SCRI Research Symposium

Mast cells (MC) are granule-containing immune cells that primarily reside at mucosal surfaces and the skin. MC are strategically poised to initiate immune responses to allergic and pathogenic threats by releasing an impressive arsenal of proteases, cytokines, chemokines, and other inflammatory mediators. As major effector cells of inflammation, the activation of MC must be tightly regulated, and one type of inhibitory receptor on MC are Sialic acid-binding immunoglobin-like lectins (Siglecs). Siglecs are type I transmembrane lectins that are characterized by the presence of an N-terminal V-set immunoglobin domain that binds sialic acid and a C-terminal cytoplasmic region that typically contains one or more immunoreceptor tyrosine-based inhibitory motifs (ITIM) that dampen immune cell activation. Human MC express Siglec 2, 3, 5, 6, 7, 8, and 10. Ligation of Siglec 6, 7, or 8 prior to stimulation by IgE results in decreased activation of MC and, subsequently, decreased release of pro-inflammatory mediators. Siglec 9 is expressed on neutrophils, monocytes, B cells, and NK cells and plays an important role in limiting the baseline activation of these immune cells. A high throughput screen of lung MC detected Siglec 9 on their surface, but this result was not validated, and it is unknown whether Siglec 9 plays an important role in MC functions. Thus, we hypothesize that MC express Siglec 9 and that signaling through this receptor can dampen the activation of MC. Our data shows that human MC lines LAD2, LUVA, and HMC-1 express high levels of Siglec 9 by mRNA and surface protein expression. Moreover, pre-treatment of LAD2 MC with Siglec 9 agonists Glycophorin A or High molecular weight hyaluronic acid (HMW-HA) followed by stimulation with IgE results in decreased release of granule content. These agonists showed no inhibitory effect against CRISPR/CAS edited LAD2 cells deficient in Siglec 9. Additionally, we tested whether human primary cells express Siglec 9 by using peripheral blood-derived cultured MC (PBCMC) as a model. The expression of Siglec 9 peaks at week 5 of differentiation of PBCMC and correlates positively with the expression of the high affinity receptor for IgE, FCƐRI. Interestingly, incubating PBCMC with IgE increased the expression of both FCƐRI and Siglec 9. Together, our data shows that human MC express Siglec 9 and that engaging this inhibitory receptor can reduce the immune cell’s activation. This is significant because targeting Siglec 9 may be a powerful tool to limit inflammation by dampening innate immune responses.

Irina Miralda

Center for Immunity and Immunotherapies

CIIT, 2:30-3:00

Background: The expansion of telehealth services during the COVID-19 pandemic has allowed for continued access to care for many patients, but little is known about whether it has removed or exacerbated existing barriers in access to gender-affirming care for transgender and gender diverse (TGD) youth. Hypothesis: The purpose of this study was to understand whether demographic characteristics of the TGD patient population receiving care in the Seattle Children’s Gender Clinic (SCGC) changed once telehealth services became available beginning in March 2020. Methods: We used electronic health record data to identify all patients who had an ICD-10 diagnosis of gender dysphoria and completed a medical visit at SCGC, either in-person or through telehealth, between March 2019-February 2021. Chi-squared analyses were used to assess differences in demographic characteristics and care utilization among those receiving care between March 2019-February 2020 (pre-telehealth) and March 2020-February 2021 (post-telehealth). We also tested for differences among those receiving care in the post-telehealth period after stratifying by telehealth and non-telehealth users. Results: Of the 1,061 unique patients during the study period (mean age at first visit: 15.6 years), 61.8% identified as transmasculine/male, 24.2% as transfeminine/female, and 7.8% as non-binary. The majority (67.7%) identified as non-Hispanic White, followed by Hispanic/Latinx (11%), Asian (3.2%), Black/African American (2.2%) and American Indian/Alaska Native and Native Hawaiian/Pacific Islander (0.8%). More than two-thirds (66.8%) resided within 35 miles of SCGC. Comparing the pre- and post-telehealth periods, a smaller proportion of patients identified using she/her and he/him pronouns and a greater proportion used she/they and he/they pronouns (p=0.007). Similarly, smaller proportions used all insurance types except for self-pay/charity care which increased post-telehealth (p<0.001). In the post-telehealth period, 48.6% of all visits were conducted via telehealth; however, there were no significant differences in telehealth utilization by demographic characteristics. Conclusion: Our findings indicate that demographic characteristics of SCGC patients remained relatively consistent with the introduction of telehealth services. Although this suggests that telehealth has not introduced new disparities in access to gender-affirming healthcare, the SCGC patient population continues to be disproportionately comprised of White privately insured patients living in urban areas, indicating ongoing disparities in access to gender-affirming care for youth of color and rural youth. Overall, these findings improve our understanding of how telehealth has affected access to gender-affirming care and can be used to advocate for evidence-based policies to improve healthcare delivery and reduce health inequities among TGD youth.

