricardo Pastori1,2, Jay Sosenko2, Shari Messinger Cayetano3, and

Alberto Pugliese1,2,4

1Diabetes Research Institute

2Department of Medicine, Division of Endocrinology and Metabolism,

3Department of Public Health Sciences,

4Department of Microbiology and Immunology,

Leonard Miller School of Medicine,

University of Miami,

Miami, Florida

Corresponding AuthorAlberto Pugliese, MDDiabetes Research InstituteLeonard Miller School of Medicine, University of Miami1450 NW 10th Avenue, Miami, FL 33136 USATel. 305-243-5348; Fax 305-243-4404;E-mail: [email protected]

Words: 4,204

of 50 Diabetes

Diabetes Publish Ahead of Print, published online December 4, 2020

2

Abstract

Type 1 diabetes is an autoimmune disease resulting in severely impaired insulin

secretion. We investigated whether circulating microRNAs (miRNAs) are associated with

residual insulin secretion at diagnosis and predict the severity of its future decline. We

studied 53 newly diagnosed subjects enrolled in placebo groups of TrialNet clinical trials.

We measured serum levels of 2,083 miRNAs using RNAseq technology, in fasting

samples from the baseline visit (<100 days from diagnosis), during which residual insulin

secretion was measured with a mixed meal tolerance test (MMTT). Area under the curve

(AUC) C-peptide and peak C-peptide were stratified by quartiles of expression of 31

miRNAs. After adjustment for baseline C-peptide, age, BMI and sex, baseline levels of

miR-3187-3p, miR-4302, and the miRNA combination of miR-3187-3p/miR-103a-3p

predicted differences in MMTT C-peptide AUC/peak levels at the 12-month visit; the

combination miR-3187-3p/miR-4723-5p predicted proportions of subjects above/below

the 200 pmol/L clinical trial eligibility threshold at the 12-month visit. Thus, miRNA

assessment at baseline identifies associations with C-peptide and stratifies subjects for

future severity of C-peptide loss after 1 year. We suggest that miRNAs may be useful in

predicting future C-peptide decline for improved subject stratification in clinical trials.

of 50Diabetes

3

Type 1 diabetes is a chronic autoimmune disease leading to progressive (yet

heterogeneous) loss and dysfunction of pancreatic -cells (1). The TrialNet organization

has completed clinical trials in participants with newly diagnosed type 1 diabetes (<100

days) and stimulated peak C-peptide >200 pmol/L. A 2-year follow-up of 191 participants

in placebo groups demonstrated greater C-peptide loss during the first year post-

diagnosis (2); yet 88% and 66% had stimulated peak C-peptide >200 pmol/L 1 year and

2 years after diagnosis, respectively. Several agents preserve insulin secretion in

individuals with newly diagnosed type 1 diabetes, in some cases for up to 2-7 years after

treatment (3). Novel biomarkers to stratify trial participants at baseline and predict decline

of their insulin secretion would afford gains in trial design and efficiency, facilitating the

identification of effective therapies.

miRNAs are small, non-coding RNAs that regulate gene expression (4) and are

emerging as disease biomarkers. Circulating miRNAs are stable and measurable in

serum and plasma with similar results (5). Twenty-nine circulating miRNAs were

associated with type 1 diabetes by 2-8 studies (supplemental Table S1) (6-24),

suggesting reproducible associations despite heterogeneity in study design, cohorts,

assays, and analysis methods. Cellular miRNAs were also linked to both human and

experimental diabetes (23; 25-30).

An outstanding question is whether circulating miRNAs predict C-peptide decline

after diagnosis. Samandari et al. (12) assessed plasma levels using Exiqon RT-PCR

assays for 179 miRNAs; 3-month visit plasma levels of several miRNAs (miR-24-3p, miR-

146a-5p, miR-194-5p, miR-197-3p, miR-301a-3p and miR-375) correlated with residual

-cell function at the 6- and 12-month visits; miR-197-3p levels at the 3-month visit

of 50 Diabetes

4

predicted -cell function 12 months after diagnosis. Garavelli et al. (24) reported that miR-

23a-3p, miR-23b-3p, miR-24-3p, miR-27a-3p and miR-27b-3p predicted fasting C-

peptide loss <10% or >90% 12 months after diagnosis. In other studies, Let-7g was

associated with C-peptide levels during the first year post-diagnosis (19), levels of the -

cell enriched miR-204 correlated with C-peptide AUC at diagnosis (31), and levels of miR-

142-5p, miR-29c-3p, and miR320 differed in children with recent onset type 1 diabetes

according to residual fasting C-peptide (23).

We assessed levels of 2,083 miRNAs in baseline serum samples from 53

participants with newly diagnosed type 1 diabetes randomized to placebo groups in

TrialNet clinical trials. We report miRNA associations with stimulated C-peptide which are

maintained at the 12-month visit and stratify participants for future severity of C-peptide

loss.

of 50Diabetes

5

MATERIAL AND METHODS

Subjects. We examined baseline serum samples from 53 individuals with newly

diagnosed type 1 diabetes who were enrolled in TrialNet clinical trials as placebos, <100

days from diagnosis and with stimulated peak C-peptide >200 pmol/L. Table 1 shows

their baseline characteristics. Subjects were included in this study based on availability

of a fasting serum sample that could be used for miRNA assessment, obtained on the

day of the baseline MMTT. Supplemental Fig. S1 illustrates C-peptide AUC and peak

levels during the 2-hour MMTTs performed at the baseline, 6-month, and 12-month visits

(2). The baseline MMTT was performed within an average of 1.9 + SD 0.1 months from

diagnosis or 58 days.

MMTT. The MMTT was described previously (32); serum C-peptide levels were

measured using a TOSOH 900 AIA analyzer. The trapezoidal rule was used to calculate

the C-peptide AUC in nmol/L; peak levels are reported in pmol/L (2).

miRNA Assay. We tested 15 l serum aliquots. Samples were collected and

processed according to TrialNet protocols (supplemental methods). miRNAs were

assayed using the HTG Molecular Diagnostics EdgeSeq miRNA assay (33), which

combines a quantitative nuclease protection assay with Next Generation Sequencing. It

does not require miRNA isolation, reverse transcription, adenylation or ligation, which

could introduce bias. The assay has a broad dynamic range with high reproducibility,

sensitivity and specificity (33). Testing of blind replicate samples from 4 individuals with

type 1 diabetes provided by the JDRF Biomarker Working Group Core for Assay

Validation confirmed excellent reproducibility (r=0.94-0.96, CV%=1.06%). Undiluted RNA

is bound to corresponding target-specific nuclease protection probes, after treatment with

of 50 Diabetes

6

lysis buffer. The probe set contains complementary sequences for 2,083 specific miRNAs

or ~78% of the published mature transcripts in miRbase V22 (34). Probes hybridized to

cognate miRNAs are protected from S1 nuclease digestion, amplified with the addition of

barcodes, and sequenced. After amplification the library was quantified according to the

HTG EdgeSeq KAPA Library Quantification protocol for Illumina Sequencing.

Sample Processing and Batch Control. Processing controls include 4 negative and

1 positive control, and a human brain RNA standard. All samples were run as singletons

except the standard was run in triplicate. Samples were randomized before placement

to reduce inter-plate and intra-plate biases, which were assessed using both Pearson and

Spearman correlation coefficients.

Post Sequencing Quality Control. Each well in HTG EdgeSeq assays includes 4

negative and 1 positive control probes with unique sequences. All samples and controls

were quantified in triplicate with the inclusion of no template control (NTC) reactions

during the qPCR process. A PhiX control adaptor-ligated library was spiked into the

pooled library to confirm labeling efficiency and each well was spiked with four unique

plant sequences that are digested by the S1 nuclease during the protection assay.

miRNA Data Management and Analysis. The output is a read count, as in small

RNA-seq, but unlike small RNA-seq, the read count reflects the quantity of probes bound

by miRNAs and protected from digestion. The HTG EdgeSeq Parser aligned the FASTQ

files to the probe list to collate the data. Data tables included raw, QC raw, log2 CPM

(counts per million), and median normalized.

CPM Standardization and Normalization. CPM standardization was used for

evaluation between samples, replicate comparisons, batch effects, and quality control

of 50Diabetes

7

metrics. The Log2 transformation was used for standardizing, or scaling, gene-level data

within a sample. CPM standardization allows the evaluation of probe-level expression as

a proportion of total counts on a sample level and between samples, as described in the

𝑙𝑖𝑚𝑚𝑎 package (35; 36). Inter-sample normalization was achieved by scaling raw read

counts in each lane by a single lane-specific factor reflecting its library size (37). Gene

counts were divided by the median of mapped reads (or library size) associated with their

lane and multiplied by the median total count across all samples. Normalization was

performed using the 𝐷𝐸𝑆𝑒𝑞2 package from Bioconductor and the 𝑅 statistical program

(www.r- project.org).

Statistical analysis. We investigated whether C-peptide levels (fasting, AUC and

peak) were associated with miRNA levels at the baseline MMTT; then, we examined

whether baseline associations were maintained at the 12-month MMTT.

Baseline analysis. We estimated associations between baseline MMTT outcomes

(fasting C-peptide, AUC and peak levels) and miRNA levels. For each miRNA we fit a

linear model to MMTT outcomes, including a nominal indicator of quartile of expression,

with adjustments for age, BMI, and sex. Global F tests of significance determined a list of

miRNAs where variability in baseline MMTT outcome was significantly explained by

miRNA quartile. Bootstrapped resampling with 1,000 replications provided correction for

multiple comparisons, and estimated comparisons among quartiles for differences in

MMTT outcomes. For those miRNAs having significant bootstrapped associations

between expression quartiles and MMTT outcomes, we evaluated specific comparisons

between quartiles for significant differences in MMTT outcomes.

of 50 Diabetes

8

Longitudinal analysis. Those miRNAs with baseline associations with MMTT

outcomes were investigated to evaluate maintained associations with 12-month C-

peptide AUC and peak. This was accomplished by fitting a linear model to each MMTT,

considering quartile of miRNA expression with identified association with baseline MMTT,

adjusting for corresponding baseline C-peptide, age at draw, sex, and BMI. Results

illustrating evidence of longitudinal associations with specific miRNAs were then

considered in stepwise regression to identify those miRNAs which, in combination, have

significant association and best predict 12-month MMTT after adjustment for baseline C-

peptide AUC, age, sex and BMI.

Receiving Operator Curves (ROC). ROC were constructed from predicted values

from logistic regression models fit to binary outcome defined by percentage decline from

baseline (<25% vs >25%). These models were fit with/without miRNA information in

addition to baseline C-peptide AUC and were adjusted for BMI and gender. ROC were

compared to determine whether miRNAs improve prediction of C-peptide decline above

baseline C-peptide AUC using the DeLong’s test for correlated ROC (38).

