Date of Award

Summer 8-15-2018

Author's Department

Biomedical Engineering

Degree Name

Doctor of Philosophy (PhD)

Degree Type

Dissertation

Abstract

Immune dysregulation has been implicated in pediatric autoimmune disease

pathogenesis, but the underlying mechanisms still remain poorly understood. We explored

immune cell signaling dysregulation in two different pediatric autoimmune diseases:

polyarticular juvenile idiopathic arthritis (JIA), a disease characterized by pain, swelling, and

limited range of motion in five or more joints and juvenile dermatomyositis (JDM), a disease

characterized by pathognomonic skin rashes and proximal muscle weakness. Both polyarticular

JIA and JDM can be controlled with newer biologic medications, but a substantial fraction of

patients still experiences refractory disease courses. To explore potential immune cell signaling

dysregulation, mass cytometry was performed on peripheral blood mononuclear cells (PBMCs)

from treatment-naïve patient, remission patient, and control blood samples. Cryopreserved

samples were thawed, rested, stimulated with different cytokines (polyarticular JIA) or a cocktail

of stimuli over a time course (JDM), fixed, barcoded, stained with surface marker antibodies,

methanol permeabilized, stained with antibodies for intracellular molecules, and analyzed on a

CyTOF2/Helios mass cytometer.

No differences in immune cell signaling were detected at baseline or after 15 minutes of

IL-6 stimulation between treatment-naïve polyarticular JIA patients and controls. With IFN�

stimulation, treatment-naïve polyarticular JIA patient classical monocytes and naïve CD4 T cells

more strongly phosphorylated STAT1 and/or STAT3 than those from controls. These stratifying

cell populations were heterogeneous. Enhanced IFN� responsiveness in naïve CD4 T cells was

associated with elevated levels of JAK1 and SOCS1.

Examination of the JDM data set at 0, 3, and 15 minutes with a cocktail of multiple

stimuli with Significance Analysis of Microarrays (SAM) yielded 292 results with a false

discovery rate (FDR) of less than 5%. As a large number of results is difficult to manually

interpret, Citrus paired with LASSO feature selection was also employed to model differences in

immune cell signaling between treatment-naïve JDM patients and controls over three time

points. LASSO selected 12 Citrus features (signaling molecule in specific immune cell subset);

10 out of 12 were in PLC�2 phosphorylation with 4 out of 12 in NK cell PLC�2

phosphorylation. NK cell PLC�2 phosphorylation was lower in treatment-naïve JDM patients in

comparison to controls across all analyzed time points. Decreased treatment-naïve JDM patient

NK cell PLC�2 phosphorylation was associated with lower levels of calcium flux in NK cells

upon receptor crosslinking in comparison to a healthy control.

While similar analytical approaches were applied to the polyarticular JIA and JDM

cohorts, different stimulation protocols were utilized. However, quantitative analytical

techniques revealed previously undescribed signaling abnormalities in both diseases. These

signaling differences provide a mechanism for tofactinib (a JAK inhibitor) treatment in a clinical

trial for polyarticular JIA and a potential new target for therapy in JDM. In addition, these

signaling differences may be valuable for confirmation of polyarticular JIA and JDM diagnoses,

as well as biomarkers to predict which treatment-naïve patients may respond to a given therapy

with expansion of the original models.

Language

English (en)

Chair

Anthony R. French

Committee Members

Lori Setton, Kristen Naegle, Todd Fehniger, Stephen Oh,

Comments

Permanent URL: https://doi.org/10.7936/q14t-sm67

Share

COinS