Establishment and Characterization of Novel Canine Organoids with Organ-Specific Physiological Similarity
Curation statements for this article:-
Curated by eLife
eLife assessment
Organoids mimic the architecture and function of their cognate organs and have potential as replacements for animal models. Here the authors generated canine organoids from multiple adult tissues, including endometrium, lung, and pancreas, in addition to previously generated organoids from the kidney, bladder, and liver. However more methodological detail and functional characterization are required before this toolbox can be optimally utilized by wider scientific community.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Genetics and Genomics (eLife)
Abstract
Organoids are 3-dimensional (3D) stem cell-derived cell culture lines that offer a variety of technical advantages compared to traditional 2-dimensional (2D) cell cultures. Although murine models have proved useful in biomedical research, rodent models often fail to adequately mimic human physiology and disease progression, resulting in poor preclinical prediction of therapeutic drug efficacy and toxicity. With the advent of organoid technology, many of these challenges can be overcome. Previously, the use of canine organoids in drug testing and disease modeling was limited to organoids originating from the intestine, liver, kidney, and urinary bladder. Here, we report the cultivation, maintenance, and molecular characterization of three novel adult-stem cell-derived canine organoid cell lines, including the endometrium, lung, and pancreas, in addition to previously reported kidney, bladder, and liver organoids from two genetically related canines. Six tissues and organoid lines from each donor were characterized using bulk RNASeq, allowing for a unique, multi-organ comparison between these two individuals and identification of specific cell types such as glandular epithelial cells in endometrial organoids.
Article activity feed
-
Author Response
Reviewer #1 (Public Review):
The authors sought to establish canine tissue-specific organoids for propagation, storage and potential use in biomedical and translational medicine.
Strengths - The project is ambitious in aim, seeking to raise 6 tissue-specific, stem cell-derived organoid lines.
Weaknesses -
- While the manuscript refers to stem cell lines, no evidence of progressive organoid morphogenesis has been shown from undifferentiated single stem cells or stem cell clusters. This omission makes it difficult to distinguish true organoids from surviving pieces of parental tissue that the authors actually include within their cultures. The authors infer that high order tissue complexity can be generated within in short term 3D cultures. For example, their kidney organoids contained glomeruli, renal tubules and a …
Author Response
Reviewer #1 (Public Review):
The authors sought to establish canine tissue-specific organoids for propagation, storage and potential use in biomedical and translational medicine.
Strengths - The project is ambitious in aim, seeking to raise 6 tissue-specific, stem cell-derived organoid lines.
Weaknesses -
- While the manuscript refers to stem cell lines, no evidence of progressive organoid morphogenesis has been shown from undifferentiated single stem cells or stem cell clusters. This omission makes it difficult to distinguish true organoids from surviving pieces of parental tissue that the authors actually include within their cultures. The authors infer that high order tissue complexity can be generated within in short term 3D cultures. For example, their kidney organoids contained glomeruli, renal tubules and a Bowman's'capsule. These remarkable findings contrast with a previous study by Chen et al 2019 that showed kidney organoids had restricted morphogenic capacity, forming only simple epithelial dome-like structures (Chen et al 2019). Although the Chen study was cited, the major differences in study findings were not discussed. In the current study, no compelling evidence is provided for the integrated assembly of the glomerular microvascular capillary network, the glomerular epithelial capsule and complex tubular epithelial collecting ducts, during organoid growth.
Thank you, clarification was made regarding the differences between Chen et al. 2019 and our organoids in our revision of the manuscript (Lines 445-447). The sentence regarding glomeruli, renal tubules, and Bowman's capsules was modified to specifically state the morphological resemblance our organoids have to these structures. We further clarified in the text that we are not stating these structures are complete (glomerular microvascular capillary) (Line 236), as our data are too preliminary to support this statement. However, in future publications we are excited to complete more in-depth characterization and investigation, along with functional assessment. To aid in characterization, three immunohistochemistry (IHC) antibodies were added to Figure 2.
- The potential of the organoids for freezing, storage and re-culture is unclear from the data presented.
We did not present data regarding the re-culturing of organoids from this manuscript. However, we are working on additional publications which have already thawed and regrown multiple cell lines from this manuscript leading us to believe it is possible for all lines cultured in this manuscript. Further investigation into long term expression changes after thawing is warranted in future investigations.
- Organoid capacity for regenerative growth in xenograft models has not been tested.
We did not investigate this in the current manuscript, from our understanding, manuscripts which describe a new organoid model typically do not utilize xenografts to confirm the regenerative growth capacity. The use of organoids in xenograft models is an exciting avenue to explore in the future.
- Figure 4 lacks appropriate positive and negative tissue controls.