Ruby Lucas

Center for Child Health, Behavior and Development

CHBD, gender care, telehealth, 1:00-1:30

Background: Ecological Momentary Assessment (EMA) is a tool that is well used in research to gather information remotely from individuals about their mood, behaviors, and symptoms in real time. Utilizing remote assessment became increasingly important as a result of social distancing guidelines. Preliminary data about the feasibility and acceptability of using EMA as part of psychiatric outpatient therapy services during the COVID-19 Pandemic are discussed in this abstract.

Hypothesis: We hypothesize that EMA will be a feasible and acceptable method for use in outpatient psychiatry groups.

Methods: Spanish and English-speaking parents and their children who were enrolled in 1 of 6 outpatient psychiatry groups at Seattle Children’s North Clinics were invited to participate in the study. After informed consent was obtained, participating families downloaded the Metric Wire application, designed for EMA data collection. Families received EMA notifications to complete a survey between 2-3 times per week on either a specified time designated by the participants or on randomly selected days throughout the week. EMA questions asked about the child’s mood, behavior or a specific parenting skill used on a particular day. Participating families and clinicians completed semi-structured qualitative interviews to obtain feedback about the acceptability of these procedures as part of outpatient psychiatric care.

Results: 15 families enrolled in the study. 47% of participants were Caucasian, 33% were Hispanic, and 20% were Asian/Pacific Islander. 27% of participants were primarily Spanish-speaking families. On average, families responded to 55% of EMA surveys. Based on the qualitative interviews, participating families reported that each EMA survey served as a “reminder” to utilize one of the parenting skills taught in their outpatient group. Based on qualitative interviews with clinicians, they expressed difficulty in incorporating EMA as part of their outpatient psychiatry group. Challenges centered around the time-commitment and remembering to review the EMA surveys for each family participating in the study.

Conclusions: Preliminary data indicates that EMA is acceptable to families but there are issues with feasibility that may pose a challenge for integration in outpatient psychiatry services. Our findings will provide recommendations for overcoming feasibility challenges and recommendations for future iterations of EMA procedures in clinical care.

Yesenia Garcia

Center for Child Health, Behavior and Development

CHBD, 1:30-2:00

Background: During HIV infection, liver disease is a major cause of morbidity and mortality even in individuals treated with combination antiretroviral therapy (cART), but its pathogenesis is not completely understood. The simian immunodeficiency virus (SIV), infection of rhesus macaques, represents an optimal research model, as SIV is genetically related to HIV and results in similar pathology. Using this animal model, we are able to collect repeated laparoscopic liver biopsies, allowing for a longitudinal evaluation of immune cell subsets during SIV infection. Previous research found an increase in CD8+ T cells in the liver during SIV infection as well as a correlation between Ki67+ (replicating) T cells in the bloodstream and gastrointestinal tract of macaques. CXCR3+ T cells have been shown to be associated with T cell recruitment to inflamed and infected tissues, while CCR6+ T cells are associated with wound healing and fibrosis. We hypothesized that early in SIV infection, T cells would be replicating locally and thus we expected to find more hepatic T cells expressing Ki67 than before infection or in samples from naïve macaques. We further hypothesized that T cell recruitment and inflammation would be evidenced by an increased number of hepatic CXCR3+ T cells early in SIV infection. Finally, we predicted to find a greater number of CCR6+ T cells during late-stage SIV infection, which is associated with liver fibrosis.

Methods: Liver biopsies from rhesus macaques were fluorescently labeled for CD3 (T cell), Ki67, CXCR3, and CCR6 antibodies. CD3+ Ki67+ cells were analyzed in naïve and SIV+ animals at baseline (week 4 before infection) and acute/early infection (week 2 after infection). Both CD3+ CXCR3+ and CD3+ CCR6+ cells were investigated at acute infection and necropsy (week 15-32 after infection). Immunofluorescence analysis was performed with Keyence BZ-X710 microscope. Computational analysis was conducted using Fiji/Image J, to quantify the percentage of hepatic Ki67+, CXCR3+, and CCR6+ T cells. GraphPad Prism was used to plot the results and perform statistical analyses.

Results: During acute SIV infection, the expression of Ki67+ T cells in all SIV-infected macaques was significantly higher than at baseline (P< 0.0001) and in naïve samples (P< 0.0001). Two SIV-infected macaques had a greater proportion of CXCR3 at necropsy compared to week 2, while two other SIV-infected macaques showed the opposite trend. In the CCR6+ subset, the same relationship was observed, however not for corresponding animals. Overall, comparing week 2 and necropsy, in the SIV cohort there was no significant change of expression of both CXCR3+ and CCR6+ T cells. At week 2 and necropsy, CXCR3+ T cells averaged 45% and 40%, respectively, whereas CCR6+ T cells averaged 10% and 11%, respectively.

