Human lung organoids develop into adult airway-like structures directed by physico-chemical biomaterial properties

Human lung organoids develop into adult airway-like structures directed by physico-chemical biomaterial properties

Journal Pre-proof Human lung organoids develop into adult airway-like structures directed by physicochemical biomaterial properties Briana R. Dye, Ric...

4MB Sizes 0 Downloads 8 Views

Journal Pre-proof Human lung organoids develop into adult airway-like structures directed by physicochemical biomaterial properties Briana R. Dye, Richard L. Youngblood, Robert S. Oakes, Tadas Kasputis, Daniel W. Clough, Jason R. Spence, Lonnie D. Shea PII:

S0142-9612(20)30003-X

DOI:

https://doi.org/10.1016/j.biomaterials.2020.119757

Reference:

JBMT 119757

To appear in:

Biomaterials

Received Date: 9 July 2019 Revised Date:

15 November 2019

Accepted Date: 3 January 2020

Please cite this article as: Dye BR, Youngblood RL, Oakes RS, Kasputis T, Clough DW, Spence JR, Shea LD, Human lung organoids develop into adult airway-like structures directed by physico-chemical biomaterial properties, Biomaterials (2020), doi: https://doi.org/10.1016/j.biomaterials.2020.119757. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Published by Elsevier Ltd.

Human lung organoids develop into adult airway-like structures directed by physicochemical biomaterial properties Briana R. Dye1, Richard L. Youngblood1, Robert S. Oakes1, Tadas Kasputis1, Daniel W. Clough1, Jason R. Spence2, Lonnie D. Shea1 1. Department of Biomedical Engineering; 2. Department of Internal Medicine, University of Michigan, Ann Arbor, MI

1

Abstract Tissues derived from human pluripotent stem cells (hPSCs) often represent early stages of fetal development, but mature at the molecular and structural level when transplanted into immunocompromised mice. hPSC-derived lung organoids (HLOs) transplantation has been further enhanced with biomaterial scaffolds, where HLOs had improved tissue structure and cellular differentiation. Here, our goal was to define the physico-chemical biomaterial properties that maximally enhanced transplant efficiency, including features such as the polymer type, degradation, and pore interconnectivity of the scaffolds. We found that transplantation of HLOs on microporous scaffolds formed from poly(ethylene glycol) (PEG) hydrogel scaffolds inhibit growth and maturation, and the transplanted HLOs possessed mostly immature lung progenitors. On the other hand, HLOs transplanted on poly(lactide-co-glycolide) (PLG) scaffolds or polycaprolactone (PCL) led to tube-like structures that resembled both the structure and cellular diversity of an adult airway. Our data suggests that scaffold pore interconnectivity and polymer degradation contributed to the maturation, and we found that the size of the airway structures and the total size of the transplanted tissue was influenced by the material degradation rate. Collectively, these biomaterial platforms provide a set of tools to promote maturation of the tissues and to control the size and structure of the organoids.

2

Introduction Human lung organoid models facilitate the study of cell fate decisions during development and for modeling diseases such as cystic fibrosis and goblet cell metaplasia, and infections such as respiratory syncytial virus [1-8]. We have previously demonstrated that human pluripotent stem cell (hPSC)-derived human lung organoids (HLOs) possess a complex tissue structure in vitro, which includes both the epithelium and supporting tissue (cartilage, smooth muscle, fibroblasts) [1,9]. Notably, in vitro HLO cultures reflect the fetal airway, with adult airway-like structures generated only after in vivo transplantation [9]. Maturation following in vivo transplantation of HLOs reflects observations with numerous other organoid and hPSC-based systems [1,10-13],. While many other studies have shown successful transplantation of hPSC-derived tissues under the kidney capsule or other vascular sites within the murine host, HLOs required the assistance of a PLG microporous polymer scaffold to support engraftment and vascularization following transplant into the epididymal fat pad of immunocompromised mice. After 8 weeks, the transplanted HLO (tHLO) had airway-like structures that resembled native adult airways including proper cellular organization, epithelial cellular ratios and airway cell types. Airway-like structures were also surrounded by smooth muscle and possessed cartilage, as would be the case in the human airways [9]. However, these previous studies did not identify the polymer scaffolds design parameters that conferred an engraftment and growth advantage for HLOs. In this report, we investigated the physico-chemical properties of microporous scaffolds that support HLO maturation into airway structures. Polymers have different degradation rates and may have distinct interactions with the host, so microporous scaffold support of transplanted HLO

were tested

using diverse materials including

poly(lactide-co-glycolide)