Longitudinal assessment of MMTT C-peptide/Glucose Response (CGR) curves

after stratification for baseline miRNA levels. We plotted glucose against C-peptide values

(30 to 120 minutes) from baseline, 6-month and 12-month MMTTs for the two groups of

subjects defined by baseline miRNA levels. Changes/differences in the curves position

(shifts to the left, lower C-peptide; upwards, higher glucose), directionality and shape (a

narrower horizontal spread and upward straightening of the curve, or monotonic shape)

illustrates progressive worsening over time and differences between miRNA-stratified

groups. We used the T-test to assess statistical significance in comparisons of baseline

of 50Diabetes

9

mean AUC C-peptide/AUC glucose ratios of the two miRNA expression groups (curves)

to 6- and 12-month ratios, and in comparisons of miRNA-stratified groups (curves) within

each MMTT.

Other analyses. For statistical comparisons involving binary outcomes, groups

were compared using the 2-tailed Fisher’s exact test.

Bioinformatic prediction of putative gene targets and pathways. We

interrogated the reference database KEGG (Kyoto Encyclopedia of Genes and Genomes,

https://www.genome.jp/kegg/) and used miRWalk 2.0 (39) (http://zmf.umm.uni-

heidelberg.de/apps/zmf/mirwalk2/) to identify gene pathways predicted to be targeted by

miRNAs of interest (accessed April-May 2020). miRWalk 2.0 hosts predicted and

experimentally validated miRNA-target interaction pairs, documents miRNA-binding sites

within the complete sequence of a gene and combines this information with a comparison

of binding sites resulting from use of miRanda-rel2010, Targetscan 6.2, miRWalk 2.0 and

RNA22 v2. Statistical significance of these predictions is reported after the Benjamini–

Hochberg correction for multiple comparisons.

Data resource sharing and availability. The datasets generated during and/or

analyzed during the current study were deposited in the Gene Expression Omnibus

(GEO) repository https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE157177). No

applicable resources were generated or analyzed during the current study.

of 50 Diabetes

https://www.genome.jp/kegg/

http://zmf.umm.uni-heidelberg.de/apps/zmf/mirwalk2/

http://zmf.umm.uni-heidelberg.de/apps/zmf/mirwalk2/

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE157177

10

RESULTS

Detection of miRNAs. The EdgeSeq assay collectively detected all 2,083

miRNAs. Detection rates were >95% for all but 4 miRNAs (miR-128-2-5p, 2%; miR-1282,

27%; miR-4525, 6%; miR-6752-3p, 38%). The average detection rate was 2,077 ± 12 SD

miRNAs, which accounts for 99.7% ± SD 0.5% of the panel. The mean normalized log2

CPM values were 7.92 ± SD 0.33 (range:0.5-22.4) for all miRNAs. We observed very

robust raw counts (mean 3.2 ± SD 1.05, range 1.4-7.7 million reads) in 15 L of serum.

Baseline associations. We estimated associations of baseline MMTT outcomes

(fasting C-peptide, AUC, and peak levels) with miRNA expression quartiles. Bootstrapped

resampling with 1,000 replications provided adjusted p values for associated miRNAs, to

correct for multiple comparisons, and estimated comparisons among quartiles for

differences in MMTT outcomes. Among statistically significant associations of C-peptide

with mRNAs after bootstrapping, lowest or highest quartiles vs. the other 3 quartiles were

generally best at discriminating C-peptide differences. Table 2 lists miRNAs and quartile

comparisons that identified significant differences in baseline C-peptide AUC and/or peak:

differences in C-peptide AUC or peak were identified by 25 and 22 miRNAs, respectively,

among which 16 miRNAs had associations with both outcomes, which are naturally

correlated (Table 2A). There were 9 and 6 miRNAs associated with either C-peptide AUC

or peak levels (Table 2B). There were no significant miRNA associations with fasting C-

peptide. miRNA expression quartiles identified baseline C-peptide AUC differences

ranging from 25.92 to 42.3 nmol/L, and peak C-peptide differences ranging from 276 to

442 pmol/L, after bootstrapping. The miRNA that identified the largest C-peptide AUC

and peak differences was miR-3187-3p (42.3 nmol/L and 442 pmol/L, respectively); the

of 50Diabetes

11

previously reported miR-197-3p (6) identified C-peptide AUC and peak differences of

33.58 nmol/L and 351 pmol/L, respectively. All values are reported in Table 2; Fig. 1

shows C-peptide AUC and peak levels according to miRNA expression quartiles for 6

representative miRNAs. Supplemental Table S5 reports total raw counts, raw CPM and

normalized CPM (log2) for the 31 miRNAs associated with C-peptide.

Longitudinal associations. We investigated whether baseline miRNAs predicted

C-peptide at the 12-month MMTT. Specifically, we examined whether any miRNA

associated with baseline C-peptide AUC and/or peak (Table 2) remained associated and

predicted C-peptide AUC and/or peak at the 12-month MMTT, after adjusting for baseline

C-peptide, age, sex, and BMI. Two miRNAs remained associated with both C-peptide

AUC (miR-3187-3p, p=0.037; miR-4302, p=0.047) and peak (miR-3187-3p, p=0.038;

miR-4302, p=0.039) at the 12-month MMTT (Table 3). Baseline expression quartiles of

miR-3187-3p and miR-4302 defined groups of participants with a mean difference in the

12-month C-peptide AUC of 22.49 and 21.05 nmol/L, respectively; mean differences for

12-month peak C-peptide were 236 and 229 pmol/L, respectively. miR-1292-5p was

associated with peak C-peptide only (mean difference=212 pmol/L). Fig. 2 illustrates

longitudinal associations and C-peptide decline for representative miRNAs: miR-3187-3p

and miR-4302 (which stratified participants with significantly different 12-month C-peptide

AUC and peak) and miR-103a-3p and miR-197-3p (which stratified participants only at

baseline).

Stepwise regression modeling to identify combinations of predictive

miRNAs. miRNAs associated with baseline C-peptide AUC were evaluated in a stepwise

regression to identify miRNA combinations with improved prediction of 12-month C-

of 50 Diabetes

12

peptide AUC, after adjustment for baseline C-peptide AUC, age, sex and BMI: 12/25

miRNAs with baseline associations were included in the model. The combination of miR-

3187-3p and miR-103a-3p discriminated C-peptide AUC (Fig. 3A) and peak (Fig. 3B)

levels at 12 months. Stratification according to baseline miRNA expression quartiles for

this combination demonstrated differences in the baseline to 12-month AUC and peak C-

peptide levels of 37.95 nmol/L (p=0.001) and 39 pmol/L (p=0.001) between groups

(supplemental Table S2). Eleven participants with low expression (1st quartile) of miR-

3187-3p combined with high expression (2nd-4th quartile) of miR-103a-3p had higher 12-

month C-peptide AUC compared to the other 42 participants. This combination was

superior to miR-3187-3p alone as 2 more subjects were stratified to the lower C-peptide

group. The miR-3187-3p/miR-4302 combination identified differences in AUC C-peptide

decline of 44.86 nmol/L (p=0.001) (Table S2, Fig. 3C) and assigned two additional

individuals to the lower C-peptide group than miR-3187-3p alone.

We also investigated whether any single miRNA or combination could stratify

participants at the 12-month visit by peak C-peptide levels above/below the clinical trial

eligibility threshold. No individual miRNA from Table 2 was predictive. However, the

combination of miR-3187-3p and miR-4723-5p predicted that 94% (17/18) of the

participants with baseline expression levels in the lower quartile for both miRNAs would

have peak C-peptide >200 pmol/L at the 12-month visit compared to 64% (22/34) of those

with miRNA expression levels in the second to fourth quartiles (p=0.021, Fisher’s exact

test, 2-tailed; relative risk=1.4, 95% C.I.=1.086-1.993; sensitivity=0.4359, 95%

C.I.=0.2930-0.5902; specificity=0.9231, 95% C.I.=0.6669-0.9961).

of 50Diabetes

13

We examined whether miRNAs improved prediction of C-peptide decline

compared to baseline C-peptide AUC. ROC in Fig. 3D illustrates improved ability to

separate between groups with decline greater/lower than 25% when the miR-3187-3p/

miR-103a-3p combination is considered in the model; ROC AUC with/without miRNAs

were 0.82 and 0.70, respectively (p=0.04).

Longitudinal assessment of MMTT CGR curves after stratification for

baseline miRNA levels. CGR curves for baseline, 6-month and 12-month MMTTs

stratified participants into two groups (curves) by their baseline levels of associated

miRNAs (Fig. 4). Curves evolved at 6 months and 12 months, demonstrating progressive

worsening of insulin secretion (shift to the left) and higher glucose levels (shift upwards),

with greater separation of the curves for the miRNA combinations. Overall, subject

stratification by baseline miRNA expression quartiles demonstrated differences in

disease severity at diagnosis which persisted during further progression and involved

both C-peptide and glucose responses.

Longitudinally, we compared baseline mean AUC C-peptide/AUC Glucose ratios

of the two miRNA expression groups (curves) to the ratios of the corresponding curves at

later time points; for example, the ratio from the Q1 curve at baseline was compared to

the ratio of the Q1 curves at 6 month and the same comparison was made between 6

and 12 months. These were all significantly different from each other, demonstrating

worsening in both groups (range p<0.0001-p=0.02, 2-tailed paired T-test, supplemental

Table S3). However, 6-month and 12-month curves of miR-3187-3p/miR-4302 were

statistically different from each other for participants in combination group 0, but not for

those in combination group 1, suggesting that the latter did not experience significant

of 50 Diabetes

14

worsening in this time interval. Cross-sectionally, we compared AUC C-peptide/AUC

Glucose ratios of the two groups (curves) in each panel; these were significantly different

at all time points for all miRNAs or combinations analyzed (range p<0.0001-p=0.0397; T-

test, unpaired, 2-tailed; supplemental Table S4); the only exception were the miR-103a-

3p 12-month curves (p=0.05). The findings suggest significant differences in disease

progression identified by stratification in groups defined by baseline miRNA levels.

Bioinformatic prediction of target gene pathways. We used miRWalk 2.0 (39)

to examine whether any of the 31 miRNAs associated with C-peptide AUC and/or peak

at baseline are predicted to modulate gene pathways relevant to type 1 diabetes. Results

are reported in Table 4, in which we list four major gene pathways relevant to type 1

diabetes and/or type 2 diabetes, specifically the insulin signaling, SNARE interactions in

vesicular transport, type II diabetes mellitus, and the T-cell receptor (TCR) signaling

pathways. Nineteen miRNAs were predicted to target either the insulin or TCR signaling

pathways, and remarkably 9 miRNAs were predicted to target both (miR-103a-3p, miR-

193b-5p, miR-197-3p, miR-3187-3p, miR-4302, miR-622, miR-6748-3p, miR-1208, and

miR-1292-5p). Among the 5 miRNAs which alone or in combination predicted 12-month

C-peptide outcomes (miR-3187-3p, miR-4302, miR-1292-5p, miR-103a-3p, and miR-

4723-5p), 4 targeted both insulin and TCR signaling pathways and potentially may

modulate a large number of genes (71 to 104/139 genes and 65-80/110 genes,

respectively). For miR-3187-3p, the TCR signaling pathway was predicted as the first of

16 pathways.

of 50Diabetes

15

DISCUSSION

Despite the emergence of reproducible associations of circulating miRNAs with

type 1 diabetes, there are limited data about miRNA prediction of C-peptide decline after

diagnosis. Moreover, virtually all published studies of circulating miRNAs in islet

autoimmunity and type 1 diabetes used RT-PCR assays investigating a fraction of the

known miRNAs (supplemental Table S1); Nielsen et al. (6) sequenced pooled samples

to identify differentially expressed miRNAs between individuals with type 1 diabetes and

controls, then assessed levels of 24 miRNAs by RT-PCR. With 2,656 transcripts in

miRbase V22 (34), there is much potential for discovery. Thus, we profiled 2,083 miRNAs

using RNAseq technology. To date, our study has examined the largest number of

miRNAs concerning residual C-peptide at diagnosis.

