Abstract
To investigate the response of primary nasal epithelial cells (NECs) to novel alpha haemolytic Streptococcus and lactobacilli strains, isolated from the upper respiratory tract of children.
Submerged cultures of NECs from healthy adult donors were exposed to either novel strains; Lactobacillus rhamnosus D3189, D3160, Streptococcus salivarius D3837; or commercially available probiotic strains L. rhamnosus LB21, S. salivarius K12; or a pathogenic strain (S. pneumoniae 49619). Cytotoxicity (measured through lactate dehydrogenase release) and cytokine release were quantified 24 hours post-exposure. Exposure to novel and commercially available strains did not induce the production of IFN-β, IFN-λ1/3, IL-1β, IL-6, IL-8, or TNF-α production or the release of LDH. Conversely, the pathogenic strain S. pneumoniae 49 619 significantly elevated the expression of IL-1β, IL-8, TNF-α, and LDH in NECs.
The findings within this study highlight the non-pathogenic nature of these novel strains and support further investigation of the potential to develop nasally administered probiotics.
Upper respiratory tract infections and otitis media represent a substantial socioeconomic burden, highlighting a critical need for effective preventative strategies. Previous research indicated that alpha haemolytic Streptococcus and lactobacilli isolates were associated with a reduced incidence of these conditions, suggesting potential probiotic characteristics for nasal delivery. This study further underscores their safety, showing that the novel strains do not induce cytotoxic effects or inflammatory responses in NECs. These results pave the way for exploring the use of these strains in developing effective nasal probiotic therapies.
Introduction
The upper respiratory tract (URT) supports a respiratory microbiota that is likely to play a pivotal role in maintaining a healthy URT environment (Hakansson et al. 2018, Man et al. 2017). As such, interest has grown in understanding how commensal bacteria may be used as probiotics to maintain a healthy nasal microbiome and therefore reduce the risk and incidence of upper respiratory conditions such as otitis media (OM) and rhinosinusitis (RS). There is published evidence from clinical trials (Brugger et al. 2020, De Boeck et al. 2020, De Boeck et al. 2017, DeMuri et al. 2018, Endam et al. 2020, Man et al. 2017, Man et al. 2019, Marchisio et al. 2015, Ortiz Moyano et al. 2020, Roos et al. 2001, 2011, Skovbjerg et al. 2009) and in vivo mouse studies (Gabryszewski et al. 2011, Harata et al. 2010, 2011, Hori et al. 2001, Izumo et al. 2010, Park et al. 2013, Tomosada et al. 2013, Yeo et al. 2014, Youn et al. 2012), which demonstrate that nasal administration of probiotic bacteria can modulate both infection by pathogenic bacteria and viruses, respectively, and mucosal inflammation. Moreover, in vitro studies using immortalized airway epithelial cells (AECs) lines have demonstrated that probiotic bacteria can positively influence resistance against viral infection by differentially modulating innate immune responses in AECs (Islam et al. 2021, Olímpio et al. 2022). However, it is important to select and use bacterial strains that can be correlated with a healthy microbiome and reduced URT disease that are also not potentially damaging to the nasal mucosa.
Lactobacillus spp. and alpha haemolytic Streptococcus (AHS) spp. are the most investigated microbes with potential probiotic characteristics within the URT (Tapiovaara 2016). Lactobacilli are acid-producing bacteria that are ubiquitous in the gastrointestinal tract but also detected in the URT (Bogaert et al. 2011, De Boeck et al. 2020, Stearns et al. 2015). Lactobacillus spp. have been studied widely, with several strains, including model probiotic L. rhamnosus GG, displaying strong and direct inhibitory activity against Moraxella catarrhalis in cultured human AECs (van den Broek et al. 2018). Moreover, some Lactobacillus casei strains inhibit the growth and virulence of Haemophilus influenzae, M. catarrhalis, and Staphylococcus aureus by conferring anti-inflammatory benefits in vitro (De Boeck et al. 2020). Similarly, AHS are abundant URT commensals, and are often found in the oral cavity from birth (Pearce et al. 1995). Colonization with AHS in the URT has been inversely correlated to respiratory pathogen colonization (Brook 2005), with several strains including alpha 4 (Streptococcus oralis), alpha U2 (S. oralis), and alpha 89 (S. sanguis) demonstrating in vitro inhibition against key bacterial respiratory pathogens, S. pneumoniae, H. influenzae, and M. catarrhalis (Tano et al. 2003; Tano et al. 2002). Many AHS species and strains produce bacteriocins, hydrogen peroxide, and other inhibitory substances that contribute to their bacterial interference characteristics (Brook 2005, Tano et al. 2003). To date, various Lactobacillus (L. rhamnosus GG, L. rhamnosus LB21, L. rhamnosus AMBR1, L. casei AMBR2, L. rhamnosus AMBR3, L. rhamnosus AMBR4, L. rhamnosus AMBR5, L. rhamnosus AMBR6, L. rhamnosus AMBR7, L. sakei AMBR8, and L. plantarum AMBR9,) (Coleman and Cervin 2019, De Boeck et al. 2020) and AHS (S. salivarius 24SMB, S. salivarius K12, and S. sanguinis 89a) (Coleman and Cervin 2019, Marchisio et al. 2015, Roos, Hakansson et al. 2001, Skovbjerg et al. 2009) strains have been developed for commercial use orally and have been administered intranasally in both children and adults in clinical trials and demonstrate an excellent safety profile.