Please see Figure 5-figure supplement 1 for all negative control images. The tissues of origin in Figure 4 (now Figure 5) were used as positive controls for the antibody.
- Gene expression differences between tissues and organoids are inadequately explained.
Our apologies for the lack of clarity of our original manuscript. Differences of gene expression between tissues and organoids were compared in the revised Results section of each organ. To better describe the differences seen between tissues and organoids, information was added to the discussion elaborating on cell types present and missing from our samples (Lines 315-321).
- Methodological detail is sparse. It is not clear how tissue biopsies are obtained, what size they are and how they are processed for organoid preparation.
Our apologies. Information regarding biopsies was added in Experimental Procedures specifically in the Tissue collection section (Lines 509-510, 519-535). Additional details on protocols for organoid preparation and culturing were added to the Experimental Procedures and are cited in Gabriel et al. 2022 (Lines 535-548).
- The manuscript as a whole is poorly focussed and difficult to follow. The introduction is repetitive with only weak relevance to the main experiments.
We appreciate the reviewer’s concern. To better focus this manuscript, we re-ordered the introduction to be more linear and improve the focus regarding the main experiments. We hope our revisions will satisfy the reviewer.
Appraisal - The lack of morphogenesis and xenograft data undermines confidence that the authors have achieved their aims. The above concerns are also likely to hamper utility of the methods for the scientific community.
We appreciate the listed concerns. These novel organoid models are not limited to applications pertaining to xenografts. Our aim was to develop novel organoid lines that we believe can be of use to a variety of fields including pharmacology, virology, and basic research. The testing of these organoids in xenograft models is outside the scope of the current manuscript.
Reviewer #2 (Public Review):
Zydryski et al. develop a comprehensive toolbox of organ-specific canine organoids. Building on previous work on kidney, urinary bladder, and liver organoids, they now report on lung, endometrium, and pancreatic organoids; all six organoid lines are derived from two canines. The authors attempt to benchmark these organoids via histological, transcriptomic, and immunofluorescence characterization to their cognate organs. These efforts are a welcome development for the organoid field, broaden the scope of use to studies with canine models, and seek to establish robust standards. The organ specific RNAseq dataset is also likely to be useful to other researchers working with the canine model.
A key methodological advance would appear to be that the authors culture these organ-specific organoids using a common cell culture media. This is not the typical protocol in the organoid field; however, the authors do not provide enough information in the manuscript to evaluate if this is a good choice. Furthermore, it is likely that the authors were successful because they included additional tissue components in the co-culture for the organoids which might have provided the necessary tissue specific cues, but the methodological details to reproduce this and the technical evaluation of this approach are missing.
This is an excellent point and details were added to the methods section to better explain the embedding process in our revision of the manuscript (Lines 519-532). Your hypothesis about the tissue-specific cues is very intriguing and something we should explore in the future. Previous publications have isolated ECM from the native tissue (Giobbe et al. 2019), this may be a similar mechanism as you stated.
The authors also directly compare the transcriptional responses of the organoids with the organs, but this is a challenging enterprise given that the organoid models do not incorporate resident immune cells and typically are composed only of epithelial cells. This lack of an 'apples to apples' comparison might explain why in many cases the organoids and organs are highly divergent; however, it could also be that the common cell culture media did not lead to specific maturation of cell types.
We agree, this manuscript aimed to derive epithelial organoids, and we acknowledge the lack of all cell types present in the tissues. The comparison was meant to identify similarities (epithelial cells) and the current limitations of the organoid model. We added to the Discussion, specifically the Insights into organ-specific genes section to further clarify this point (Lines 315-321).
Reviewer #3 (Public Review):
Zydrski et al. describe the generation and characterization of multiple adult tissues from canines. While canine derived organoids could potentially be advantageous over murine and human organoids, the novelty of generation and characterization is limited, as organoid systems are now being rapidly genetically editing using CRISPR technologies and modeled within immunocompetent environments. Certain points limit my enthusiasm.
First, the authors do not support the use of serum (FBS) in their media and why they include the same growth and differentiation factors across all tissue types.
We added a sentence to the Discussion (Canine organoids as biomedical models) to further clarify the reasoning behind the inclusion of the same growth factors for all tissue types
“The use of the same media composition lends itself to future applications of co-culture or use in assembloid models where multiple organoid lines are combined and continued growth in a shared media is required”
As this media is based on canine intestinal organoid media, the FBS was included in case of potential applications require the co-culture of intestinal organoids.
Second, while bulk RNA sequencing data shows similarity per certain genes to the corresponding tissue, there is a lack of detailed characterization of what passage these organoids were harvested and how they change over time. Do they become more stem like and are they genetically stable?