To investigate whether circulating miRNAs are associated with and predict loss of

insulin secretion after diagnosis, we examined fasting serum samples obtained on the

same day of the baseline MMTT from 53 individuals. Several miRNAs were associated

with C-peptide AUC and/or peak at the baseline MMTT (Table 2, Fig. 1); miRNA

expression quartiles identified participants with better or worse residual insulin secretion,

after adjustment for age, sex, and BMI. The observed differences were not explained by

variation in time from diagnosis to baseline MMTT (not shown). These associations

survived correction for multiple comparisons by bootstrapping.

In longitudinal analyses, baseline levels of 5 of these miRNAs, alone or in

combination, predicted MMTT C-peptide outcomes at the 12-month visit: miR-3187-3p

and miR-4302 predicted C-peptide AUC, miR-3187-3p/miR-103a-3p predicted AUC and

peak, and miR-1292-5p predicted peak C-peptide; . In addition, miR-3187-3p/miR-4723-

of 50 Diabetes

16

5p predicted participants being above or below the peak C-peptide trial eligibility threshold

at the 12-month visit. The miR-3187-3p/miR-103a-3p combination improved prediction of

C-peptide decline compared to baseline C-peptide AUC alone (Fig. 3D). Baseline

differences in C-peptide AUC or peak were maintained on follow-up, after correction for

baseline C-peptide, age, sex and BMI, and decline occurred with similar slopes (Figs. 2

and 3). We cannot discern whether this reflects differences in physical/functional -cell

mass at baseline, in the severity of the autoimmune process, or both.

A prior study of relatives at-risk for type 1 diabetes showed that plotting C-peptide

AUC against glucose AUC values at the time points of the oral glucose tolerance test

(OGTT) helps assessing metabolic impairment during progression to clinical diagnosis

(40). Changes in the curves position, shape and direction demonstrated progressive

worsening during the progression. For the first time, we applied this approach to visualize

these relationships during the MMTT and analyze differences in metabolic responses at

baseline and on follow-up in groups stratified by baseline miRNA levels (Fig. 4).

Participants having higher baseline C-peptide AUC after miRNA stratification had

less pathological curves: their position on the grid indicated higher C-peptide and lower

glucose levels, and their shape indicated more C-peptide secretion relative to glucose

levels. On follow-up, their curves remained distinct from those of the other participants.

The comparisons of the C-peptide AUC/Glucose AUC ratios from baseline ratios

quantified the significant worsening and the differences among the groups stratified by

the baseline miRNA levels persisted over time.

For most of the 31 miRNAs associated with C-peptide at baseline there is no prior

involvement in disease-relevant pathways. This is not surprising given that we identified

of 50Diabetes

17

many miRNAs never before examined in this setting. However, miR-Walk 2.0 predicted

that 19/31 miRNAs may target disease-relevant gene pathways; 7/19 had previous

disease-relevant literature associations (4 with type 1 and type 2 diabetes, 2 with type 2

diabetes only, and 1 with -cell differentiation): remarkably, 19 miRNAs could target either

the insulin or TCR signaling pathways; 9 miRNAs may target both. These included 4/5

miRNAs associated with C-peptide at the 12-month MMTT, alone or in combination (miR-

3187-3p, miR-4302, miR-103a-3p, miR-1292-5p and miR-4723-5p). There were no

previous associations, except for miR-589-5p, for 13 miRNAs with no relevant predictions.

miR-3187-3p had the strongest association with C-peptide. The TCR signaling

pathway ranked first of 16 predicted pathways. It could target genes involved in AKT

(Serine/Threonine Kinase)/PI3K (Phosphatidylinositol 3-kinase) signaling, which is critical

for the development, differentiation and function of effector (41) and regulatory T-cells

(42). Other predicted targets include the CD3 -chain, the MAPK13 and MAPK14 genes

in the mitogen-activated protein kinase signaling pathway, LAT (linker for the activation

of T-cells, SOS1 (son of sevenless factor 1, an exchange factor recruited by LAT), NFAT

(transcription factor nuclear factor of activated T-cells), and the tyrosine phosphatase

CD45 (43).

Plasma levels of miR-103a-3p were increased in individuals with type 1 diabetes

(<5 years duration compared to healthy subjects) (18). We show that higher miR-103a-

3p levels are associated with higher residual insulin secretion near diagnosis, and that

baseline miR-103a-3p levels can aid in predicting C-peptide AUC at 12 months; the

combination miR-3187-3p/miR-103a-3p was the stronger predictor of C-peptide AUC.

This miRNA was linked to type 2 diabetes, obesity, and HFN1A-MODY (44-46). In the

of 50 Diabetes

18

Coronary Diet Intervention with Olive Oil and Cardiovascular Prevention study, low

circulating levels of miR-103a-3p were associated with increased likelihood of type 2

diabetes (47). miR-107 and miR-103a-3p are negative regulators of insulin sensitivity

(48). Validated gene targets of miR-103a-3p include SFRP4 (the secreted frizzled-related

protein 4), which suppresses insulin exocytosis (49), and Cav1 (caveolin-1), which inhibits

insulin signaling by decreasing insulin receptors in caveolae-enriched plasma membrane

domains (48). miR-103a-3p regulates the autophagy gene ATG5 (50), and autophagy

regulates transport-competent secretory peptide precursors, including proinsulin (51). In

our analysis, this miRNA may target both the TCR and insulin signaling pathways.

Other miRNAs were associated with C-peptide at baseline but not on follow-up.

These included miR-197-3p, which in a previous report predicted future C-peptide AUC

(12); the different outcomes may reflect assay type, sample size (most likely), or sample

type (serum vs plasma). miR-197-3p is predicted to target several genes in both the

insulin and TCR signaling pathways. Its plasma levels were reduced in subjects with type

2 diabetes (52).

The gene coding for miR-342-3p on 14q32 contains a cluster of glucose-

responsive miRNAs expressed in pancreatic islet cells (53); miR-342-3p also regulates

the expression of the autoantigen IA-2β (54). We previously reported that miR-342-3p

levels were associated with increased risk of progression to type 1 diabetes among

autoantibody-positive relatives and levels correlated with OGTT outcomes (13); in other

studies, miR-342-3p was differentially expressed in individuals with type 1 diabetes

compared to healthy subjects and at-risk relatives (14); its levels were reduced in

regulatory T-cells in affected individuals compared to healthy subjects (25).

of 50Diabetes

19

miR-127-3p is enriched in human -cells (55) and involved in endocrine

differentiation (56); miR-99a-5p, exhibited increased levels during the first 12 months

post-diagnosis in children with recent onset type 1 diabetes (20) and targets the mTOR

pathway (57). miR-589-5p (58) and miR-193b-5p were associated with type 2 diabetes

and prediabetes, respectively; miR-193b-5p was linked to islet autoimmunity as it was

differentially expressed in autoantibody-positive vs -negative individuals (10).

In closing, trial participants with higher or lower MMTT C-peptide AUC and peak

were stratified by baseline miRNA levels. Selected miRNAs/miRNA combinations

predicted future decline of C-peptide AUC and peak. Predicting future C-peptide at

baseline is critical for subject stratification early after diagnosis when impactful decisions

about trial participation or treatment need to be made. A miRNA combination predicted

12-month C-peptide peak above/below the clinical trial eligibility threshold, which is of

particular importance given increased consideration for trial enrollment up to 2 years from

diagnosis if meeting the peak C-peptide threshold. Many associated miRNAs were

examined for the first time, but some were previously linked to type 1 diabetes; several

are predicted to impact gene pathways relevant to -cell function and T-cells, both critical

to disease pathogenesis. Future studies may explore possible links of miRNAs with

disease endotypes. Limitations of this study are the limited sample size and the lack of a

validation cohort, which require future investigations. These miRNAs are excellent

candidates for validation studies and may become useful biomarkers for advancing

therapeutic discoveries for type 1 diabetes.

of 50 Diabetes

20

ACKNOWLEDGEMENTS

Author contributions. Authors generated (I.S.), analyzed and interpreted data (I.S., J.S.

R.L.P., S.M.C., A.P.) participated in the preparation of the manuscript and approved its

final version (all authors). S.M.C. was responsible for statistical design and analysis plan.

J.S. R.L.P., S.M.C., and A.P conceived the study or parts of the study.

Guarantor statement. A.P. and S.M.C. are the guarantors of this work and, as such, had

full access to all the data in the study and take responsibility for the integrity of the data

and the accuracy of the data analysis.

Conflict of interest. Authors declare no conflict of interest relevant to this study.

Funding. The study supported by JDRF (2-SRA-2015-122-Q-R) and the Diabetes

Research Institute Foundation, Hollywood, Florida, USA. We acknowledge the support of

the Type 1 Diabetes TrialNet Study Group, which identified study participants and

provided samples and follow-up data for this study. The Type 1 Diabetes TrialNet Study

Group is a clinical trials network funded by the National Institutes of Health (NIH) through

the National Institute of Diabetes and Digestive and Kidney Diseases, the National

Institute of Allergy and Infectious Diseases, and The Eunice Kennedy Shriver National

Institute of Child Health and Human Development, through the cooperative agreements

U01 DK061010, U01 DK061034, U01 DK061042, U01 DK061058, U01 DK085453, U01

DK085461, U01 DK085465, U01 DK085466, U01 DK085476, U01 DK085499, U01

of 50Diabetes

21

DK085504, U01 DK085509, U01 DK103180, U01 DK103153, U01 DK103266, U01

DK103282, U01 DK106984, U01 DK106994, U01 DK107013, U01 DK107014, UC4

DK106993, UC4 DK11700901, U01 DK 106693-02, and the JDRF. We acknowledge Dr.