The effects of beneficial microbes on the prevention and treatment of disease are strain-specific and disease-specific (McFarland et al. 2018). Therefore, isolating health-associated microbes from healthy subjects in comparison to individuals prone to OM and RS is key for the identification of effective inhibitors of respiratory pathogens. Previous studies collected swabs from the URT of healthy children, cultured them under aerobic and anaerobic conditions, and identified morphologically unique colonies using Matrix-Assisted Laser Desorption/Ionisation Time-of-Flight Mass Spectometry (MALDI-TOF) or 16S rRNA V3–V4 sequencing. Notably, L. rhamnosus D3189 and D3160 and S. salivarius D3837 demonstrated strong inhibitory effects against S. pneumoniae, H. influenzae, and M. catarrhalis, pathogens commonly associated with OM and RS, using bacterial interference assays (Coleman et al. 2021, 2022). Whole-genome sequencing and phenotypic testing confirmed their safety by revealing the absence of virulence genes, tropism to the URT, and antibiotic susceptibility to commonly used URT antibiotics. Furthermore, as strains isolated from healthy URTs, they are likely phenotypically adapted to this environment, enhancing their probiotic potential (Coleman et al. 2022). These findings highlight the novelty of these strains as promising candidates for reducing pathogen carriage, a reduction strongly associated with decreased prevalence of OM and RS (Sun et al. 2012, Coleman et al. 2018).
However, a key knowledge gap concerning these potential probiotic bacteria, and indeed most probiotics that are currently in clinical trials, or considered for trial via intranasal delivery, is how they each interact with nasal epithelial cells (NECs). As such, it is important to understand how NECs may react to different potentially probiotic strains to optimize treatment selection and avoid deleterious responses.
In this context, the aim of this work was to evaluate the response of primary human NECs to L. rhamnosus D3189 and D3160, and S. salivarius D3837, in comparison to commercially available probiotics L. rhamnosus LB21 and S. salivarius K12 and bacterial pathogen S. pnemoniae ATCC strain 49619. As such, the effect of bacterial exposure on cytotoxic responses and cytokine release of NECs were investigated to identify strains that may induce harmful effects when administered to nasal cells.
Materials and methods
Ethics statement and collection of primary NECs
Donors were recruited and nasal cells collected in accordance with the Queensland University of Technology Human Research Ethics Committee (2021000292). Primary NECs were collected from healthy adult (aged 18–50) donors (N = 5, 4 females, 1 male, 32.8 ± 10.9 years old), with informed consent, by scraping a sterile nasal mucosal curette (Arlington Scientific, USA) across the inferior turbinate. NEC cultures were established and expanded as submerged monolayers as previous described (Kicic et al. 2010, Kicic et al. 2016, Spann et al. 2014), and stored in aliquots in freezing media (foetal bovine serum with 10% dimethyl sulfoxide) at passage 1 or 2 in liquid nitrogen until utilized for experiments. Prior to experimentation, PCR testing was performed for a panel of common respiratory viruses and bacteria to confirm their pathogen-free status.