(PLG),

polycaprolactone (PCL), and poly(ethylene glycol) (PEG). The interconnected pore size was varied, as well, through the initial scaffold fabrication and also through the degradation rate of

3

the polymers. For these material platforms, we investigated airway maturation, immune response, as well as overall explant and airway size. Identifying the biomaterial design parameters that influence airway maturation and structure will enable the development of platforms that can direct the structure to better model airway homeostasis and disease environments. Methods Maintenance of hESCs, generation of HLOs, and seeding on scaffolds H1 human embryonic stem cell (hESC) line (NIH registry #0043) and H9 (NIH registry #0062) was obtained from the WiCell Research Institute. H1 hESC line was used to derive all HLOs for these experiments except for Figure 2 where H9 hESC and H9 GFP hESC lines were used to derive HLOs. H9 GFP hESC line was generated by infecting H9 hESCs with pLenti PGK GFP Puro virus generated from the plasmid purchased from AddGene (Cat#: 19070)[14]. H9 GFP hESC clonal line was generated by puromycin selection flow cytometry analysis sorting (FACS) for GFP high expressing cells. All hESC lines were approved by the University of Michigan Human Pluripotent Stem Cell Research Oversight Committee. hESCs were maintained as previously described [15]. HLOs were derived as previously described [1]. Foregut spheroids, which grow into HLOs, were seeded on scaffolds as previously described [9]. Scaffold fabrication 75:25 PLG scaffolds were fabricated as previously described [9,16]. 85:15 (Resomer® RG 858 S, Poly(D,L-lactide-co-glycolide), Sigma, Cat#: 739979-1G) and 50:50 PLG (Resomer® RG 505, Poly(D,L-lactide-co-glycolide) ester terminated, MW: 54,000-69,000, Sigma,Cat#: 739960) were fabricated the same as 75:25 PLG scaffolds. 20% (w/v) 4-arm PEG maleimide, 20,000MW (JenChem, Cat#: A7029-1) hydrogels were fabricated as previously described[17]. PCL scaffolds were fabricated as previously described[18]. Large interconnected PCL were commercially bought (National Institute of Standards and Technology, Cat#:8394). Scaffold transplantation

4

Scaffolds were implanted as previously described [9]. Briefly, mice were anesthetized and prepped as for omental transplants. The epididymal fat pads of male 7–10 week old NOD-scid IL2Rgnull (NSG) were exposed using a lower midline incision. Scaffolds were then placed along the epididymal blood vessels and covered with epididymal fat. An intraperitoneal flush of Zosyn (100 mg/kg; Pfizer Inc.) was administered after which the incision was closed in 2-layers using absorbable suture. Mice were euthanized between 1-8 weeks post-transplant. Immunohistochemistry, hematoxylin and eosin stain (H&E), and imaging Immunostaining and H&E were carried out as previously described [19]. Antibody information and dilutions can be found in Table 1. All images and videos were taken on a Nikon A1 confocal microscope or the Zeiss Axio Observer.Z1. Flow Cytometry Cell disassociation and flow cytometry was previously described[18]. Antibodies used are as follow: Fluor® 700 anti-CD45 (1:125, clone 30-F11, Biolegend), V500 anti-CD11b (1:100, clone M1/70, BD Biosciences), FITC anti-Ly6C (1:100, clone HK.14, Biolegend), PE-Cy7 anti-F4/80 (1:80, clone BM8, Biolegend), APC anti-CD11c (1:80, clone N418, Biolegend), and Pacific Blue™ anti- Ly-6G/Ly-6C (Gr-1) (1:70, clone RB6-8C5, Biolegend). Quantification Airway diameters were measured using ImageJ software. The longest and shortest diameter was measured per airway structure and then averaged together. Explant size was measured using a ruler by placing the longest side of the tHLO on the ruler. Ki67+ cells and ECAD+, Ki67+ cells were quantified using a program developed in lab by Kevin Rychel, previously described[20]. Experimental replicates and statistics All experiments were done on at least three (N=3) independent biological samples for each experiment. All error bars represented SEM while the long bar represented the average. Statistical differences were assessed with Prism software using unpaired t-test.