Simi Ahmed (JDRF) for programmatic support. The contents of this article are solely the

responsibility of the authors and do not necessarily represent the official views of the NIH

or the JDRF.

of 50 Diabetes

22

REFERENCES

1. Rodriguez-Calvo T, Richardson SJ, Pugliese A: Pancreas Pathology During the Natural History of Type 1 Diabetes. Current diabetes reports 2018;18:124

2. Greenbaum CJ, Beam CA, Boulware D, Gitelman SE, Gottlieb PA, Herold KC, Lachin JM, McGee P, Palmer JP, Pescovitz MD, Krause-Steinrauf H, Skyler JS, Sosenko JM: Fall in C-peptide during first 2 years from diagnosis: evidence of at least two distinct phases from composite Type 1 Diabetes TrialNet data. Diabetes 2012;61:2066-2073

3. Warshauer JT, Bluestone JA, Anderson MS: New Frontiers in the Treatment of Type 1 Diabetes. Cell metabolism 2020;31:46-61

4. Yates LA, Norbury CJ, Gilbert RJ: The long and short of microRNA. Cell 2013;153:516-519

5. Kroh EM, Parkin RK, Mitchell PS, Tewari M: Analysis of circulating microRNA biomarkers in plasma and serum using quantitative reverse transcription-PCR (qRT-PCR). Methods (San Diego, Calif) 2010;50:298-301

6. Nielsen LB, Wang C, Sørensen K, Bang-Berthelsen CH, Hansen L, Andersen ML, Hougaard P, Juul A, Zhang CY, Pociot F, Mortensen HB: Circulating levels of microRNA from children with newly diagnosed type 1 diabetes and healthy controls: evidence that miR-25 associates to residual beta-cell function and glycaemic control during disease progression. Exp Diabetes Res 2012;2012:896362

7. Latreille M, Herrmanns K, Renwick N, Tuschl T, Malecki MT, McCarthy MI, Owen KR, Rulicke T, Stoffel M: miR-375 gene dosage in pancreatic beta-cells: implications for regulation of beta-cell mass and biomarker development. Journal of molecular medicine (Berlin, Germany) 2015;93:1159-1169

8. Marchand L, Jalabert A, Meugnier E, Van den Hende K, Fabien N, Nicolino M, Madec AM, Thivolet C, Rome S: miRNA-375 a Sensor of Glucotoxicity Is Altered in the Serum of Children with Newly Diagnosed Type 1 Diabetes. Journal of diabetes research 2016;2016:1869082

9. Seyhan AA, Nunez Lopez YO, Xie H, Yi F, Mathews C, Pasarica M, Pratley RE: Pancreas-enriched miRNAs are altered in the circulation of subjects with diabetes: a pilot cross-sectional study. Scientific reports 2016;6:31479

10. Yin CW, M; Li, J; She, R; Zhiu, L; Mi, Q: Serum miRNAs as potential biomarkers for early prediction of type 1 diabetes. Faseb Journal 2016;30

11. Erener S, Marwaha A, Tan R, Panagiotopoulos C, Kieffer TJ: Profiling of circulating microRNAs in children with recent onset of type 1 diabetes. JCI Insight 2017;2:e89656

12. Samandari N, Mirza AH, Nielsen LB, Kaur S, Hougaard P, Fredheim S, Mortensen HB, Pociot F: Circulating microRNA levels predict residual beta cell function and

of 50Diabetes

23

glycaemic control in children with type 1 diabetes mellitus. Diabetologia 2017;60:354-363

13. Snowhite IV, Allende G, Sosenko J, Pastori RL, Messinger Cayetano S, Pugliese A: Association of serum microRNAs with islet autoimmunity, disease progression and metabolic impairment in relatives at risk of type 1 diabetes. Diabetologia 2017;60:1409-1422

14. Akerman L, Casas R, Ludvigsson J, Tavira B, Skoglund C: Serum miRNA levels are related to glucose homeostasis and islet autoantibodies in children with high risk for type 1 diabetes. PloS one 2018;13:e0191067

15. Lakhter AJ, Pratt RE, Moore RE, Doucette KK, Maier BF, DiMeglio LA, Sims EK: Beta cell extracellular vesicle miR-21-5p cargo is increased in response to inflammatory cytokines and serves as a biomarker of type 1 diabetes. Diabetologia 2018;61:1124-1134

16. Grieco GE, Cataldo D, Ceccarelli E, Nigi L, Catalano G, Brusco N, Mancarella F, Ventriglia G, Fondelli C, Guarino E, Crisci I, Sebastiani G, Dotta F: Serum Levels of miR-148a and miR-21-5p Are Increased in Type 1 Diabetic Patients and Correlated with Markers of Bone Strength and Metabolism. Noncoding RNA 2018;4

17. Liu L, Yan J, Xu H, Zhu Y, Liang H, Pan W, Yao B, Han X, Ye J, Weng J: Two Novel MicroRNA Biomarkers Related to beta-Cell Damage and Their Potential Values for Early Diagnosis of Type 1 Diabetes. The Journal of clinical endocrinology and metabolism 2018;103:1320-1329

18. Assmann TS, Recamonde-Mendoza M, Punales M, Tschiedel B, Canani LH, Crispim D: MicroRNA expression profile in plasma from type 1 diabetic patients: Case-control study and bioinformatic analysis. Diabetes research and clinical practice 2018;141:35-46

19. Malachowska B, Wyka K, Nowicka Z, Bartlomiejczyk MA, Mlynarski W, Fendler W: Temporal dynamics of serum let-7g expression mirror the decline of residual beta-cell function in longitudinal observation of children with type 1 diabetes. Pediatric diabetes 2018;19:1407-1415

20. Samandari N, Mirza AH, Kaur S, Hougaard P, Nielsen LB, Fredheim S, Mortensen HB, Pociot F: Influence of Disease Duration on Circulating Levels of miRNAs in Children and Adolescents with New Onset Type 1 Diabetes. Noncoding RNA 2018;4

21. Bertoccini L, Sentinelli F, Incani M, Bailetti D, Cimini FA, Barchetta I, Lenzi A, Cavallo MG, Cossu E, Baroni MG: Circulating miRNA-375 levels are increased in autoantibodies-positive first-degree relatives of type 1 diabetes patients. Acta Diabetol 2019;56:707-710

22. Liu Y, Ma M, Yu J, Ping F, Zhang H, Li W, Xu L, Li Y: Decreased Serum microRNA-21, microRNA-25, microRNA-146a, and microRNA-181a in Autoimmune Diabetes:

of 50 Diabetes

24

Potential Biomarkers for Diagnosis and Possible Involvement in Pathogenesis. Int J Endocrinol 2019;2019:8406438

23. Garavelli S, Bruzzaniti S, Tagliabue E, Prattichizzo F, Di Silvestre D, Perna F, La Sala L, Ceriello A, Mozzillo E, Fattorusso V, Mauri P, Puca AA, Franzese A, Matarese G, Galgani M, de Candia P: Blood Co-Circulating Extracellular microRNAs and Immune Cell Subsets Associate with Type 1 Diabetes Severity. International journal of molecular sciences 2020;21

24. Garavelli S, Bruzzaniti S, Tagliabue E, Di Silvestre D, Prattichizzo F, Mozzillo E, Fattorusso V, La Sala L, Ceriello A, Puca AA, Mauri P, Strollo R, Marigliano M, Maffeis C, Petrelli A, Bosi E, Franzese A, Galgani M, Matarese G, de Candia P: Plasma circulating miR-23~27~24 clusters correlate with the immunometabolic derangement and predict C-peptide loss in children with type 1 diabetes. Diabetologia 2020;

25. Hezova R, Slaby O, Faltejskova P, Mikulkova Z, Buresova I, Raja KR, Hodek J, Ovesna J, Michalek J: microRNA-342, microRNA-191 and microRNA-510 are differentially expressed in T regulatory cells of type 1 diabetic patients. Cell Immunol 2010;260:70-74

26. Roggli E, Gattesco S, Caille D, Briet C, Boitard C, Meda P, Regazzi R: Changes in MicroRNA Expression Contribute to Pancreatic beta-Cell Dysfunction in Prediabetic NOD Mice. Diabetes 2012;61:1742-1751

27. Serr I, Furst RW, Ott VB, Scherm MG, Nikolaev A, Gokmen F, Kalin S, Zillmer S, Bunk M, Weigmann B, Kunschke N, Loretz B, Lehr CM, Kirchner B, Haase B, Pfaffl M, Waisman A, Willis RA, Ziegler AG, Daniel C: miRNA92a targets KLF2 and the phosphatase PTEN signaling to promote human T follicular helper precursors in T1D islet autoimmunity. Proc Natl Acad Sci U S A 2016;113:E6659-e6668

28. Serr I, Scherm MG, Zahm AM, Schug J, Flynn VK, Hippich M, Kalin S, Becker M, Achenbach P, Nikolaev A, Gerlach K, Liebsch N, Loretz B, Lehr CM, Kirchner B, Spornraft M, Haase B, Segars J, Kuper C, Palmisano R, Waisman A, Willis RA, Kim WU, Weigmann B, Kaestner KH, Ziegler AG, Daniel C: A miRNA181a/NFAT5 axis links impaired T cell tolerance induction with autoimmune type 1 diabetes. Science translational medicine 2018;10

29. Scherm MG, Serr I, Zahm AM, Schug J, Bellusci S, Manfredini R, Salb VK, Gerlach K, Weigmann B, Ziegler AG, Kaestner KH, Daniel C: miRNA142-3p targets Tet2 and impairs Treg differentiation and stability in models of type 1 diabetes. Nature communications 2019;10:5697

30. Guay C, Kruit JK, Rome S, Menoud V, Mulder NL, Jurdzinski A, Mancarella F, Sebastiani G, Donda A, Gonzalez BJ, Jandus C, Bouzakri K, Pinget M, Boitard C, Romero P, Dotta F, Regazzi R: Lymphocyte-Derived Exosomal MicroRNAs Promote Pancreatic beta Cell Death and May Contribute to Type 1 Diabetes Development. Cell metabolism 2019;29:348-361.e346

of 50Diabetes

25

31. Xu G, Thielen LA, Chen J, Grayson TB, Grimes T, Bridges SL, Jr., Tse HM, Smith B, Patel R, Li P, Evans-Molina C, Ovalle F, Shalev A: Serum miR-204 is an early biomarker of type 1 diabetes-associated pancreatic beta-cell loss. Am J Physiol Endocrinol Metab 2019;317:E723-e730

32. Greenbaum CJ, Mandrup-Poulsen T, McGee PF, Battelino T, Haastert B, Ludvigsson J, Pozzilli P, Lachin JM, Kolb H: Mixed-meal tolerance test versus glucagon stimulation test for the assessment of beta-cell function in therapeutic trials in type 1 diabetes. Diabetes Care 2008;31:1966-1971

33. Godoy PM, Barczak AJ, DeHoff P, Srinivasan S, Etheridge A, Galas D, Das S, Erle DJ, Laurent LC: Comparison of Reproducibility, Accuracy, Sensitivity, and Specificity of miRNA Quantification Platforms. Cell reports 2019;29:4212-4222.e4215

34. Kozomara A, Birgaoanu M, Griffiths-Jones S: miRBase: from microRNA sequences to function. Nucleic acids research 2019;47:D155-d162

35. Law CW, Chen Y, Shi W, Smyth GK: voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol 2014;15:R29

36. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK: limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic acids research 2015;43:e47

37. Anders S, Pyl PT, Huber W: HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics (Oxford, England) 2015;31:166-169

38. DeLong ER, DeLong DM, Clarke-Pearson DL: Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988;44:837-845

39. Dweep H, Gretz N: miRWalk2.0: a comprehensive atlas of microRNA-target interactions. Nature methods 2015;12:697

40. Sosenko JM, Skyler JS, Herold KC, Schatz DA, Haller MJ, Pugliese A, Cleves M, Geyer S, Rafkin LE, Matheson D, Palmer JP: Slowed Metabolic Decline After 1 Year of Oral Insulin Treatment Among Individuals at High Risk for Type 1 Diabetes in the Diabetes Prevention Trial-Type 1 (DPT-1) and TrialNet Oral Insulin Prevention Trials. Diabetes 2020;69:1827-1832