Submerged culture of primary NECs
Primary NECs were seeded at a density of 5 × 104 cells per well of tissue culture chamber slides (Sigma Aldrich, Germany) and Costar® 24 Well Clear TC-Treated Multiple Well Plates (Corning, USA). Cells were cultured in 300 µl of PneumaCult™-Ex Plus Medium (Stemcell Technologies, Canada) and 500 µl for 24 well plates. The medium was supplemented with 0.1% hydrocortisone (HC; Stemcell Technologies, Canada) and 1% Antibiotic-Antimycotic (Anti-Anti; Thermo Fisher Scientific, USA). Cell culture media was changed three times a week until the cells reached >80% confluency. Once confluent, HC and Anti-Anti were removed 24 h prior to probiotic exposure. Cell cultures were selected at random to confirm the epithelial phenotype of the cells using immunofluorescent staining for cytokeratin 5 and confocal imaging. Cultures were stained with mouse anti-human monoclonal antibody against cytokeratin 5 (1:100, Abcam), followed by detection using secondary antibody goat anti-mouse IgG H + L Alexa Fluor 488 (1:250, Thermo Fisher Scientific). Nuclei were visualized with Hoechst staining, and epifluorescence microscopy (Zeiss Axio Vert.A1, Germany) confirmed the presence of basal NECs in the cultures (See Fig. S1).
Bacterial culture preparation
Lactobacillus rhamnosus strains D3189 and D3160 and S. salivarius D3837, isolated from URT of healthy children (Andrea Coleman et al. 2021), and widely used commercial probiotics L. rhamnosus LB21 (provided by Essum AB, Umeå, Sweden) and S. salivarius K12 (ATCC® BAA-1024™) were selected for this study. Pathogenic S. pneumoniae ATCC strain 49619 was also used to compare cellular responses to probiotic and pathogenic strains. Each isolate was cultured in Man-Rogosa-Sharpe (MRS) broth, for Lactobacillus spp, or Todd-Hewitt broth, for Streptococcus spp. Broth preparations were seeded from one colony forming unit (CFU), selected from solid MRS or HBA agar, respectively, and were grown in pure culture until the end of the lag phase at 37°C. Broth cultures were then harvested by centrifugation and washed twice with sterile phosphate-buffered saline (PBS). Each strain was labelled with carboxyfluorescein diacetate, succinimidyl ester for confocal imaging to confirm bacterial survival, as described previously (Yarlagadda et al. 2024). Following labelling, isolates were resuspended in PneumaCult™-Ex Plus Medium (Stemcell Technologies) at a concentration of 2.5 × 107 CFU ml−1.
Probiotic exposure
Duplicate cultures of NECs from five healthy adult donors, grown in both tissue culture chamber slides and 24 well plates were exposed to 300 µl of each of 5 probiotic strains and 1 pathogenic strain. Duplicate cultures of NECs were also exposed to cell culture media containing no bacteria (control). After incubation for 4 h at 37°C in 5% CO2, the bacterial inoculum was replaced with PneumaCult™-Ex Plus Media and NECs incubated under the same conditions for a further 20 h. Supernatants were collected 24 h post bacterial exposure and stored at −80°C.
NECs in chamber slides were fixed, and immunofluorescent staining and confocal imaging was performed. Cultures in 24 well plates were washed three times with PBS to remove non-adherent bacteria, followed by disruption of the cell monolayers using 200 µl of 0.1 M Triton X-100 in 0.1 M PBS with gentle vortexing. Serial dilutions were performed to enumerate the total number of cell-associated bacteria.
Immunofluorescence
NECs cultured on tissue culture chambers and exposed to Carboxyfluorescein succinimidyl ester (CFSE)-labelled bacterial strains for 24 h were fixed with 4% paraformaldehyde/PBS, permeabilized with 0.5% (v/v) Triton X-100 (Sigma–Aldrich; X100) in PBS for 10 mins at RT and blocked with 2% (w/v) bovine serum albumin (BSA; Sigma–Aldrich). Alexa Fluor™ 594 Phalloidin (Invitrogen, 1:200 in PBS) was applied for 1 h at RT, followed by washing three times in PBS, application of Hoechst stain (1:2000 dilution of 1.0 mg ml−1 stock solution; Sigma Aldrich) for 10 mins at RT, and washing three times in PBS. Following staining, the chambers were manually removed, and the slides mounted using ProLong™ Glass Antifade Mountant (Thermo Fisher Scientific). Images of the mounted cultures were captured (in x, y, z dimensions) using an inverted SP5 5-channel confocal microscope (Lecia Microsystems) with a 63× objective (numerical aperture 1.25). To avoid non-specific spill over, each laser line was excited and detected sequentially. Z-stacks were captured with a line average of 3 and a Z-step thickness of 0.13 µm.