The passage number of samples when they were harvested are listed in Supplemental file 1. The question of being genetically stable is an excellent point regarding organoids. We have not examined that yet in these canine organoids; however, we can leverage previous publications regarding organoids and how they are genomically stable over time regarding chromosome number and base pair changes, we added these citations into the introduction (Line 48). However, this current manuscript focuses on the derivation and initial characterization; future work will focus on the re-growth, genetic stability, and functional assessment of canine organoid lines.
Third, it would be important to demonstrate that these organoids can be genetically manipulated or be exposed to drugs and how they might be beneficial over murine and human organoids.
The genetic editing of twelve organoid lines is outside the scope of this paper and we plan to include this element in future publications. We believe that the organoids can be useful for veterinary medicine as well as being an important model or human disease as canines typically better represent humans better than mice (Lines 28-34, 77-95).
Fourth, the organoid complexity is not clear and cannot be ascertained from bulk RNA sequencing- for example, do kidney organoids recapitulate canonical markers at the protein level of proximal tubules, distal convoluted tubules, etc. Are different lung cells represented (AT1/AT2/club) and what is the composition of these cells? Why are these cells selected for?
Thank you. We agree that bulk RNA sequencing has its limitations when it comes to heterogenous cell populations. This was meant to give a first insight into whether the organoid lines resemble their tissue of origin, the addition of single cell RNA-sequencing in the future is worth investigating.
Fifth, as the authors note, methodically these canine organoids have been developed before from other tissues. For these reasons, my enthusiasm is diminished and unfortunately many of the necessary experiments for further consideration appear out of the scope of the study.
Three of these organoid lines have previously been published in canines. However, the growth and characterization of three novel organoid lines is included in this manuscript, while typically a manuscript focuses on one novel organoid line. Furthermore, unique to this study is the multi-organ comparisons of expression across both tissues and organoids from the same animal, with a biological replicate being a related individual which is unique to this study. In human and murine field, the organoid media must be adjusted to each individual organ the stem cells are isolated from. We show that our media composition, which is similar to that which previously supported hepatic and intestinal canine organoids, can now support organoids from six different tissues, bringing a novel approach to the field. To our knowledge, this is the most comprehensive comparison across tissue types of canine organoids. Additionally we have not seen any literature of the comparison of six different organoid lines from the same individual, with a related biological replicate in any other species.
-
eLife assessment
Organoids mimic the architecture and function of their cognate organs and have potential as replacements for animal models. Here the authors generated canine organoids from multiple adult tissues, including endometrium, lung, and pancreas, in addition to previously generated organoids from the kidney, bladder, and liver. However more methodological detail and functional characterization are required before this toolbox can be optimally utilized by wider scientific community.
-
Reviewer #1 (Public Review):
The authors sought to establish canine tissue-specific organoids for propagation, storage and potential use in biomedical and translational medicine.
Strengths - The project is ambitious in aim, seeking to raise 6 tissue-specific, stem cell-derived organoid lines.
Weaknesses -
1. While the manuscript refers to stem cell lines, no evidence of progressive organoid morphogenesis has been shown from undifferentiated single stem cells or stem cell clusters. This omission makes it difficult to distinguish true organoids from surviving pieces of parental tissue that the authors actually include within their cultures. The authors infer that high order tissue complexity can be generated within in short term 3D cultures. For example, their kidney organoids contained glomeruli, renal tubules and a Bowman's'capsule. …
Reviewer #1 (Public Review):
The authors sought to establish canine tissue-specific organoids for propagation, storage and potential use in biomedical and translational medicine.
Strengths - The project is ambitious in aim, seeking to raise 6 tissue-specific, stem cell-derived organoid lines.
Weaknesses -
1. While the manuscript refers to stem cell lines, no evidence of progressive organoid morphogenesis has been shown from undifferentiated single stem cells or stem cell clusters. This omission makes it difficult to distinguish true organoids from surviving pieces of parental tissue that the authors actually include within their cultures. The authors infer that high order tissue complexity can be generated within in short term 3D cultures. For example, their kidney organoids contained glomeruli, renal tubules and a Bowman's'capsule. These remarkable findings contrast with a previous study by Chen et al 2019 that showed kidney organoids had restricted morphogenic capacity, forming only simple epithelial dome-like structures (Chen et al 2019). Although the Chen study was cited, the major differences in study findings were not discussed. In the current study, no compelling evidence is provided for the integrated assembly of the glomerular microvascular capillary network, the glomerular epithelial capsule and complex tubular epithelial collecting ducts, during organoid growth.