41. Han JM, Patterson SJ, Levings MK: The Role of the PI3K Signaling Pathway in CD4(+) T Cell Differentiation and Function. Frontiers in immunology 2012;3:245

42. Sauer S, Bruno L, Hertweck A, Finlay D, Leleu M, Spivakov M, Knight ZA, Cobb BS, Cantrell D, O'Connor E, Shokat KM, Fisher AG, Merkenschlager M: T cell receptor signaling controls Foxp3 expression via PI3K, Akt, and mTOR. Proc Natl Acad Sci U S A 2008;105:7797-7802

of 50 Diabetes

26

43. Vignard V, Labbé M, Marec N, André-Grégoire G, Jouand N, Fonteneau JF, Labarrière N, Fradin D: MicroRNAs in Tumor Exosomes Drive Immune Escape in Melanoma. Cancer Immunol Res 2020;8:255-267

44. Luo M, Xu C, Luo Y, Wang G, Wu J, Wan Q: Circulating miR-103 family as potential biomarkers for type 2 diabetes through targeting CAV-1 and SFRP4. Acta Diabetol 2020;57:309-322

45. Zhu H, Leung SW: Identification of microRNA biomarkers in type 2 diabetes: a meta-analysis of controlled profiling studies. Diabetologia 2015;58:900-911

46. Bonner C, Nyhan KC, Bacon S, Kyithar MP, Schmid J, Concannon CG, Bray IM, Stallings RL, Prehn JH, Byrne MM: Identification of circulating microRNAs in HNF1A-MODY carriers. Diabetologia 2013;56:1743-1751

47. Jiménez-Lucena R, Rangel-Zúñiga OA, Alcalá-Díaz JF, López-Moreno J, Roncero-Ramos I, Molina-Abril H, Yubero-Serrano EM, Caballero-Villarraso J, Delgado-Lista J, Castaño JP, Ordovás JM, Pérez-Martinez P, Camargo A, López-Miranda J: Circulating miRNAs as Predictive Biomarkers of Type 2 Diabetes Mellitus Development in Coronary Heart Disease Patients from the CORDIOPREV Study. Molecular therapy Nucleic acids 2018;12:146-157

48. Trajkovski M, Hausser J, Soutschek J, Bhat B, Akin A, Zavolan M, Heim MH, Stoffel M: MicroRNAs 103 and 107 regulate insulin sensitivity. Nature 2011;474:649-653

49. Mahdi T, Hänzelmann S, Salehi A, Muhammed SJ, Reinbothe TM, Tang Y, Axelsson AS, Zhou Y, Jing X, Almgren P, Krus U, Taneera J, Blom AM, Lyssenko V, Esguerra JL, Hansson O, Eliasson L, Derry J, Zhang E, Wollheim CB, Groop L, Renström E, Rosengren AH: Secreted frizzled-related protein 4 reduces insulin secretion and is overexpressed in type 2 diabetes. Cell metabolism 2012;16:625-633

50. Zhang C, Zhang C, Wang H, Qi Y, Kan Y, Ge Z: Effects of miR‑103a‑3p on the autophagy and apoptosis of cardiomyocytes by regulating Atg5. Int J Mol Med 2019;43:1951-1960

51. Riahi Y, Wikstrom JD, Bachar-Wikstrom E, Polin N, Zucker H, Lee MS, Quan W, Haataja L, Liu M, Arvan P, Cerasi E, Leibowitz G: Autophagy is a major regulator of beta cell insulin homeostasis. Diabetologia 2016;59:1480-1491

52. Zampetaki A, Kiechl S, Drozdov I, Willeit P, Mayr U, Prokopi M, Mayr A, Weger S, Oberhollenzer F, Bonora E, Shah A, Willeit J, Mayr M: Plasma microRNA profiling reveals loss of endothelial miR-126 and other microRNAs in type 2 diabetes. CircRes 2010;107:810-817

53. Kameswaran V, Bramswig NC, McKenna LB, Penn M, Schug J, Hand NJ, Chen Y, Choi I, Vourekas A, Won KJ, Liu C, Vivek K, Naji A, Friedman JR, Kaestner KH: Epigenetic regulation of the DLK1-MEG3 microRNA cluster in human type 2 diabetic islets. Cell metabolism 2014;19:135-145

of 50Diabetes

27

54. Abuhatzira L, Xu H, Tahhan G, Boulougoura A, Schaffer AA, Notkins AL: Multiple microRNAs within the 14q32 cluster target the mRNAs of major type 1 diabetes autoantigens IA-2, IA-2beta, and GAD65. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2015;29:4374-4383

55. Klein D, Misawa R, Bravo-Egana V, Vargas N, Rosero S, Piroso J, Ichii H, Umland O, Zhijie J, Tsinoremas N, Ricordi C, Inverardi L, Dominguez-Bendala J, Pastori RL: MicroRNA expression in alpha and beta cells of human pancreatic islets. PLoSONE 2013;8:e55064

56. Jin W, Mulas F, Gaertner B, Sui Y, Wang J, Matta I, Zeng C, Vinckier N, Wang A, Nguyen-Ngoc KV, Chiou J, Kaestner KH, Frazer KA, Carrano AC, Shih HP, Sander M: A Network of microRNAs Acts to Promote Cell Cycle Exit and Differentiation of Human Pancreatic Endocrine Cells. iScience 2019;21:681-694

57. Tsai TF, Lin JF, Chou KY, Lin YC, Chen HE, Hwang TI: miR-99a-5p acts as tumor suppressor via targeting to mTOR and enhances RAD001-induced apoptosis in human urinary bladder urothelial carcinoma cells. Onco Targets Ther 2018;11:239-252

58. Mononen N, Lyytikäinen LP, Seppälä I, Mishra PP, Juonala M, Waldenberger M, Klopp N, Illig T, Leiviskä J, Loo BM, Laaksonen R, Oksala N, Kähönen M, Hutri-Kähönen N, Raitakari O, Lehtimäki T, Raitoharju E: Whole blood microRNA levels associate with glycemic status and correlate with target mRNAs in pathways important to type 2 diabetes. Scientific reports 2019;9:8887

of 50 Diabetes

28

Table 1. Baseline characteristics of 53 Participants with new-onset type 1 diabetes.

______________________________________________________________________

N (M/F) 53 (34/19)

TrialNet Trials N (M/F)

TN02 MMF/DZB 16 (11/5)

TN08 GAD 19 (12/7)

TN09 CTLA-4Ig 8 (6/2)

TN14 Anti-IL-1 Beta 10 (5/5)

Mean ± SD

Age of Diagnosis (Years) 16.8 ± 10.0

Type 1 diabetes duration at MMTT (Months) 1.9 ± 0.1

BMI (kg/m2) 20.8 ± 4.2

HbA1c [mmol/mol (%)] 60 ± 26 (7.6% ± 1.7)

2-Hours MMTT Mean ± SD

Fasting C-Peptide (pmol/L) 364.1 ± 216.3

AUC C-Peptide (nmol/L) 82.2 ± 35.3

Peak C-Peptide (pmol/L) 886.2 ± 385.1

Fasting Glucose (mmol/L) 112.8 ± 29.6

Peak Glucose (mmol/L) 2.7 ± 1.2

______________________________________________________________________

Essential baseline characteristics of the study cohort and baseline MMTT C-peptide

outcomes.

of 50Diabetes

29

Table 2. miRNAs associated with baseline MMTT AUC and/or peak C-peptide.

miRNA Quartiles Comparison

C-peptide AUC Estimated Difference

(nmol/L)P

ValueC-peptide Peak

Estimated Difference (pmol/L)

P Value

miRNAs associated with AUC and peak C-peptidemiR-3187-3p Q2-4 vs Q1 42.30 0.0018 442 0.0070miR-4302 Q2-4 vs Q1 35.19 0.0218 344 0.0065miR-8079 Q2-4 vs Q1 34.46 0.0156 387 0.0258miR-197-3p Q1 vs Q2-4 33.58 0.0150 351 0.0086miR-193b-5p Q2-4 vs Q1 32.58 0.0272 366 0.0178miR-4669 Q2-4 vs Q1 31.75 0.0279 351 0.0419miR-494-5p Q1-3 vs Q4 31.72 0.0233 325 0.0377miR-103a-3p Q1 vs Q2-4 31.44 0.0231 332 0.0560miR-4304 Q1-3 vs Q4 29.83 0.0257 324 0.0360miR-4701-3p Q1-3 vs Q4 29.75 0.0269 272 0.0501miR-98-3p Q2-4 vs Q1 29.23 0.0387 337 0.0480miR-99a-5p Q4 vs Q1-3 28.42 0.0367 312 0.0492miR-3678-3p Q1-3 vs Q4 27.77 0.0292 290 0.0487miR-5682 Q1-3 vs Q4 26.91 0.0436 288 0.0343miR-7154-3p Q1-3 vs Q4 26.55 0.0457 294 0.0194miR-3191-3p Q1-3 vs Q4 26.14 0.0455 276 0.0106

miRNAs associated with C-peptide AUCmiR-8058 Q2-4 vs Q1 32.30 0.0216 - nsmiR-2355-3p Q1-3 vs Q4 32.13 0.0162 - nsmiR-934 Q2-4 vs Q1 28.77 0.0443 - nsmiR-6748-3p Q1 vs Q2-4 28.71 0.0275 - nsmiR-6073 Q1-3 vs Q4 28.33 0.0416 - nsmiR-342-3p Q4 vs Q1-3 26.96 0.0572 - nsmiR-622 Q2-4 vs Q1 26.92 0.0369 - nsmiR-215-5p Q2-4 vs Q1 26.89 0.0563 - nsmiR-568 Q1-3 vs Q4 25.92 0.0456 - ns

miRNAs associated with peak C-peptidemiR-1208 Q2-4 vs Q1 - ns 361 0.0406miR-1292-5p Q2-4 vs Q1 - ns 311 0.0326miR-589-5p Q1-3 vs Q4 - ns 297 0.0159miR-4723-5p Q2-4 vs Q1 - ns 283 0.0260miR-127-3p Q1 vs Q2-4 - ns 282 0.0321miR-6506-5p Q1-3 vs Q4 - ns 281 0.0559

of 50 Diabetes

30

The Table reports estimated differences in C-peptide levels between patient groups

defined by quartiles of miRNA expression. Estimated differences and p values are

corrected for multiple comparisons by bootstrapping analysis. The estimated differences

are those between the quartile comparisons; the quartiles on the left side of the

comparison are those associated with higher C-peptide AUC or peak levels. Either the

highest or lowest quartile of miRNA expression was associated with higher or lower C-

peptide levels compared to the remaining quartiles, which did not differ amongst

themselves. C-peptide AUC and peak levels showed associations with 16 miRNAs, and

miRNAs are ranked by the estimated difference in C-peptide AUC (upper Table); 9

miRNAs were associated with AUC and 6 miRNAs with peak C-peptide, respectively, and

miRNAs are ranked by estimated difference in AUC or peak (middle and lower Table).