LDH activity
Lactate dehydrogenase (LDH) release was measured in the supernatants collected after 24 h of exposure as an indication of cytotoxicity, using the LDH-Glo Cytotoxicity Assay Kit (Promega) as per the manufacturer’s instructions. The luminescence of the samples was recorded using a CLARIOstar Omega reader. Enzyme release was expressed as mU ml−1.
Measurement of IFN-β, IFN-λ1/3, IL-1β, IL-6, IL-8, and TNF-α
Supernatants from NECs were used to quantify the release of IFN-β, IFN-λ1/3, and TNF-α (Human DuoSet ELISA kits, R&D Systems) and IL-1β, IL-6, and IL-8 (AlphaLISA Immunoassays, Perkin Elmer, Australia) as per manufacturer’s instructions. Protein secretion was expressed as pg ml−1.
Statistical analysis
Statistical analyses were conducted using the GraphPad Prism 9.4.1 software. For the treatment of the obtained data, we used the Friedman’s test followed by Dunn’s multiple comparisons test. The results indicated by “P < 0.05″ were considered significant, with a 95% confidence interval.
Results
Bacterial survival and colonization of NECs.
To establish the survival and colonization of all bacterial isolates in a submerged model of NECs, we quantified the number of NEC-associated bacteria and performed confocal microscopy using CFSE-labelled bacterial strains (Fig. 1a-f). The results confirmed that all isolates successfully colonized the NECs 24 h post-exposure. Notably, Streptococcus species exhibited significantly higher colonization levels than L. rhamnosus, which showed consistent colonization across isolates (Fig. 1g). Similarly, all Streptococcus isolates showed comparable colonization patterns, with no significant difference observed between S. pneumoniae and S. salivarius.
Survival and colonization of L. rhamnosus, S. salivarius, and S. pneumoniae on NECs 24 h post-exposure. NECs were exposed to L. rhamnosus D3189, D3160, or LB21, S. salivarius D3837 or K12 or S. pneumoniae 49 619 (2.5 × 107 CFU ml−1). (a–f) Representative confocal microscopy images of NECs from one healthy donor exposed to CFSE-labelled L. rhamnosus (a) D3189, (b) D3160, or (c) LB21; S. salivarius (d) D3837 or (e) K12; or (f) S. pneumoniae 49619. Actin cytoskeleton was labelled with Alexa Fluor™ 594 phalloidin, nuclei are stained with Hoescht. Confocal microscopy images were taken with 63x objective lens (oil) (numerical aperture 1:40). Scale bars = 50 µM. (g) NEC-associated bacteria were enumerated using cell lysates at 24 h post-exposure. Each data point represents the mean concentrations of bacteria from duplicate NEC cultures of each subject, with each donor represented by a colour (n = 5). The limits of the box represent the maximum and minimum values, and the line represents the median. Statistical analysis between treatment groups was performed with the Friedman test and the Dunn's multiple comparisons test, using GraphPad Prism 9.4.1 software. *, P < 0.05; **, P < 0.01.
IFN-β and IFN-λ1/3 production by NECs exposed to bacterial strains
Type I (β) and III (λ) IFNs are principal antiviral proteins induced by viral infection of AECs. Therefore, we were interested in the potential of bacteria to induce an antiviral state in NECs. However, none of the bacteria induced IFN-β or IFN-λ 1/3 production in any NEC cultures (data not shown).
Inflammatory cytokine (IL-1β, IL-6, IL-8, and TNF-α) release by NECs exposed to bacterial strains
A deleterious inflammatory response within the URT is one of the defining features of URTIs and the development of OM and RS. Therefore, it was important to identify strains that may induce a harmful inflammatory response in a clinical setting. IL-1β, IL-6, IL-8, and TNF-α, key inflammatory markers produced by AECs, were quantified using supernatants collected from NECs at 24 h post exposure to bacteria and mock-exposed NECs (bacteria-free control).
Cytokine release varied among donors; however, the median secretion of all cytokines in response to both novel and commercial probiotics was comparable to that of mock-exposed NECs, even though S. salivarius exhibited significantly higher colonization levels compared to L. rhamnosus. In contrast, S. pneumoniae 49 619 induced a significantly higher median secretion of IL-1β, IL-8 and TNF-α (Fig. 2a, 2b and 2c, respectively), despite colonization levels comparable to the other two Streptococcus isolates. While IL-6 levels did not show a significant increase in response to any bacterial strains (Fig. 2d), there was a trend towards elevated IL-6 release from NECs exposed to S. pneumoniae 49619. This observation suggests that strains isolated from the URT of children exhibit a lower level of inflammation compared to well-established pathogenic bacteria. This characteristic is particularly advantageous as is aligns with favourable conditions for their potential use as probiotics.