Conclusion: Our findings indicate that hepatic T cell replication in SIV-infected macaques increases in early infection compared to baseline and in naïve samples. Contrary to our hypotheses, there was not a significant difference in the proportion of hepatic T cells expressing CXCR3 and CCR6 early and at late stage SIV infection. Further investigation will assess CXCR3+ and CCR6+ T cells at baseline and intermittent timepoints between week 2 and necropsy. In addition, more research will quantify CXCR3+ and CCR6+ T cells within the CD4+ and CD8+ T cell populations.

Yohannes M. Abraham

Center for Global Infectious Disease Research

CGIDR, 2:00-2:30

Background Therapists and supervisors use LYSSN, an AI-based tool that provides feedback on therapy audio tapes, and Care4 (a dashboard that provides supervisors with a library of resources for preparation of their clients and treatment manual). These online systems help support the implementation of STAND, (Supporting Teens Autonomy Daily), an evidence-based practice modular therapy that allows families to choose activities that target ADHD (Attention Deficit Hyperactivity Disorder) symptoms that focus on building skills as: academic, interpersonal, and planning with goal settings and implementation to help reduce ADHD symptomatology. The implementation of STAND is a cost-effective manner to implement evidence-based practices compared to the usual-care psychological services and medication. The purpose of this study is to demonstrate the utilization of resources on an online system and attitudes of therapists and supervisors when interacting with LYSSN and Care4 to implement STAND. Methods From a collection of four supervisors and six therapists, each therapist having a minimum of one patient and at maximum two patients, I use 49 therapist sessions and 3 supervision sessions to determine the attitudes and engagement. Therapists would login onto the Care4 system after a therapy session and answer questions about feedback they received from LYSSN and the preparation for their session. Supervisors could use the data collected from the therapist’s LYSSN score and formulate their feedback. Results On average it took supervisors 48.6 days to upload their tapes after having their sessions. 33% of supervisors would upload their session tapers within 48 hours. Therapists would take on average 10.9 days to upload their tapes after having their sessions. 25% of therapists would upload their session tapes within 48 hours. There was an average of 1.4 (SD = .77; 1= no time to 4-more than thirty minutes) for therapists reviewing feedback from their supervisors. 1.7 (SD = 1.2) for seeking consultation from the STAND research team. 1.9 (SD = .83) for reviewing feedback from LYSSN. 2.8 (SD = .93) for watching STAND demonstration videos. 3.2 (SD = .69) for preparing client worksheets. 3.4 (SD = .54) for reading through STAND descriptions. 2.8 (SD = .84; 1 – never to 5 – always) when it came to finding feedback credible. 2.7 (SD = .8) when it came to therapists finding their feedback to be accurate in summarizing their actions in their session. 3 (SD = 0; 1 – much easier to 5 – more difficult) to indicate the difficulty of supervision in comparison to their typical cases. 3.3 (SD = .57; 1 – very uncomfortable to 5 – very comfortable) in the comfortability of providing supervision. 1.6 (SD = .57; 1 - being not at all to 5 – being very much) from the extent supervisors would incorporate LYSSN feedback. 2 (SD = 1) from how credible LYSSN feedback was. 2 (SD = 1) from how useful LYSSN feedback was to provide supervision feedback. 3.6 (SD = .57) from how receptive the therapist was to feedback. Discussion Collection of this research will provide insight about the implementation of STAND through an online format and about the general appeal of the use of AI to provide feedback to therapists and help supervisors with the feedback they provide. Using AI feedback, the workload of supervisors and therapists could be reduced and give feedback more efficiently to them with regards to STAND treatment and other evidence-based treatments.