5

Results PLG, PEG and PCL have varying extents of HLO derived airway maturation Microporous

scaffolds

with

similar

architectures

and

formed

from

either

75:25

(lactide:glycolide) PLG or 20% (w/v) 4-arm PEG-maleimide microporous scaffold were tested for their ability to support transplantation of foregut spheroids. Both scaffolds had pores ranging in size from 225 µm to 450µm, and were cylinder shaped with a diameter of 5mm diameter and a thickness of 2mm. PLG is a degradable, hydrophobic polyester that will adsorb proteins while PEG scaffolds are non-degradable hydrogels and were formed with or without the adhesion peptide RGD. PLG and PEG scaffolds were seeded with foregut spheroids and cultured for 7 days in vitro, during which time the foregut spheroids grew to fill the pores (Figure 1A). Scaffolds were then transplanted into the epididymal fat pad of NSG mice. This highly vascularized implant site [21] is accessible by a minimally invasive surgery, has a large surface area [22] and the presence of pro-angiogenesis cytokines [23], which has supported the use of this site for transplantation. After 8 weeks, tissue was found within and around the PLG scaffold, and histological examination revealed that the tHLO tissue contained airway-like structures. The majority of the PLG scaffold should be degraded by this time point [24,25] and, accordingly, the material was not detected in histological sections (Figure 1C). Growth of the transplanted PLG scaffolds contrasted with the PEG-seeded scaffolds. 8 weeks after transplantation, PEG scaffolds appeared intact and the spheroids remained within the pores independent of whether the scaffold was modified with RGD (Figure 1B). We hypothesized that the individual HLOs would grow together on the surface of the PEG scaffold and form airway-like structures. However, the transplanted spheroids remained within the separate pores. We investigated the possibility that smaller airways formed within the pores of the PEG scaffold;

6

however, the PEG hydrogels only had clusters of cells within the pores and did not possess tissue resembling airway structures (Figure 1C). The tissue within the PEG pores was derived from both spheroids and host (murine) cells, as we observed cells expressing the humanspecific mitochondria marker (huMITO) along with huMITO-negative cells. huMITO+ cells coexpressed NKX2.1, an early lung epithelial–specific transcription factor[26,27], yet no mature cell types or airway-like structures were observed in the PEG explants (Figure 1D). In comparison, the organized airway structures observed in the PLG explants expressed the lung marker NKX2.1 and were huMITO+. Further characterization of the lung tissue was previously described in Dye et al. [9]. We next tested the ability of PCL scaffolds to support transplantation and lung organoid growth in vivo. Both PLG and PCL are polyester polymers that degrade predominantly by simple hydrolysis of the ester bond into acidic monomers. However, due to its higher molecular weight and higher hydrophobicity, PCL has a slower degradation rate than PLG [28,29]. PCL scaffolds were seeded with spheroids and transplanted for 8 weeks. Similar to the PEG, PCL scaffolds were still intact after 8 weeks in vivo (Figure 1C). No organized epithelial structures were observed within the pores of the PCL grafts but the scaffolds had clusters of cells similar to what is observed in the PEG transplants (Figure 1C). On the other hand, there were airway-like structures that formed on the outside of the scaffolds where tHLOs had expanded. Collectively, these results suggest that the polyester polymers (PLG, PCL) supported the spheroid engraftment and development of airway structures after 8 weeks, but that failure of the scaffold to degrade prevented the growth and development of spheroids into airway-like structures. Initial immune response at microporous scaffolds may contribute to HLO responses: The initial immune response within the microporous scaffolds was investigated as a potential mechanism underlying the differential maturation on the various materials. Note that these studies were performed within NSG mice that lack an adaptive immune system in order to prevent rejection, yet these mice retain innate immune cells that respond to transplantation of