of 50Diabetes

31

Table 3. miRNAs assessed at baseline with association with C-peptide AUC

and/or peak at the 12-month MMTT.

miRNA Quartile Comparison

C-peptide AUC Estimated Difference (nmol/L)

P Value

C-peptide Peak Estimated Difference (pmol/L)

P Value

miR-3187-3p Q2-4 vs Q1 22.49 0.0371 236 0.0383

miR-4302 Q2-4 vs Q1 21.05 0.0479 229 0.0391

miR-1292-5p Q2-4 vs Q1 n/a n/a 212 0.0496

The Table reports estimated differences in the 12-month MMTT C-peptide levels between

participant groups defined by baseline quartiles of miRNA expression. The estimated

differences are those between the quartile comparisons, as described in the legend of

Table 2.

of 50 Diabetes

32

Table 4. Prediction of targeted gene pathways by miRNAs associated with

baseline MMTT AUC and/or peak C-peptide.

miRNAPredicted Disease Relevant KEGG Pathways

Predicted Gene Targets/ Genes in

PathwayPathway Ranking P Value

miR-98-3p Insulin signaling 63/139 17 of 18 0.0320miR-99a-5p Insulin signaling 34/139 5 of 6 0.0569

Insulin signaling 104/139 2 of 20 8.33E-06miR-103a-3pTCR signaling 80/110 11 of 20 0.0018

miR-127-3p Insulin signaling 80/139 5 of 21 0.0002TCR signaling 94/110 13 of 27 0.0005miR-193b-5pInsulin signaling 115/139 16 of 27 0.0013TCR signaling 59/110 8 of 18 0.0046miR-197-3pInsulin signaling 71/139 10 of 18 0.0064

miR-2355-3p Insulin signaling 84/139 14 of 14 0.0469miR-342-3p Insulin signaling 90/139 1 of 23 5.00E-06

TCR signaling 82/110 15 of 33 0.0002Insulin signaling 100/139 16 of 33 0.0003miR-622Type II diabetes mellitus 38/49 31 of 33 0.0381

miR-934 Insulin signaling 76/139 5 of 17 0.0017Insulin signaling 91/139 4 of 27 5.14E-05TCR signaling 72/110 14 of 27 0.0008miR-1208Type II diabetes mellitus 36/49 17 of 27 0.0049Insulin signaling 91/139 10 of 22 0.0008miR-1292-5pTCR signaling 71/110 17 of 22 0.0171Insulin signaling 101/139 1 of 16 4.76E-06miR-3187-3pTCR signaling 74/110 12 of 16 0.0230Insulin signaling 85/139 6 of 19 0.0001miR-4302TCR signaling 65/110 12 of 19 0.0094

miR-4304 Insulin signaling 57/139 3 of 15 0.0001Insulin signaling 71/139 2 of 20 3.40E-06miR-4723-5pType II diabetes mellitus 28/49 16 of 20 4.72E-03TCR signaling 49/110 9 of 21 0.0005miR-6748-3pInsulin signaling 57/139 12 of 21 0.0021Insulin signaling 76/139 3 of 12 0.0008

miR-7154-3p SNARE interactions in vesicular transport 26/29 8 of 12 0.0160

miR-8058 Insulin signaling 60/139 5 of 12 0.0002

of 50Diabetes

33

Disease-relevant gene pathways predicted to be targeted by miRNAs associated with C-

peptide in the primary analysis. Number of predicted genes, ranking of the reported

pathways and corrected p values are also shown.

of 50 Diabetes

34

FIGURE LEGENDS

Fig. 1. Baseline miRNA quartile levels association with baseline C-peptide AUC and

peak levels. The figure illustrates individual C-peptide AUC and peak levels according to

miRNA expression quartiles. Data shown are mean + SD. Data are shown for 6

representative miRNAs. Differences between groups and p values after bootstrapping are

reported from Table 2.

Fig. 2. Baseline miRNA levels association with 12-month C-peptide AUC and peak

levels. The dot plots illustrate C-peptide AUC and peak levels at the 12-month visit, for

participants stratified by baseline miRNA expression quartiles. Their C-peptide AUC and

peak values are shown from baseline to the 12-month visit, and slope values are reported.

Data shown are mean + SD. Data are shown for 4 representative miRNAs: the baseline

associations of miR-3187-3p and miR-4302 with C-peptide values remained significant at

the 12-month visit, for which differences between groups and p values are reported from

Table 3. The baseline levels of miR-103a-3p and miR-197-3p were not associated with

statistically significant C-peptide differences at the 12-month visit.

Fig. 3. Combined baseline levels of selected miRNAs predict C-peptide at the

12-month visit. 12-month AUC (A, C) and peak C-peptide (B) in participant groups

defined by baseline combined levels of miR-3187-3p and miR-103a-3p (A, B) or miR-

3187-3p and miR-4302 (C). The Whisker boxes show median, 1st and 4th quartiles, and

maximum/minim values. The dot plots show individual values, mean and SD. Black

circles in the dot plots mark subjects with lower C-peptide outcome uniquely identified by

the miRNA combinations. By multivariable analysis, the miR-3187-3p/miR-103a-3p

combination identified differences between groups in AUC and peak C-peptide decline

of 50Diabetes

35

from baseline to 12 months of 37.95 nmol/L (p=0.001) and 39 pmol/L (p=0.001),

respectively. Likewise, the miR-3187-3p/miR-4302 combination identified differences

between groups in AUC C-peptide decline of 44.86 nmol/L (p=0.001). In panels A and B,

combination 0=Q1 miR-103a-3p + Q2-4 miR-3187-3p, combination 1=Q2-4 miR-103a-3p

+ Q1 miR-3187-3p; in panel C, combination 0=Q2-4 miR-3187-3p + Q2-4 miR-4302,

combination 1=Q1 miR-3187-3p + Q1 miR-4302. Results of these analyses are reported

in supplemental Table S2. Panel D illustrates the improved ability to separate between

groups with C-peptide decline lesser or greater than 25% when the miR-3187-3p/miR-

103a combination is considered in the model. ROC AUC in the model with and without

miRNAs were 0.82 and 0.70, respectively (p=0.04).

Fig. 4. MMTT CGR curves after stratification for baseline miRNA expression

quartiles. The curves plot the mean values at 30, 60, 90, and 120 minutes (left to right)

for the baseline, 6-month and 12-month MMTTs. CGR curves are shown for miR-3187-

3p, miR-103a-3p, miR-4302, and the combination of miR-3187-3p with miR-103a-3p

(combination 0=Q1 miR-103a-3p + Q2-4 miR-3187-3p, combination 1=Q2-4 miR-103a-

3p + Q1 miR-3187-3p) and the combination of miR-3187-3p with miR-4302 (combination

0=Q2-4 miR-3187-3p + Q2-4 miR-4302, combination 1=Q1 miR-3187-3p + Q1 miR-

4302). Participants were stratified for baseline expression levels of associated miRNAs;

those in the Q1 group had a more monotonic shape of the CGR curves, which is located

upward and to the left of participants in the Q2-Q4 groups. The 6- and 12-month panels

shows the metabolic deterioration over time in both groups as is evident by the increasing

monotonicity in both and by the upward and leftward movement of the CGR curves.

of 50 Diabetes

p=0.0018 p=0.0070 p=0.0218 p=0.0065

p=0.0156 p=0.0258 p=0.0150 p=0.0086

p=0.0272 p=0.0178 p=0.0231 p=0.0560

of 50Diabetes

C-peptide AUC C-peptide Peak

p=0.0371 p=0.0383

p=0.0479 p=0.0391

p=ns p=ns

p=ns p=ns

of 50 Diabetes

C

A

B

Sensitivity

Specificity

- - with miRNA (AUC=0.82)… without miRNA (AUC=0.72)

p=0.04

p=0.001 p=0.001

p=0.001 p=0.001

p=0.001 p=0.001

D

of 50Diabetes

of 50 Diabetes

Snowhite et al.

1

Supplementary Information

Supplementary Table S1





Supplementary Fig. S1

Supplementary Material

Supplementary Fig. S2

of 50Diabetes

Snowhite et al.

2

Table S1. Circulating miRNAs associated with type 1 diabetes by at least 2 independent studies.

Table S1. Circulating miRNAs associated with T1D by at least 2 independent studies published (2012-2020).

Control At-risk Associated miRNAs (n=30)T1D

First author Year Study DesignNew

Onset(Y/N)

n(meanage)

n(meanage)

n(meanage)

Sample type Assay type

Numberof

miRNAsstudied

miR

-24-

3pm

iR-3

75m

iR-2

5-3p

miR

-146

a-5p

miR

-21-

5ple

t-7g-

5pm

iR-1

48a-

3pm

iR-1

81a-

5pm

iR-2

6b-5

pm

iR-2

9a-3

pm

iR-3

0e-5

pm

iR-1

03a-

3pm

iR-1

06a-

5pm

iR-1

39-5

pm

iR-1

40-5

pm

iR-1

44-5

pm

iR-1

52-3

pm

iR-1

6-5p

miR

-342

-3p

miR

-19a

-3p

miR

-20a

-5p

miR

-200

a-3p

miR

-21-

3pm

iR-2

10-3

pm

iR-2

22-3

pm

iR-2

7b-3

pm

iR-3

0c-5

pm

iR-3

4a-5

pm

iR-4

51a

miR

-93-

5p

Nielsen et al. 2012 Case-Control, C-peptide levels Y 275 (12) 151 n/a serum Sequencing/qPCR 240/47 X X X X X X X X X X X XLatreille et al. 2015 Case-Control N 38 (43.6) 51 (40.8) n/a plasma TLDA qPCR 1 XMarchand et al. 2016 Case-Control Y 22 (9.8) 10 (9.9) n/a serum TLDA qPCR 1 XSeyhan et al. 2016 Case-Control N 16 (25.9) 27 (25.3) n/a plasma TLDA qPCR 28 X X X X XYin et al. 2016 At-risk relatives n/a n/a n/a 35 (n/a) serum TLDA qPCR 754 X XErener et al. 2017 Case-Control Y 38 (8.9) 32 (8.8) n/a plasma Exiqon LNA qPCR 745 X X X X X X X XSamandari et al. 2017 C-peptide levels Y 40 (8.7) n/a n/a plasma Exiqon LNA qPCR 745 X X X XSnowhite et al. 2017 At-risk relatives n/a n/a n/a 150 (11) serum Exiqon LNA qPCR 93 X X X X X X X X XAkerman et al. 2018 Case-Control, At-risk relatives Y 8 (11.7) 17 (11.8) 21 (10.2) serum Exiqon LNA qPCR 179 X X X X X X X X X X X X X X X X X X X X X X XLakhter et al. 2018 Case-Control Y 19 (10.5) 16 (10.5) n/a serum/exosomes Digital droplet PCR 1 X X X XGrieco et al. 2018 Case-Control N 15 (32) 14 (28) n/a serum TLDA qPCR 6 X XLiu et al. 2018 Case-Control Y 73 (22) 85 (21) n/a serum TLDA qPCR 6 XAssmann et al. 2018 Case-Control N 33 (19.5) 26 (21.5) 0 plasma TLDA qPCR 45 X X X XMałachowska et al. 2018 Case-Control Y 9 (n/a) 10 (n/a) n/a serum Exiqon LNA qPCR 752 X XSamandari et al. 2018 C-peptide levels Y 40 (8.7) n/a n/a serum/plasma Exiqon LNA qPCR 179 XBertoccini et al. 2019 Case-Control, At-risk relatives Y 49 (16.1) 48 (41.2) 46 (26.8) serum TLDA qPCR 1 XLiu et al. 2019 Case-Control N 29 (24) 19 (30) n/a serum In house qPCR 4 X X X XGaravelli et al. 2020 Case-Control, C-peptide levels Y 88 (8.9) 47 (8.4) n/a plasma Exiqon LNA qPCR 60 XGaravelli et al. 2020 Case-Control, C-peptide levels Y 88 (8.9) 47 (8.4) n/a plasma Exiqon LNA qPCR 60 X