Jessica Monroy <jmonroy@uw.edu>

Center for Child Health, Behavior and Development

CHBD, 2:00-2:30

Background: Families who speak a language other than English have not been considered in the design of existing digital health technologies. The way in which we have designed these technologies do not cater to every family, which means they cannot recognize the benefits of these technologies for improving family-provider interactions. Purpose/Objective: To analyze and examine the health information technology disparities within Seattle Children’s and how it affects families whose first language is not English. Design/Methods: We conducted 25-minute interviews with families of patients hospitalized at Seattle Children’s Hospital (SCH) whose preferred language for medical communication is not English. The interview consisted of questions relating to how families access the SCH’s patient portal, how accessible interpreter services are, and how they connect to SCH providers from home. Our team also created login prototypes exhibiting various methods of logging into patient portals and helpline prototypes to triage questions from patients to their providers. We presented families scenarios in which they would give responses to how they would contact providers and any difficulties with telehealth services they have faced throughout their stay at Seattle Children’s. Results: We spoke to two Spanish speaking families, one Somali speaking family, and one Persian family that was conversational in English. All of the families stated that they do not use MyChart to contact their providers due to the difficulty and lack of understanding with the tool. On average it took 20 minutes to connect with their provider through interpreter services. 50% of the families have another person (e.g. child, spouse, family member) call in lieu of themselves to make their child's appointment. 75% of the families preferred login three (see figure 1) because the icons for passcode were easier to understand. The families chose to use the phone option for the helpline because it would be easier to call their providers and less error would occur over the phone (see figure 3).

Conclusion/Discussion: Health information technologies are an ongoing issue in healthcare. Further methods of reducing these issues need to be put in place to break down this barrier to promote equity within hospitals.

Jordyn Alyssa Warner <jordynwarner0830@gmail.com>

Center for Child Health, Behavior and Development

CHBD, 1:30-2:00

Background: Leigh Syndrome is a pediatric neurodegenerative disease that predominantly affects infants. It is associated with loss-of-function mutations in NADH-ubiquinone iron-sulfur protein 4 (NDUFS4) which resides within Complex I (CI) of the electron transport chain. The conditional knockout (KO) of Ndufs4 in only GABAergic interneurons leads to a severe epilepsy phenotype, suggesting GABAergic interneurons drive the severe and often fatal epilepsy phenotype which is commonly reported in patients. This present study aims to target a subtype of GABAergic interneurons called parvalbumin (PV) interneurons which display fast spiking firing patterns and are known to be expressed in the cortex and hippocampus, key regions often associated with epilepsy.

Hypothesis: We hypothesized that since PV interneurons display fast spiking patterns, the Ndufs4 KO will impair these cells, leading to severe spontaneous seizures, cell loss, and neurocognitive impairments.

Methods: Experimental mice models with Ndufs4flx/flx/PVCreflx/+ genotype for the mutants, and Ndufs4flx/flx/PVCre+/+ genotype for the controls were used in all of the experiments except imaging. For imaging experiments, Ndufs4flx/flx/Ai14flx/+/PVCreflx/+ were used for mutants and Ndufs4+/+/Ai14flx/+/PVCreflx/+ were used for controls.

For open field, mice of both genotypes were placed into a 50x50 cm open field chamber for 10 minutes and their movements were tracked. For EEG recordings, mice were implanted with EEG/EMG electrodes and placed in recording chambers where we assessed occurrence, frequency and duration of seizures and epileptiform events. For the PTZ challenge, mice were injected (subcutaneous) with 20 mg/kg PTZ and then placed in recording chambers for approximately 30 minutes where we assessed occurrence, frequency, and duration of seizures. For imaging, we counted Ai14-labeled PV interneurons in key areas associated with epilepsy (for example, cortex and hippocampus) between the two groups. Brains were harvested, sliced, mounted in triplicates, and imaged at 10X with a confocal microscope.

Results: PV mutants showed an increase in the frequency of myoclonic seizures and interictal spikes. There was no difference in seizure susceptibility (from the PTZ challenge) between mutants and controls. Additionally, we did not observe major impairments in locomotor activity nor anxiety like behavior in PV mutants. Finally, we did not detect cell loss changes in PV mutants.

Conclusion: These experiments show that PV mutants display a mild seizure phenotype, with no cognitive or motor abnormalities, suggesting targeted Ndufs4 KO in PV interneurons does not drive the severe epilepsy phenotype in LS.

Devika Gandhay

Center for Integrative Brain Research

CIBR, 2:00-2:30

We have created robots to automate the laborious task of dissecting mosquitoes and extracting malaria parasites. Producing malaria parasites is necessary for malaria research and vaccine development. Extracting even small quantities of the parasites by hand requires hours of difficult and tedious mosquito dissection, which limits the efficiency and extent of malaria research. Our team of scientists and engineers has built mosquito dissection and processing robots to automate parasite. Our robots dissect mosquitoes approximately eight times faster than by hand, and users can successfully dissect mosquitoes with significantly less training compared to manual methods. To test the usability of our robots, we sent to researchers at the Pennsylvania State University, who successfully dissected mosquitoes. In vitro studies of our robots show that the parasites produced are infectious but exhibit reduced immunogenicity when compared to hand dissected parasites. By replacing the inefficient process of dissecting mosquitoes by hand with our automatic dissection robots, we aim to remove a major barrier to researching malaria.

Zephyr Pitre

Center for Global Infectious Disease Research

CGIDR, 2:30-3:00