7

the construct. Lung organoids were transplanted and collected after 1 week in vivo, with analysis of the innate immune response. PEG tHLOs had a significantly greater percentage of leukocytes (CD45+) than PLG and PCL tHLOs (Figure 2A), which indicates greater cell recruitment from the host tissue. From the CD45+ population, the PEG tHLOs had significantly higher percent of CD11b+GR1+Ly6c- cells, often referred to as myeloid derived suppressor cells, and significantly less CD11c+F4-80- (dendritic) cells and Ly6c+F4-80- (monocyte) cells compared to PLG and PCL tHLOs (Figure 2B). No differences were observed between the PLG and PCL scaffolds. While the immune cell population at the graft may not fully recapitulate the immune-response of biomaterials in an immunocompetent environment, these data suggest that the immune response could be a contributing factor to the inhibition of HLO maturation in PEG scaffolds relative to PLG and PCL scaffolds. HLO fusion during formation of airway-like structures We subsequently investigated the contribution of HLO interaction in adjacent pores to the formation of airway structures, which was motivated by the observations of airway structures forming on the surface of slow degrading PCL scaffolds, but not within the pores. In order to analyze the fusion of multiple organoids into an airway structure, scaffolds were seeded with +

-

HLOs constitutively expressing GFP (GFP ) and GFP HLOs. Following culture of HLOs in PLG and PCL scaffolds for 1 week in vitro, we observed pores seeded with GFP+ HLOs that were -

adjacent to pores containing GFP HLOs (Figure 3A). Scaffold were transplanted and retrieved after 4 and 8 weeks in vivo. After 4 weeks in vivo, airway structures were not present, yet there -

were populations expressing the lung marker NKX2.1 that were either GFP+ or GFP indicating both HLO populations survived and successfully generated lung progenitors (Figure 3B). After 8 weeks, airway structures were observed in PLG and on the outside of PCL tHLOs, and these structures contained airway-like structures that possessed multiciliated cells (ACTTUB+) and -

basal cells (P63+). These airway structures had both GFP+ and GFP cells both in PCL and PLG

8

tHLOs (Figure 3C), indicating that the airway structures form by the HLOs fusing from adjacent pores in both the surface pores of PCL and in the degrading pores of the PLG tHLOs. Scaffold degradation affects the HLO derived airway size We next investigated the contribution of polymer degradation to the number of airway-like structures with the hypothesis that faster degradation would permit more HLOs fusing in adjacent pores, thus, support larger airway-like structures. Previously, the 75:25 PLG scaffolds were used to transplant HLOs and were thoroughly characterized in Dye et al. 2016 [9]. Yet, now with the use of multiple types of polymers with varying degrees of degradation, we quantified the impact on airway diameter size. We first investigated polymers with faster and slower degradation rates than 75:25 PLG. The airway diameters in the PCL tHLOs trended towards being slightly smaller (276 µm) compared to 75:25 PLG tHLOs (333 µm, Figure 4A). We also transplanted HLOs onto 85:15 PLG polymer which degrades at a rate intermediate of that between 75:25 PLG and PCL[30]. The 85:15 PLG tHLOs had significantly smaller airway diameter (224µm, p=0.049) than the 75:25 PLG (Figure 4A, C). In addition, the 85:15 PLG tHLO had a similar phenotype to the PCL tHLO, with the airway-like structures present adjacent to the scaffold and the tissue within the pores remaining as cell clusters (Figure 4A, C). We then fabricated a faster degrading polymer scaffold consisting of 50:50 PLG to further investigate whether degradation could support larger airway-like structures [30]. The 50:50 PLG tHLOs trended towards larger airways (414µm) relative to the 75:25 PLG (Figure 4A,C), though the difference was not significant. Collectively, the size of the airway structures was influenced by the degradation rate of the scaffold, with slower degrading polymers leading to smaller airwaylike structures. We next tested the hypothesis that an increase in the size of pore interconnections could increase HLO fusion while creating larger airway structure, as HLO fusion may be limited for slow degrading materials that led to smaller airway structures. We obtained PCL scaffolds constructed by 3D printing rods of PCL in a cross-hatch pattern within a 5mm wide, 2mm tall