Effect of L. rhamnosus, S. salivarius, and S. pneumoniae on IL-1β, IL-8, TNF-α, and IL-8 release. NECs were exposed to L. rhamnosus D3189, D3160, or LB21, S. salivarius D3837 or K12 or S. pneumoniae 49 619 (2.5 × 107 CFU ml−1) or exposed to cell culture media as a negative (no probiotic) control. Secreted (a) IL-1β, (b) IL-8, (c) TNF-α, and (d) IL-6 was quantified by AlphaLISAs and standard ELISAs using cell supernatant collected from NECs at 24 h post-exposure. Each data point represents the mean concentrations of IL-1β, IL-8, TNF-α, and IL-6 secreted from duplicate NEC cultures of each subject, with each donor represented by a colour (n = 5). The limits of the box represent the maximum and minimum values and the line represents the median. Statistical analysis between treatment groups was performed with the Friedman test and the Dunn’s multiple comparisons test, using GraphPad Prism 9.4.1 software. **, P < 0.01; NS, not significant.
LDH release by NECs exposed to bacterial strains
LDH, a cytosolic enzyme released upon damage to the plasma membrane, was measured in supernatants as an indicator of cytotoxic effects of bacterial exposure. As for IL-1β, IL-6, IL-8, and TNF-α, LDH release was highly variable from donor-to-donor. However, the colour designation assigned to each donor (Fig. 3) highlights the consistent cellular response observed within the same donor across various treatments. The only strain that induced significantly elevated LDH release compared to mock-treatment was S. pneumoniae 49 619, despite exhibiting bacterial colonization levels similar to probiotics of the same species.
Effect of L. rhamnosus on LDH activity. NECs were exposed to L. rhamnosus D3189, D3160, or LB21, S. salivarius D3837 or K12 or S. pneumoniae 49 619 (2.5 × 107 CFU ml−1) or exposed to cell culture media as a negative control. Secreted LDH was quantified by LDH-Glo Cytotoxicity Assay Kit using cell supernatant collected 24 h post-exposure. Each data point represents the mean concentrations of LDH secreted from duplicate NEC cultures of each subject, with each donor represented by a colour (n = 5). The limits of the box represent the maximum and minimum values and the line represents the median. Statistical analysis between treatment groups was performed with the Friedman test and the Dunn’s multiple comparisons test, using GraphPad Prism 9.4.1 software. *, P < 0.05.
Discussion
The present study demonstrates the benign nature of commensal and probiotic bacteria, isolated from the URT of healthy children, in a submerged model of NECs. These data address a key knowledge gap, as there is a limited understanding of how NECs respond when directly exposed to probiotics. It is important to understand this response to identify specific beneficial microbes that are effective inhibitors of respiratory pathogens and can modulate disease, and to exclude those that display cytotoxic effects. The commensal strains in this study were chosen based on their potential as nasally administered probiotics, as demonstrated by health and disease-associated nasal microbiome studies and pathogenic bacterial interference assays (Coleman et al. 2022). While previous findings show a direct inhibitory effect against bacterial pathogens during agar overlay and cell-free supernatant bacterial interference studies, the microbiological and immunological environment of the URT is more complex and requires further testing using appropriate cell culture models.
In investigating the effects of nasal probiotics, NEC models offer a significant advantage as they effectively capture the interactions between host cells and probiotics, allowing for the assessment of cellular responses such as cytokine production and cellular viability. Consequently, NECs provide a more integrated understanding of the effects of probiotic bacteria that gene expression analyses for toxins and antibiotic resistance alone cannot offer. While gene expression data can identify potential risks, it lacks the capacity to reveal host responses and dynamic interactions. Despite the challenges in maintaining NEC cultures due to their variability, their use enables a more comprehensive and realistic evaluation of nasal probiotics, therefore advancing our understanding beyond the limitations of gene expression analyses alone.