8 6 6 6 4 3 3 3 3 3 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2TOTAL STUDIES REPORTING

In a review of the published literature last updated on October 29, 2020 we find that circulating levels of 30 miRNAs were associated with type 1 diabetes by at least 2 published studies and 11 miRNAs were reported by at least 3 studies, as indicated in the last row of the table. These 11 miRNAs are: miR-24-3p, miR-375, mir-25-3p, miR-146-3p, miR-21-5p, let-7g-5p, miR-148a-3p, miR-181a-5p, miR-26b-5p, miR-29a-3p, and miR-30e-5p. This Table only includes 19 studies examining association of circulating miRNAs in autoantibody-positive at-risk relatives (4 studies), case-control studies (n=15), or investigations of miRNAs in relation to C-peptide levels after the onset of type 1 diabetes (n=5). Virtually all studies used RT-PCR assays and the number of miRNAs examined ranged between 1 and 754, with 13/19 studies examining fewer than 100 miRNAs, 7<10 miRNAs; Nielsen et al (2012) used sequencing of pooled samples and then RT-PCR to assess levels of 47 miRNAs. All papers listed in this table are referenced in the manuscript.

of 50 Diabetes

Snowhite et al.

3

Table S2. Combinations of miRNAs predict change in C-peptide AUC and Peak between

baseline and 12 months.

______________________________________________________________________

Estimate Std. Error t value Pr(>|t|)Intercept -39.5511 20.95 -1.888 0.06551BMI 2.6036 1.1169 2.331 0.02429 *Age at draw 0.3008 0.4802 0.626 0.53419Sex -3.8242 8.0189 -0.477 0.63574Baseline C-Peptide AUC -0.7337 0.1314 -5.583 1.3e-06 ***COMBINATION 103a-3p/3187-3p 37.9503 11.3019 3.358 0.0016 **

Multiple R-squared: 0.4418, Adjusted R-squared: 0.3798 F-statistic: 7.124 on 5 and 45 DF, p-value: 5.551e-05Baseline-12 months C-peptide AUC difference: 37.9 mmol/l

Estimate Std. Error t value Pr(>|t|)Intercept -0.404444 0.220724 -1.832 0.07352BMI 0.024643 0.011931 2.065 0.04467 *Age at draw 0.00538 0.005116 1.052 0.29854Sex -0.031963 0.084778 -0.377 0.70793Baseline C-Peptide Peak -0.73719 0.130452 -5.651 1.03e-06 ***COMBINATION 103a-3p/3187-3p 0.39958 0.120168 3.325 0.00176 *

Multiple R-squared: 0.4413, Adjusted R-squared: 0.3792F-statistic: 7.108 on 5 and 45 DF, p-value: 5.67e-05Baseline-12 months C-peptide Peak difference: 0.39 mmol/l

Estimate Std. Error t value Pr(>|t|)Intercept -38.1881 20.8928 -1.828 0.07421BMI 2.6424 1.1132 2.374 0.02194 *Age at draw 0.3338 0.4759 0.701 0.48666Sex -4.2175 7.9992 -0.527 0.60062Baseline C-Peptide AUC -0.7489 0.1326 -5.65 1.03e-06 ***COMBINATION 3187-3p/4302 44.8617 13.0642 3.434 0.00129 **

Multiple R-squared: 0.4469, Adjusted R-squared: 0.3855F-statistic: 7.272 on 5 and 45 DF, p-value: 4.59e-05Baseline-12 months C-peptide AUC difference: 44.86 mmol/L

______________________________________________________________________

The table reports the detailed results of the analysis for the data shown in Fig. 3.

of 50Diabetes

Snowhite et al.

4

Table S3. Longitudinal comparisons of AUC C-peptide/Glucose ratios.

miR-3187-3p miR-3187-3p miR-3187-3pQ1: Baseline Q1: 6-Month Q1: Baseline Q1: 12-Month Q1: 6-Month Q1: 12-Month

N 13 13 13 13 13 13Mean + SD 1.06 + 0.30 0.09 + 0.05 1.06 + 0.30 0.07 + 0.05 0.09 + 0.06 0.07 + 0.05

Paired T-test p=0.0002 p=0.0002 p=0.0063Q2-4: Baseline Q2-4: 6-Month Q2-4: Baseline Q2-4: 12-Month Q2-4: 6-Month Q2-4: 12-Month

N 35 35 39 39 35 35Mean + SD 0.57 + 0.26 0.04 + 0.03 0.58 + 0.26 0.03 + 0.02 0.04 0.03 + 0.02

Paired T-test p<0.0001 p<0.0001 p<0.0001

miR-103a-3p miR-103a-3p miR-103a-3p Q1: Baseline Q1: 6-Month Q1: Baseline Q1: 12-Month Q1: 6-Month Q1: 12-Month

N 12 12 13 13 12 12Mean + SD 0.55 + 0.34 0.03 + 0.03 0.38 + 0.34 0.02 + 0.01 0.03 + 0.02 0.01 + 0.01

Paired T-test p=0.0005 p=0.0002 p=0.0256 Q2-4: Baseline Q2-4: 6-Month Q2-4: Baseline Q2-4: 12-Month Q2-4: 6-Month Q2-4: 12-Month

N 36 36 39 39 35 35Mean 0.76 + 0.33 0.06 + 0.05 0.75 + 0.32 0.04 + 0.03 0.06 + 0.06 0.04 + 0.04

Paired T-test p<0.0001 p<0.0001 p=0.0007

3187-3p/miR-103a-3p 3187-3p/miR-103a-3p 3187-3p/miR-103a-3p 1: Baseline 1: 6-Month 1: Baseline 1: 12-Month 1: 6-Month 1: 12-Month

N 11 11 11 11 11 11Mean + SD 1.05 + 0.29 0.09 + 0.06 1.05 + 0.05 0.10 + 0.05 0.09 + 0.06 0.10 + 0.06

Paired T-test p=0.0010 p=0.0010 p=0.0264 0: Baseline 0: 6-Month 0: Baseline 0: 12-Month 0: 6-Month 0: 12-Month

N 37 37 41 41 36 36Mean + SD 0.60 + 0.29 0.04 + 0.03 0.61 + 0.29 0.03 + 0.02 0.04 + 0.03 0.03 + 0.01

Paired T-test p<0.0001 p<0.0001 p=0.0052 miR-4302 miR-4302 miR-4302

Q1: Baseline Q1: 6-Month Q1: Baseline Q1: 12-Month Q1: 6-Month Q1: 12-MonthN 13 13 13 13 13 13

Mean + SD 0.90 + 0.37 0.08 + 0.06 0.90 + 0.37 0.06 + 0.04 0.08 + 0.06 0.06 + 0.04Paired T-test p=0.0002 p=0.0005 p=0.0017

Q2-4: Baseline Q2-4: 6-Month Q2-4: Baseline Q2-4: 12-Month Q2-4: 6-Month Q2-4: 12-MonthN 35 35 39 39 35 35

Mean + SD 0.63 + 0.30 0.04 + 0.04 0.63 + 0.30 0.03 + 0.03 0.04 + 0.04 0.03 + 0.03Paired T-test p<0.0001 p<0.0001 p<0.0001 miR-3187-3p/4302 miR-3187-3p/4302 miR-3187-3p/4302

1: Baseline 1: 6-Month 1: Baseline 1: 12-Month 1: 6-Month 1: 12-MonthN 8 8 8 8 8 8


0: Baseline 0: 6-Month 0: Baseline 0: 12-Month 0: 6-Month 0: 12-MonthN 40 40 44 44 39 39

Mean + SD 0.63 0.04 0.64 0.03 0.04 0.03Std.

Deviation 0.30 0.04 0.29 0.03 0.04 0.03Paired T-test p<0.0001 p<0.0001 p=0.0122

of 50 Diabetes

Snowhite et al.

5

Table S3 reports full results of the longitudinal comparisons of the AUC C-peptide/Glucose ratios from the curves shown in Fig. 4, for the indicated quartiles of individual miRNAs, or for combinations of miRNAs, in which case the combined quartiles of expression are indicated as “0” or “1”. For miR-3187-3p/miR-103a-3p, combination 0= Q1 miR-103a-3p + Q2-4 miR-3187-3p, combination 1= Q2-4 miR-103a-3p + Q1 miR-3187-3p; for miR-3187-3p/miR-4302, 0= Q2-4 miR-3187-3p + Q2-4 miR-4302, combination 1= Q1 miR-3187-3p + Q1 miR-4302. Statistically significant changes occur for each of the two groups of patients defined by miRNA levels, consistent with the disease natural history. However, 6-month and 12-month curves of miR-3187-3p/miR-4302 were statistically different from each other for participants in the combination group 0, but not for those in combination group 1, suggesting that the latter did not experience significant worsening in this time interval.

of 50Diabetes

Snowhite et al.

6

Table S4. Cross-sectional comparisons of AUC C-peptide/Glucose ratios.

miR-3187-3p miR-3187-3p miR-3187-3p Q1: Baseline Q2-4: Baseline Q1: 6-Month Q2-4: 6-Month Q1: 12-Month Q2-4: 12-Month

N 13 40 13 35 13 39Mean + SD 1.06 + 0.30 0.59 + 0.26 0.09 + 0.06 0.05 + 0.03 0.07 + 0.05 0.03 + 0.02

Paired T-test p<0.0001 p=0.0004 p=0.0007

miR-103a-3p miR-103a-3p miR-103a-3p Q1: Baseline Q2-4: Baseline Q1: 6-Month Q2-4: 6-Month Q1: 12-Month Q2-4: 12-Month

N 13 40 12 36 13 39Mean + SD 0.38 + 0.64 0.60 + 0.93 0.03 + 0.02 0.06 + 0.01 0.02 + 0.01 0.05 + 0.04

Paired T-test p=0.0140 p=0.0303 p=0.0557 3187-3p/miR-103a-3p 3187-3p/miR-103a-3p 3187-3p/miR-103a-3p

1: Baseline 0: Baseline 1: 6-Month 0: 6-Month 1: 12-Month 0: 12-MonthN 11 42 11 37 11 41

Mean + SD 1.05 + 0.29 0.61 + 0.28 0.09 + 0.06 0.04 + 0.03 0.07 + 0.05 0.03 + 0.02Paired T-test p=0.0001 p=0.0005 p=0.0004 miR-4302 miR-4302 miR-4302

Q1: Baseline Q2-4: Baseline Q1: 6-Month Q2-4: 6-Month Q1: 12-Month Q2-4: 12-MonthN 13 40 13 35 13 39


miR-3187-3p/4302 miR-3187-3p/4302 miR-3187-3p/4302 1: Baseline 0: Baseline 1: 6-Month 0: 6-Month 1: 12-Month 0: 12-Month

N 8 45 8 40 8 44Mean + SD 1.08 + 0.31 0.63 + 0.29 0.10 + 0.06 0.04 + 0.04 0.07 + 0.04 0.03 + 0.03

Paired T-test p=0.0007 P<0.0001 p=0.0003

Table S4 reports full results of the cross-sectional comparison of the AUC C-peptide/Glucose ratios from the curves shown in Fig. 4, for the indicated quartiles of individual miRNAs, or for combinations of miRNAs, in which case the combined quartiles of expression are indicated as “0” or “1”. For miR-3187-3p/miR-103a-3p, combination 0= Q1 miR-103a-3p + Q2-4 miR-3187-3p, combination 1= Q2-4 miR-103a-3p + Q1 miR-3187-3p; for miR-3187-3p/miR-4302, 0= Q2-4 miR-3187-3p + Q2-4 miR-4302, combination 1= Q1 miR-3187-3p + Q1 miR-4302. The findings suggest significant differences in disease progression identified by stratification in groups defined by baseline miRNA levels.

of 50 Diabetes

Snowhite et al.