9

cylinder (same dimensions as the scaffold used in earlier experiments), which has large pore connections (300µm) relative to the PCL scaffolds described previously, which range from 10 – 100 µm [14]. In testing the large interconnected porous scaffolds, the size of the airway structures was similar between PCL scaffolds with varying pore interconnectivity (Figure 4A, B). These results suggest that while fusion of cells can be aided by adjacent pores and degradation to contribute to the formation of airway structures, pore interconnectivity does not seem to directly determine the size of airway-like structures. Controlling the tHLO explant size After 8 weeks, we found the size of the explant was a function of the polymer type and design. The size of the explant for HLOs transplanted on PLG scaffolds reached diameters up to 2.5 cm, which is five times the original scaffold diameter (5 mm). Relative to 75:25 PLG tHLOs, the other tHLO conditions had a significant reduction in explant size. The explant sizes in fast degrading 50:50 PLG, slow degrading 85:15, and large interconnected PCL scaffolds were 0.81 cm, 0.53 cm, and 0.65 cm respectively. The PCL tHLO explant size was significantly smaller than 75:25 PLG tHLOs, with the PCL explants having diameters in the range of 0.5 to 0.6 cm (Figure 5A). However, the 50:50 PLG tHLOs were significantly larger than the 85:15 PLG tHLO. All together, these data suggested that both the slow and fast degrading PLG caused a reduction in explant size, but the size reduction was more significant in the slow degrading polymers, 85:15 PLG and PCL. Both the PCL control and PCL large interconnected pores were similar in size (Figure 5A). Thus, the size of pore interconnections does not appear to significantly impact the explant size. Proliferation within the explant was subsequently investigated as a contributing factor to the explant size. A significant two-fold change in proliferation (Ki67+cells) in 75:25 PLG tHLO (19.5%) was observed relative to PCL tHLO (8.7%) (Figure 5B-C). During native lung development, both the branching airway epithelium and surrounding mesenchyme contain proliferating cells [19,31]. ECAD+ was used to differentiate between the mesenchyme and

10

epithelium structures. The increase in proliferation in PLG tHLOs was observed in both the clusters of epithelial cells marked by E-Cadherin (ECAD+) within airway-like structures, and the surrounding tissue (ECAD-). Interestingly, both in PLG and PCL tHLOs, proliferation was significantly greater in the tissue adjacent to the scaffold relative to the ECAD+ airway-like structures (Figure 5D).

Discussion In this report, we have demonstrated that the type of material and degradation of the microporous scaffold can affect lung airway formation, airway size, and explant size derived from transplanted HLOs (Table 2). Previously, 75:25 PLG microporous scaffolds were used to transplant HLOs into the epididymal fat pad [9]. Since no maturation occurred when HLOs were placed into the kidney capsule or sewn in the omentum of an immunocompromised mouse, we hypothesized that the tHLOs needed a surface to grow and expand on in order to mature in airway structures. We then tested this hypothesis and found that the HLOs did not require the support of the pores to form airway-like structures, but instead needed specific material properties to allow for the development of airway structures. More specifically, the degradation of the material affected airway size and overall explant size of the tHLO. As the airway structures formed, the individual HLOs fused together to form these structures evident by the GFP+ and GFP- HLOs forming into one airway. Interestingly, the buds forming together occurred when the scaffold held its shape (PCL) or degraded (75:25 PLG) over the 8 weeks in vivo. Thus, the degradation was not necessary for the HLOs to fuse together to form the airway structures; however, there were no structures within the pores of the scaffolds that held the scaffold structure during the 8 weeks (PCL and 85:15 PLG). The fusion of the HLOs took place either on the surface of the PCL scaffold or as the PLG scaffold degraded. After incorporating a larger interconnected porous design to the PCL scaffold, we observed airway

11

formation within the scaffold. Collectively, the HLOs could only form airway structures if they had the space to fuse and expand together into epithelial tubes. This need of expansion aligns with airway formation during lung development where the airways start as one epithelial bud that continually bifurcates and expands into the surrounding mesenchyme to ultimately form a network of airways [32,33]. The degradation of the material contributed to multiple properties of the organoids: airway size and overall explant size. Deriving variations in airway size will allow the study of airway diseases such as COPD and asthma in both large and small airway models [34]. An airway ranging in size from 200-350µm represents a 5th generation airway in a native human lung while the 4th generation ranges from 400-600µm [35]. Here by changing the degradation we were able to represent two types of airways, 4th and 5th generation size that are observed in the native adult lung. The data with PLG and PCL indicated that the airway size was maximal for 75:25 PLG, with slightly smaller structures for 50:50 PLG and 85:15 PLG, suggesting that degradation plays a role and that maximal size occurs at an intermediate rate of degradation. One mechanism by which degradation can influence airway size is through the fusion of organoids from adjacent pores. Polymer degradation would influence fusion by increasing the size of the interconnections between pores over time, which would allow for greater connectivity. Polymer degradation would also function to remove the polymer as a substrate for organoid development, which may also influence proliferation and maturation. Collectively, these results are consistent with the general idea that the polymer degradation should be matched with the rate of tissue formation. Multiple scaffolds maintained the ability to form mature airway structures, which did not form without the presence of a material, yet the explant size was a function of the scaffold properties. PCL scaffolds allowed for the formation of structures at the surface of the material, yet not within the pores of the scaffold. The increased size resulted from an increase of proliferation from the supporting tissue including the mesenchyme with a lesser extent increase of proliferation from