We acknowledge, however, that the use of the submerged NEC culture presents certain limitations. Unlike air-liquid interface (ALI) cultures, submerged NECs do not fully differentiate into the diverse cell types characteristic of nasal respiratory epithelium, potentially affecting the model’s representation of in vivo conditions. We opted to first utilize submerged cultures due to their lower cost and shorter preparation time, allowing us to rapidly screen the probiotics for potential cytotoxicity and inflammatory responses before committing to the more resource-intensive ALI cultures.
In this context, the aim of this work was to evaluate the response of primary human NECs to L. rhamnosus D3189 and D3160, and S. salivarius D3837, in comparison to commercially available probiotics L. rhamnosus LB21 and S. salivarius K12 and bacterial pathogen S. pnemoniae ATCC strain 49619. As such, the effect of bacterial exposure on cytotoxic responses and cytokine release of NECs were investigated to identify strains that may induce harmful effects when administered to nasal cells.
Before investigating the response of NECs, it was crucial to first determine whether the bacterial isolates could colonize and survive within NEC monolayers. Using confocal imaging and quantification of live NEC-associated bacteria, we demonstrated that all strains were capable of surviving in this environment for 24 hours. This finding aligns with expectations, as these strains were isolated from the URT and thus exhibit a natural tropism for this environment (Coleman et al. 2021, 2022).
Studies using in vitro and mouse models have demonstrated that probiotic bacteria can modulate cytokine release by epithelial cells. A study by Islam et al. showed increased production of Type 1 IFN and IL-6 in Calu-3 cells treated with L. Plantarum MPL16 and CRL1506 (Islam et al. 2021). Similarly, in a study conducted by Tomosado et al., mice intranasally treated with L. rhamnosus had higher levels of Type 1 IFNs and IL-6 in bronchoalveolar lavage (Tomosada et al. 2013). On the other hand, a study by Cosseau et al. demonstrated that S. salivarius K12 can inhibit baseline IL-8 secretion in cultured 16HBE140- cells (Cosseau et al. 2008). Our findings were not consistent with these studies, as exposure of NECs to potential or known probiotics did not modulate Type I or III IFN, IL-1β, IL-6, IL-8, or TNF-α production compared to mock-exposed cells. We compared these responses to those induced by known respiratory pathogen S. pneumoniae and observed that while probiotic exposure did not induce cytokine release, S. pneumoniae 49619 did induce IL-1β, IL-8, and TNF-α production. This aligns with findings from similar studies that demonstrate the capacity of S. pneumoniae to induce the expression of IL-1β, IL-8, and TNF-α in human bronchial epithelial cells (primary and BEAS-2B) (Schmeck et al. 2004, Weight et al. 2019) and human lung tissue (Szymanski et al. 2012).
We also examined whether probiotic bacteria induced cellular cytotoxicity by using released LDH as a marker. Studies have shown that L. Plantarum MPL16 and CRL1506 and S. salivarius K12 do not induce damage and consequent release of LDH in cultured oral epithelial cells and human BECs (16HBE140-) (Cosseau et al. 2008, Islam et al. 2021), while S. pneumoniae is known to cause cell death in Calu-3 cells (Weight et al. 2023). Likewise, we observed no cytotoxic effects following probiotic exposure, although a modest but statistically significant increase in LDH secretion was observed following infection with S. pneumoniae 49619.
In conclusion, the present study demonstrated, the benign nature of L. rhamnosus and S. salivarius, isolated from the URT of healthy children, in in vitro cultures of primary NECs compared to S. pneumoniae. Moreover, we have highlighted the potential for further investigation into their ability to act as effective probiotics.
Acknowledgements
We would like to acknowledge Kirsty Short, Yanshan Zhu, and Kyra Cottrell from the University of Queensland and Saeideh Hajighasemi from Queensland University of Technology for their contributions towards sample collection and culturing of nasal epithelial cells. This study was approved by the Queensland University of Technology Human Research Ethics Committee (approval number: 2021000292).
Author contributions
Tejasri Yarlagadda (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing), Alison Carey (Methodology), Emily Bryan (Methodology), Flavia Huygens (Methodology), Prasad Yarlagadda (Resources), Diane Maresco-Pennisi (Resources), Andrea Coleman (Resources), Anders Cervin (Resources), Kirsten Spann (Conceptualization, Investigation, Methodology, Resources, Writing-review & editing)
Conflict of interest
All authors do not have a commercial or other association that might pose a conflict of interest.
Funding
This work was supported by the Children's Hospital Foundation PhD Scholarship Grant (grant number: 50280).
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