7

Table S5. Total raw counts, raw CPM, and normalized CPM (log2) observed for the 31 miRNAs associated with C-peptide AUC and or peak in the primary analysis.

Total Raw Counts Raw CPM Normalized CPM (log2)miRNA Average Median SD Average Median SD Average Median SDmiR-103a-3p 708.89 552 601.22 216.09 194.99 125.07 7.43 7.54 0.84miR-1208 455.15 235 649.40 129.18 80.12 119.64 6.17 6.31 1.65miR-127-3p 80.19 54 108.08 23.36 18.58 17.55 4.04 4.17 1.06miR-1292-5p 203.58 113 266.23 59.30 42.46 52.42 5.13 5.34 1.47miR-193b-5p 272.943 183 311.91 80.59 69.30 54.68 5.85 5.87 0.99miR-197-3p 563.91 477 386.48 174.93 163.45 72.95 7.23 7.29 0.65miR-215-5p 400.75 204 555.05 115.18 78.34 105.67 6.07 6.26 1.41miR-2355-3p 417.66 227 618.02 119.82 85.13 116.27 6.11 6.13 1.35miR-3187-3p 255.96 126 367.77 72.39 49.84 65.36 5.43 5.62 1.36miR-3191-3p 276.79 173 342.07 82.02 55.49 73.73 5.61 5.59 1.35miR-342-3p 484.98 414 497.72 144.52 126.64 93.03 6.69 6.89 1.09miR-3678-3p 362.25 199 525.68 102.85 66.30 96.02 5.92 6.00 1.36miR-4302 390.58 222 499.95 113.51 71.15 93.65 6.23 6.08 1.15miR-4304 443.13 272 503.33 131.40 92.54 97.79 6.51 6.38 1.09miR-4669 404.08 222 505.62 118.02 69.82 96.81 6.32 6.02 1.10miR-4701-3p 434.49 279 588.86 124.72 92.29 109.44 6.29 6.29 1.20miR-4723-5p 247.15 124 361.32 70.25 44.24 67.29 5.34 5.22 1.39miR-494-5p 323.17 173 480.73 91.69 59.38 86.41 5.74 5.84 1.42miR-568 435.66 181 653.15 125.12 74.24 127.33 5.92 5.99 1.69miR-5682 366.94 182 545.68 105.19 62.00 103.12 5.75 5.82 1.69miR-589-5p 423.66 238 557.65 123.11 83.55 107.92 6.24 6.16 1.30miR-6073 404.34 176 608.82 116.03 71.13 118.42 5.83 5.85 1.85miR-622 397.32 222 522.23 114.30 81.30 95.62 6.24 6.23 1.14miR-6506-5p 395.55 186 563.44 112.87 71.10 104.45 6.04 6.07 1.35miR-6748-3p 509.98 390 474.48 155.39 125.36 92.58 6.88 6.94 0.99miR-7154-3p 472.23 240 646.18 137.49 89.10 124.78 6.36 6.36 1.35miR-8058 342.83 154 499.62 98.79 63.17 96.36 5.70 5.84 1.58miR-8079 379.72 213 504.70 110.03 72.06 92.16 6.17 6.10 1.22miR-934 270.96 113 426.06 76.84 41.68 81.73 5.25 5.33 1.66miR-98-3p 191.42 94 267.90 56.39 35.23 54.12 4.95 4.95 1.51miR-99a-5p 535.47 383 530.25 159.66 131.07 94.54 6.96 7.00 0.89

Average 382.3 226.5 482.9 111.6 79.4 91.5 6.0 6.0 1.3Median 397.3 204.0 504.7 114.3 71.1 95.6 6.1 6.1 1.3SD 122.8 112.0 128.5 37.8 36.8 25.0 0.7 0.7 0.3Quartile 25% 300.0 173.0 406.3 86.9 60.7 77.7 5.7 5.8 1.1Quartile 75% 439.4 239.0 560.5 127.2 87.1 106.8 6.3 6.3 1.4

of 50Diabetes

Snowhite et al.

8

Fig. S1. C-peptide AUC Decline in the Study Cohort

C-peptide AUC and peak levels for individual study participants (panel A) and as means+ SD (panel B) at the baseline, 6-month and 12-month MMTT. In panel B, statisticaldifferences among the time points demonstrate significant C-peptide decline (Mann-Whitney test). Panel C illustrates the statistically significant decline of C-peptide AUCand peak observed in the cohort from baseline to the 12-month MMTT; significance wasestimated using the Wilcoxon matched pairs signed rank test. Data are shown for 52/53participants since a single subject did not have 12-month MMTT data; at 6 months, 4subjects did not have MMTT data. Data are plotted on a Log2 scale.

Baseline 6 Month 12 Month1248

163264

128256

C-p

eptid

eA

UC

(nm

ol/L

)

N=52 N=49 N=52

Baseline 6 Month 12 Month1

4

16

64

256

1024

4096

Peak

C-p

eptid

e(p

mol

/L)

N=52 N=49 N=52

C-peptide AUC C-peptide PeakA

B

C

Baseline 12-Month0

50

100

150

200

C-p

eptid

eA

UC

(nm

ol/L

) p<0.0001

Baseline 12-Month8

163264

128256512

10242048

Peak

C-p

eptid

e(p

mol

/L)

p<0.0001

Baseline 6 Months 12 Months1248

163264

128256

C-p

eptid

eA

UC

(nm

ol/L

)

N=52 N=49 N=52

p=0.0028

p<0.0001

p<0.0001

Baseline 6 Months 12 Months1

4

16

64

256

1024

4096C

-pep

tide

AU

C(n

mol

/L)

N=52 N=49 N=52

p=0.0179

p<0.0001

p<0.0001

of 50 Diabetes

Snowhite et al.

9

Supplementary material

Blood Processing. Serum samples used in this study were provided by the Type 1

Diabetes TrialNet. Samples were obtained from participants at various TrialNet sites and were

uniformly collected and processed according to the TrialNet processing SOPs. Ot obtain

serum, blood was collected in 2.5 mL red top SST gel tube. The tube was gently inverted 5

times and placed upright in a tube rack. The blood was allowed to clot for 20-30 minutes at

room temperature, then it was centrifuged for 15 minutes. The serum was transferred into a

1.8 mL cryovial, placed upright in a 2” partitioned freezer storage box, then frozen at -70°C.

Assessment of Hemolysis. Only visually apparent hemolysis interferes with hybrid-

capture based assays and this was never observed at visual checks performed before sample

submission and before processing. In miRNA RT-PCR assays hemolysis is present when the

miR-23a-3p/miR-451a Cq ratio of is >7 (1) or >9 (2), as reported in different studies. In our

data, this translates to differential expression levels of miR-23a-3p/miR-451 greater than 128

or 512-fold, respectively, due to the binary logarithmic nature of Cq-values. The observed

difference was much lower (mean 13.6 ± SD 18.5) than 128-fold. Thus, hemolysis levels were

satisfactorily low in all samples.

Assessment of platelet contribution. Applicable to all studies of circulating miRNAs, the

study of serum or plasma samples has the limitation that circulating miRNAs reflect a variety

of cellular sources, including platelets. Serum typically contains fewer platelets than plasma

when processed by standard clinical collection protocols with normal CBC platelet counts of

~200,000/mL in whole blood, ~28,000/mL plasma, and <1000/mL serum. Thus, since we used

serum, the possible contribution from platelets would be much reduced compared to studies

that used plasma. We are not aware of reports of truly platelet-specific miRNAs. However,

several miRNAs have been linked to platelet activation or are expressed also by platelets.

of 50Diabetes

Snowhite et al.

10

Figure S2 shows median and interquartile ranges of raw CPM values, in the Log2 scale, for

the 31 miRNAs associated with C-peptide in this study and 50 miRNAs that could be commonly

contributed by platelets (albeit not exclusively) (3). Overall, the miRNAs associated with C-

peptide had significantly lower expression levels than platelet miRNAs (p<0.0001). Moreover,

miRNAs associated with platelets were not a key contributor to the main associations reported

in this study because only 3 of the miRNAs associated with C-peptide overlapped with those

linked to platelets: these are miR-103a-3p, miR-342-3p, and miR-197-3p. As described in the

main text, these three miRNAs have been linked to type 1 diabetes by multiple studies.

References1. Blondal T, Jensby Nielsen S, Baker A, Andreasen D, Mouritzen P, Wrang Teilum M, Dahlsveen IK: Assessing sample and miRNA profile quality in serum and plasma or other biofluids. Methods (San Diego, Calif) 2013;59:S1-6

2. Zhelankin AV, Vasiliev SV, Stonogina DA, Babalyan KA, Sharova EI, Doludin YV, Shchekochikhin DY, Generozov EV, Akselrod AS: Elevated Plasma Levels of Circulating Extracellular miR-320a-3p in Patients with Paroxysmal Atrial Fibrillation. International journal of molecular sciences 2020;21

3. Sunderland N, Skroblin P, Barwari T, Huntley RP, Lu R, Joshi A, Lovering RC, Mayr M: MicroRNA Biomarkers and Platelet Reactivity: The Clot Thickens. Circ Res 2017;120:418-435

of 50 Diabetes

Snowhite et al.

11

Fig. S2. Median and interquartile ranges for 31 miRNAs associated with C-peptide in this study

compared to miRNAs associated with platelets in the literature.

31C-pep

tide Ass

ociated

miRNAs

Platele

t miRNAs

0.0009765625

0.03125

125

210

215

220

Raw

Cou

nts

perM

illio

nR

eads

(CPM

) miR-103a-3pmiR-197-3pmiR-342.-3p

of 50Diabetes

page 1 of 50 diabetes...1 baseline assessment of circulating micrornas near diagnosis of type 1...

Documents