12

the organized epithelium. However, the increased size of the explant did not influence the size of the airway-like structures that formed. The scaffold properties influenced the proliferation of the progenitor cells, that subsequently influenced the overall size of the explant. A more controlled growth of the organoid would allow for longer studies to be performed, since the explant growth will not impede the mouse health. For instance, future studies could use HLOs generated from patient specific hPSCs lines that have Cystic Fibrosis (CF). Patients with CF have mutations in Cystic Fibrosis Transmembrane conductance Regulator (CFTR), which causes excess of mucus within the airways which leads to chronic infection and inflammation of the lung epithelium[36]. With this model, HLOs could be generated from CF patient specific hPCS and studied in the tHLO model that provides a human airway model. Since the explant size is smaller, longer studies can be conducted in order to understand the short and long-term effects of CF on the airway epithelium and surrounding tissue including smooth muscle, cartilage and vasculature. Microporous scaffolds composed of PEG did not support maturation over the 8 weeks and the HLOs remained as NKX2.1 progenitors. HLOs were seeded onto PEG hydrogels with and without modification with RGD, a fibronectin binding peptide, in order to investigate if ECM signaling may be a signal directing maturation. The presence of RGD peptide did not impact maturation, suggesting that either adhesion is not a limiting factor in HLO maturation or that the RGD is insufficient to trigger the necessary signaling cascades. The innate immune response, which is active in immunocompromised mice, differed significantly between PEG versus PLG and PCL. The increase in immune cell recruitment in PEG scaffolds could be due in part to hydrogel swelling, which creates a larger volume that can contain more immune cells. This difference in scaffold environments could be a contributing factor that influenced HLOs maturation since it is known that the immune system affects hPSCs and tissue regeneration including adult stem cells [37-39].

13

Overall, HLO maturation was supported by multiple microporous scaffolds that resulted from fusion of organoid clusters in adjacent pores. Our studies show the physico-chemical properties of the scaffold can be manipulated to influence the properties of explant, such as the number and size of airways structures and the size of the explant. The biomaterials, thus, provide a tool that may be capable of directing tissue formation from organoids for the purpose of modeling normal development, and also for modeling disease states. Specific to airways, controlling airway and total explant size will allow for new models for airway diseases such as asthma, COPD, and CF with the potential to perform long-term studies.

Data Availability The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.

References [1]

[2]

[3]

[4]

[5]

[6]

[7]

B.R. Dye, D.R. Hill, M.A. Ferguson, Y.-H. Tsai, M.S. Nagy, R. Dyal, et al., In vitro generation of human pluripotent stem cell derived lung organoids, Elife. 4 (2015). doi:10.7554/eLife.05098. A.J. Miller, D.R. Hill, M.S. Nagy, Y. Aoki, B.R. Dye, A.M. Chin, et al., In Vitro Induction and In Vivo Engraftment of Lung Bud Tip Progenitor Cells Derived from Human Pluripotent Stem Cells, Stem Cell Reports. (2017). doi:10.1016/j.stemcr.2017.11.012. Y.-W. Chen, S.X. Huang, A.L.R.T. de Carvalho, S.-H. Ho, M.N. Islam, S. Volpi, et al., A three-dimensional model of human lung development and disease from pluripotent stem cells, Nat. Cell Biol. 372 (2017) 3. doi:10.1038/ncb3510. T. Tadokoro, Y. Wang, L.S. Barak, Y. Bai, S.H. Randell, B.L.M. Hogan, IL-6/STAT3 promotes regeneration of airway ciliated cells from basal stem cells, Proceedings of the National Academy of Sciences. 111 (2014) E3641–9. doi:10.1073/pnas.1409781111. H. Danahay, A.D. Pessotti, J. Coote, B.E. Montgomery, D. Xia, A. Wilson, et al., Notch2 is required for inflammatory cytokine-driven goblet cell metaplasia in the lung, Cell Rep. 10 (2015) 239–252. doi:10.1016/j.celrep.2014.12.017. J.R. Rock, M.W. Onaitis, E.L. Rawlins, Y. Lu, C.P. Clark, Y. Xue, et al., Basal cells as stem cells of the mouse trachea and human airway epithelium, Proceedings of the National Academy of Sciences. 106 (2009) 12771–12775. doi:10.1073/pnas.0906850106. K.B. McCauley, F. Hawkins, M. Serra, D.C. Thomas, A. Jacob, D.N. Kotton,