2. The potential of the organoids for freezing, storage and re-culture is unclear from the data presented.
3. Organoid capacity for regenerative growth in xenograft models has not been tested.
4. Figure 4 lacks appropriate positive and negative tissue controls.
5. Gene expression differences between tissues and organoids are inadequately explained.
6. Methodological detail is sparse. It is not clear how tissue biopsies are obtained, what size they are and how they are processed for organoid preparation.
7. The manuscript as a whole is poorly focussed and difficult to follow. The introduction is repetitive with only weak relevance to the main experiments.
Appraisal - The lack of morphogenesis and xenograft data undermines confidence that the authors have achieved their aims. The above concerns are also likely to hamper utility of the methods for the scientific community.
-
Reviewer #2 (Public Review):
Zydryski et al. develop a comprehensive toolbox of organ-specific canine organoids. Building on previous work on kidney, urinary bladder, and liver organoids, they now report on lung, endometrium, and pancreatic organoids; all six organoid lines are derived from two canines. The authors attempt to benchmark these organoids via histological, transcriptomic, and immunofluorescence characterization to their cognate organs. These efforts are a welcome development for the organoid field, broaden the scope of use to studies with canine models, and seek to establish robust standards. The organ specific RNAseq dataset is also likely to be useful to other researchers working with the canine model.
A key methodological advance would appear to be that the authors culture these organ-specific organoids using a common …
Reviewer #2 (Public Review):
Zydryski et al. develop a comprehensive toolbox of organ-specific canine organoids. Building on previous work on kidney, urinary bladder, and liver organoids, they now report on lung, endometrium, and pancreatic organoids; all six organoid lines are derived from two canines. The authors attempt to benchmark these organoids via histological, transcriptomic, and immunofluorescence characterization to their cognate organs. These efforts are a welcome development for the organoid field, broaden the scope of use to studies with canine models, and seek to establish robust standards. The organ specific RNAseq dataset is also likely to be useful to other researchers working with the canine model.
A key methodological advance would appear to be that the authors culture these organ-specific organoids using a common cell culture media. This is not the typical protocol in the organoid field; however, the authors do not provide enough information in the manuscript to evaluate if this is a good choice. Furthermore, it is likely that the authors were successful because they included additional tissue components in the co-culture for the organoids which might have provided the necessary tissue specific cues, but the methodological details to reproduce this and the technical evaluation of this approach are missing.
The authors also directly compare the transcriptional responses of the organoids with the organs, but this is a challenging enterprise given that the organoid models do not incorporate resident immune cells and typically are composed only of epithelial cells. This lack of an 'apples to apples' comparison might explain why in many cases the organoids and organs are highly divergent; however, it could also be that the common cell culture media did not lead to specific maturation of cell types.
-
Reviewer #3 (Public Review):
Zydrski et al. describe the generation and characterization of multiple adult tissues from canines. While canine derived organoids could potentially be advantageous over murine and human organoids, the novelty of generation and characterization is limited, as organoid systems are now being rapidly genetically editing using CRISPR technologies and modeled within immunocompetent environments. Certain points limit my enthusiasm.
First, the authors do not support the use of serum (FBS) in their media and why they include the same growth and differentiation factors across all tissue types.
Second, while bulk RNA sequencing data shows similarity per certain genes to the corresponding tissue, there is a lack of detailed characterization of what passage these organoids were harvested and how they change over time. …
Reviewer #3 (Public Review):
Zydrski et al. describe the generation and characterization of multiple adult tissues from canines. While canine derived organoids could potentially be advantageous over murine and human organoids, the novelty of generation and characterization is limited, as organoid systems are now being rapidly genetically editing using CRISPR technologies and modeled within immunocompetent environments. Certain points limit my enthusiasm.
First, the authors do not support the use of serum (FBS) in their media and why they include the same growth and differentiation factors across all tissue types.
Second, while bulk RNA sequencing data shows similarity per certain genes to the corresponding tissue, there is a lack of detailed characterization of what passage these organoids were harvested and how they change over time. Do they become more stem like and are they genetically stable?
Third, it would be important to demonstrate that these organoids can be genetically manipulated or be exposed to drugs and how they might be beneficial over murine and human organoids.
Fourth, the organoid complexity is not clear and cannot be ascertained from bulk RNA sequencing- for example, do kidney organoids recapitulate canonical markers at the protein level of proximal tubules, distal convoluted tubules, etc. Are different lung cells represented (AT1/AT2/club) and what is the composition of these cells? Why are these cells selected for?
Fifth, as the authors note, methodically these canine organoids have been developed before from other tissues. For these reasons, my enthusiasm is diminished and unfortunately many of the necessary experiments for further consideration appear out of the scope of the study.
-
-