Oral Presentation 3 (IMAG-AND Futures)

2:00-2:20 pm               “Towards a multiscale synthesis of skin regeneration: Single cell transcriptomics and modeling reveal functional fibroblast heterogeneity in skin wounds

Christian Guerrero-Juarez, UC-Irvine

Christian Fernando Guerrero-Juarez (photo)BIO: Christian Guerrero-Juarez was born and raised in Morelia, Michoacán, México. HE received dual B.S./B.A. Degrees in Biology and Biochemistry at California State University, San Bernardino in 2012, and completed his M.S. and Ph.D. Degrees at University of California, Irvine in 2018 where he studied cellular reprogramming during skin regeneration. As a graduate student, he received funding through the MBRS-IMSD and the NSF Graduate Research Fellowship Programs and was  awarded the 2015 Susan V. Bryant Graduate Fellowship Award, 2017 Howard A. Schneiderman Graduate Fellowship Award, and Premio Michoacano de la Juventud 2018 – Mérito Académico (2018 Michoacán Youth Prize – Academic Merit) – the highest distinction given by the State Government of Michoacán, Mexico to excelling young professionals in their respective fields. Christian joined the Nie laboratory in July 2018 as a UC Irvine Chancellor’s ADVANCE Postdoctoral Fellow. Currently, he is an NSF-Simons Center for Multiscale Cell Fate Research Fellow and was a finalist in the 2019 Howard Hughes Medical Institute Hanna H. Gray Postdoctoral Fellowship Program competition. His research interests include cell fate choice and cellular reprogramming during injury and disease, and he seeks to use a multiscale approach to better understand these processes as they are realized in vivo.

 

Comment

Single cell decision making

Very interesting study! Does your analysis predict differentiation between fibroblast subtypes and whether myofibroblast differentiation is reversible? Any particular pathways implicated in these cell decisions?

Submitted by Jeff Saucerman (not verified) on Tue, 03/17/2020 - 14:14

Thank you for your question!…

Thank you for your question! Yes, we performed pseudo-temporal ordering of all fibroblasts in our data set and overlaid RNA velocity vectors of those cells to better inform their trajectories. We found three distinct trajectories, all pointing toward regions high in Acta2/Tagln, which are markers of myofibroblasts (i.e. contractile cells). These results suggest that the distinct fibroblast subtypes all have the potential to acquire a fibrotic state. We also identified genes and transcription factors that are pseudo-temporally dependent, and found also TGF-beta genes to be changing. Currently, we are developing a pipeline for inferring cell-cell interactions, network structure and function of cells based on transcript abundance in single cells and one of the goals is to see what pathways change in these and other processes in latent space. If you have any further questions you can contact me at cfguerre@uci.edu.

Submitted by Christian Guer… (not verified) on Tue, 03/17/2020 - 15:24

In reply to by Jeff Saucerman (not verified)

Question

The result that upregulating one specific lineage of fibroblasts can affect wound healing outcome is very exciting. What are the factors limiting the natural function of these in hypertrophic scars? Are they migration limited, metabolism limited, or proliferation limited? Thanks! Guy

Submitted by Guy Genin, Was… (not verified) on Tue, 03/17/2020 - 14:16

Thank you for your question!…

Thank you for your question! Keloid and hypertrophic scars are fibrotic in nature, they have high amounts of collagen-producing fibroblasts (i.e. myofibroblasts) and, hence, have high amounts of collagen bundles. This could inherently affect the micro- and macro-environment of the wound and diminish skin integrity. Currently, we are co-opting a mouse model of keloid scars to investigate other aspects that may be limiting the natural function of non-pathological fibroblasts, such as metabolism.

Submitted by Christian Guer… (not verified) on Tue, 03/17/2020 - 15:30

In reply to by Guy Genin, Was… (not verified)

timing question

great topic - can you clarify why you picked 12 days to conduct the RNA analysis? Would cell fate and wound healing outcomes not be determined earlier in the process based on your introductory slides?

Submitted by David (not verified) on Tue, 03/17/2020 - 14:16

Thanks! Yes, we used day 12…

Thanks! Yes, we used day 12 wounds for a couple of reasons. First, this time point is prior hair follicle and fat regeneration, which occurs 15 and 21 days after pounding, respectively. Second, this time point is prior complete wound closure - which corresponds with active cellular migration into the wound bed. We therefore reasoned that during this time point there would be increased cellularity and we might see a large pool of heterogeneous cell populations.

Submitted by Christian Guer… (not verified) on Tue, 03/17/2020 - 15:45

In reply to by David (not verified)

Fibroblast Subtypes pathogenic?

Great talk! Thanks so much. Did you notice any particular fibroblast types that correlate with poor repair responses?

Submitted by Ben B (not verified) on Tue, 03/17/2020 - 14:16

Thank you for your question!…

Thank you for your question! This is an active area of research that I am currently pursuing.

Mouse vs. human

Impressive work! One question that comes up a lot is the difference in hair follicle density between mice and less hairy mammals like us. Comment?

Submitted by jmhaugh on Tue, 03/17/2020 - 14:17

Thank you for your question…

Thank you for your question. Yes, this is a recurring and important question. Mouse and human hair follicles (and also skin!) are very different not only in the density of follicles, but also in their structure and cycle dynamics. In our study, we use the wound induced hair neogenesis system as a model to study cell reprogramming during wound healing, and use hair follicle and fat regeneration as hallmarks of successful skin regeneration. Despite these differences, there have been reports of humans regenerating facial hairs after dermabrasion, and we ourselves have also shown that what we learned from mouse studies can be translated into humans by successfully reprogramming fat cells in vitro using scalp hair follicles and other molecules.

Submitted by Christian Guer… (not verified) on Tue, 03/17/2020 - 15:53

In reply to by jmhaugh

model vs data - different number of cell states

Interesting data and model. It seems that the complexity of the cellular model does not mimic the complexity you found in the single cell data. How to reconcile the two?

Submitted by raj_vadigepalli on Tue, 03/17/2020 - 14:17

Excellent question! We would…

Excellent question! We would like to incorporate distinct fibroblast subsets and immune cells into the model (i.e. based on gene expression differences from single cell RNA-seq, proliferation dynamics, etc) and determine if heterogeneity can better inform, predict and resolve wound healing outcomes.