14

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

Efficient Derivation of Functional Human Airway Epithelium from Pluripotent Stem Cells via Temporal Regulation of Wnt Signaling, Cell Stem Cell. 20 (2017) 844– 857.e6. doi:10.1016/j.stem.2017.03.001. A.J. Miller, Q. Yu, M. Czerwinski, Y.H. Tsai, R.C. bioRxiv, 2018, Basal stem cell fate specification is mediated by SMAD signaling in the developing human lung, Biorxiv.org. (n.d.). B.R. Dye, P.H. Dedhia, A.J. Miller, M.S. Nagy, E.S. White, L.D. Shea, et al., A bioengineered niche promotes in vivo engraftment and maturation of pluripotent stem cell derived human lung organoids, Elife. 5 (2016) 3025. doi:10.7554/eLife.19732. S.R. Finkbeiner, D.R. Hill, C.H. Altheim, P.H. Dedhia, M.J. Taylor, Y.-H. Tsai, et al., Transcriptome-wide Analysis Reveals Hallmarks of Human Intestine Development and Maturation In Vitro and In Vivo, Stem Cell Reports. (2015). doi:10.1016/j.stemcr.2015.04.010. J.G. Camp, F. Badsha, M. Florio, S. Kanton, T. Gerber, M. Wilsch-Bräuninger, et al., Human cerebral organoids recapitulate gene expression programs of fetal neocortex development, Proceedings of the National Academy of Sciences. 112 (2015) 15672–15677. doi:10.1073/pnas.1520760112. M. Takasato, P.X. Er, H.S. Chiu, B. Maier, G.J. Baillie, C. Ferguson, et al., Kidney organoids from human iPS cells contain multiple lineages and model human nephrogenesis, Nature. 526 (2015) 564–568. doi:10.1038/nature15695. S. Hrvatin, C.W. O'Donnell, F. Deng, J.R. Millman, F.W. Pagliuca, P. DiIorio, et al., Differentiated human stem cells resemble fetal, not adult, β cells, Proceedings of the National Academy of Sciences. 111 (2014) 3038–3043. doi:10.1073/pnas.1400709111. E. Campeau, V.E. Ruhl, F. Rodier, C.L. Smith, B.L. Rahmberg, J.O. Fuss, et al., A versatile viral system for expression and depletion of proteins in mammalian cells, PLoS ONE. 4 (2009) e6529. doi:10.1371/journal.pone.0006529. J.R. Spence, C.N. Mayhew, S.A. Rankin, M.F. Kuhar, J.E. Vallance, K. Tolle, et al., Directed differentiation of human pluripotent stem cells into intestinal tissue in vitro, Nature. 470 (2011) 105–109. doi:10.1038/nature09691. H. Blomeier, X. Zhang, C. Rives, M. Brissova, E. Hughes, M. Baker, et al., Polymer scaffolds as synthetic microenvironments for extrahepatic islet transplantation, Transplantation. 82 (2006) 452–459. doi:10.1097/01.tp.0000231708.19937.21. P.D. Rios, M. Skoumal, J. Liu, R. Youngblood, E. Kniazeva, A.J. García, et al., Evaluation of encapsulating and microporous nondegradable hydrogel scaffold designs on islet engraftment in rodent models of diabetes, Biotechnol. Bioeng. 115 (2018) 2356–2364. doi:10.1002/bit.26741. S.S. Rao, G.G. Bushnell, S.M. Azarin, G. Spicer, B.A. Aguado, J.R. Stoehr, et al., Enhanced Survival with Implantable Scaffolds That Capture Metastatic Breast Cancer Cells In Vivo, Cancer Research. 76 (2016) 5209–5218. doi:10.1158/00085472.CAN-15-2106. B.E. Rockich, S.M. Hrycaj, H.P. Shih, M.S. Nagy, M.A.H. Ferguson, J.L. Kopp, et al., Sox9 plays multiple roles in the lung epithelium during branching morphogenesis, Proceedings of the National Academy of Sciences. 110 (2013) E4456–64. doi:10.1073/pnas.1311847110. T. Kasputis, D. Clough, F. Noto, K. Rychel, B. Dye, L.D. Shea, Microporous Polymer Scaffolds for the Transplantation of Embryonic Stem Cell Derived Pancreatic Progenitors to a Clinically Translatable Site for the Treatment of Type I Diabetes, ACS Biomater Sci Eng. 4 (2018) 1770–1778. doi:10.1021/acsbiomaterials.7b00912.

