Search bioRxiv⌕ Search

Biology subjects

Taghizadeh, S.

Publications and source records attributed to Taghizadeh, S..

3 recordsLinked to original sources
bioRxiv2021

AxonDeep: Automated Optic Nerve Axon Segmentation in Mice with Deep Learning.

PurposeOptic nerve damage is the principal feature of glaucoma and contributes to vision loss in many diseases. In animal models, nerve health has traditionally been assessed by human experts that grade damage qualitatively or manually quantify axons from sampling limited areas from histologic cross sections of nerve. Both approaches are prone to variability and are time consuming. First-generation automated approaches have begun to emerge, but all have significant shortcomings. Here, we seek improvements through use of deep-learning approaches for segmenting and quantifying axons from cross sections of mouse optic nerve. MethodsTwo deep-learning approaches were developed and evaluated: (1) a traditional supervised approach using a fully convolutional network trained with only labeled data and (2) a semi-supervised approach trained with both labeled and unlabeled data using a generative-adversarial-network framework. ResultsFrom comparisons with an independent test set of images with manually marked axon centers and boundaries, both deep-learning approaches outperformed an existing baseline automated approach and similarly to two independent experts. Performance of the semi-supervised approach was superior and implemented into AxonDeep. ConclusionAxonDeep performs automated quantification and segmentation of axons from healthy appearing nerves, and those with mild to moderate degrees of damage, similar to that of experts without the variability and constraints associated with manual performance. Translational RelevanceUse of deep learning for axon quantification provides rapid, objective, and higher throughput analysis of optic nerve that would otherwise not be possible.

bioengineering↗
bioRxiv2021

Characterization of a novel Fgf10CreERT2 knock-in mouse line targeting postnatal lung Fgf10 lineages

Fgf10 is a key gene during development, homeostasis and repair after injury. We previously reported a Fgf10CreERT2 line (with the CreERT2 cassette inserted in frame with the start codon of exon 1), called thereafter Fgf10Ki-v1, to target Fgf10Pos cells. While this line allowed fairly efficient and specific labeling of Fgf10Pos cells during the embryonic stage, it failed to target these cells after birth, particularly in the postnatal lung, which has been the focus on our research. We report here the generation and validation of a new Fgf10CreERT2 (called thereafter Fgf10Ki-v2) with the insertion of the expression cassette in frame with the stop codon of exon 3. This new Fgf10Ki-v2 line exhibited comparable Fgf10 expression level to their wild type counterpart. However, a disconnection between the Fgf10 and the Cre expression was observed in Fgf10Ki-v2/+ lungs. In addition, lung and limb agenesis were observed in homozygous embryos suggesting a loss of Fgf10 functional allele in Fgf10Ki-v2 mice. Bio-informatics analysis shows that the 3UTR, where the CreERT2 cassette is inserted, contains numerous putative transcription factor binding sites. By crossing this line with tdTomato reporter line, we demonstrated that tdTomato expression faithfully recapitulated Fgf10 expression during development. Significantly, Fgf10Ki-v2 mouse is capable of significantly targeting Fgf10Pos cells in the adult lung. Therefore, despite the aforementioned limitations, this new Fgf10Ki-v2 line opens the way for future mechanistic experiments involving the postnatal lung.

developmental biology↗
bioRxiv2021

Characterization in mice of the stromal niche maintaining AT2 stem cell self-renewal in homeostasis and disease

Resident mesenchymal cells (rMCs defined as Cd31NegCd45NegEpcamNeg) control the self-renewal and differentiation of alveolar epithelial type 2 (AT2) stem cells in vitro. The identity of these rMCs is still elusive. Among them, Axin2Pos mesenchymal alveolar niche cells (MANCs), which are expressing Fgf7, have been previously described. We propose that an additional population of rMCs, expressing Fgf10 (called rMC-Sca1PosFgf10Pos) are equally important to maintain AT2 stem cell self-renewal. The alveolosphere model, based on the AT2-rMC co-culture in growth factor reduced Matrigel, was used to test the efficiency of different rMC subpopulations isolated by FACS from adult murine lung to sustain the self-renewal and differentiation of AT2 stem cells. We demonstrate that rMC-Sca1PosFgf10Pos cells are efficient to promote the self-renewal and differentiation of AT2 stem cells. Co-staining of adult lung for Fgf10 mRNA and Sftpc protein respectively, indicate that 28% of Fgf10Pos cells are located close to AT2 cells. Co-ISH for Fgf7 and Fgf10 indicate that these two populations do not significantly overlap. Gene arrays comparing rMC-Sca1PosAxin2Pos and rMC-Sca1PosFgf10Pos support that these two cell subsets express differential markers. In addition, rMC function is decreased in diabetic and obese ob/ob mutant compared to WT mice with a much stronger loss of function in males compared to females. In conclusion, rMC-Sca1PosFgf10Pos cells play important role in supporting AT2 stem cells self-renewal and differentiation. This result sheds a new light on the subpopulations of rMCs contributing to the AT2 stem cell niche in homeostasis and in the context of COVID-19 pathogenesis. Key messageO_ST_ABSWhat is already known about the subject?C_ST_ABSResident mesenchymal cells (rMCs defined as Cd31NegCd45NegEpcamNeg) control the self-renewal and differentiation of alveolar epithelial type 2 (AT2) stem cells in vitro. The identity of these rMCs is still elusive. Among them, Axin2Pos mesenchymal alveolar niche cells (MANCs), which are expressing Fgf7, have been previously described. What does this study add?Our study shows that an additional population of rMCs, expressing Fgf10 (called rMC-Sca1PosFgf10Pos) is equally important to maintain AT2 stem cell self-renewal. rMC-Sca1PosFgf10Pos are LipidToxHigh and are located close to AT2s. In addition, rMC-Sca1PosFgf10Pos cells support AT2 stem cell self-renewal and differentiation thereby identifying these cells as bone fide functional lipofibroblasts (LIFs). We have previously reported that LIF can transdifferentiate into activated MYF in the context of bleomycin-induced fibrosis in mice [1] and that activated MYF isolated from the lungs of end stage idiopathic fibrosis human patients can respond to Metformin to undergo transdifferentiation back to the LIF phenotype [2]. We also show that the function of rMCs-Sca1Pos is negatively impacted by gender and obesity, which represent two major aggravating factors for COVID-19 pathogenesis, leading to either death or major complications after infection recovery such as lung fibrosis. How might this impact on clinical practice and future development?By establishing that rMC-Sca1PosFgf10Pos are different from the MANCs, our study opens the way for a new key mesenchymal cell population that should be targeted to either prevent or reverse fibrosis. In addition, as this population maintains the AT2 stem cells self-renewal and differentiation, such targeting will also allow to progressively recover the loss in respiratory function.

developmental biology↗