15

[21]

[22] [23]

[24]

[25]

[26]

[27]

[28]

[29]

[30] [31]

[32] [33] [34]

[35]

[36] [37] [38] [39]

Y. Yasunami, P.E. Lacy, E.H. Finke, A new site for islet transplantation--a peritoneal-omental pouch, Transplantation. 36 (1983) 181–182. doi:10.1097/00007890-198308000-00014. C. Schmidt, Pancreatic islets find a new transplant home in the omentum, Nature Biotechnology. 35 (2017) 8–8. doi:10.1038/nbt0117-8. N.O. Litbarg, K.P. Gudehithlu, P. Sethupathi, J.A.L. Arruda, G. Dunea, A.K. Singh, Activated omentum becomes rich in factors that promote healing and tissue regeneration, Cell Tissue Res. 328 (2007) 487–497. doi:10.1007/s00441-006-03564. Y. Park, Y. Chen, L. Ordovas, C.M. Verfaillie, Hepatic differentiation of human embryonic stem cells on microcarriers, J. Biotechnol. 174 (2014) 39–48. doi:10.1016/j.jbiotec.2014.01.025. H.E. Williams, J. Huxley, M. Claybourn, J.B.D.A. stability, 2006, The effect of γirradiation and polymer composition on the stability of PLG polymer and microspheres, Elsevier. (n.d.). E.E. Morrisey, B.L.M. Hogan, Preparing for the First Breath: Genetic and Cellular Mechanisms in Lung Development, Dev. Cell. 18 (2010) 8–23. doi:10.1016/j.devcel.2009.12.010. A.J. Miller, J.R. Spence, In Vitro Models to Study Human Lung Development, Disease and Homeostasis, Physiology (Bethesda). 32 (2017) 246–260. doi:10.1152/physiol.00041.2016. H.-J. Sung, C. Meredith, C. Johnson, Z.S. Galis, The effect of scaffold degradation rate on three-dimensional cell growth and angiogenesis, Biomaterials. 25 (2004) 5735–5742. doi:10.1016/j.biomaterials.2004.01.066. D. Mondal, M.G.I.J. of, 2016, Polycaprolactone-based biomaterials for tissue engineering and drug delivery: Current scenario and challenges, Taylor & Francis. (n.d.). H.K. Makadia, S.S. Polymers, 2011, Poly lactic-co-glycolic acid (PLGA) as biodegradable controlled drug delivery carrier, Mdpi.com. (n.d.). B.R. Dye, A.J. Miller, J.R. Spence, How to Grow a Lung: Applying Principles of Developmental Biology to Generate Lung Lineages from Human Pluripotent Stem Cells, Curr Pathobiol Rep. 4 (2016) 47–57. doi:10.1007/s40139-016-0102-x. R.J. Metzger, O.D. Klein, G.R. Martin, M.A. Krasnow, The branching programme of mouse lung development, Nature. 453 (2008) 745–750. doi:10.1038/nature07005. E.L. Rawlins, The building blocks of mammalian lung development, Dev. Dyn. 240 (2010) 463–476. doi:10.1002/dvdy.22482. R. Athanazio, Airway disease: similarities and differences between asthma, COPD and bronchiectasis, Clinics (Sao Paulo). 67 (2012) 1335–1343. doi:10.6061/clinics/2012(11)19. T.A. Lewis, Y.-S. Tzeng, E.L. McKinstry, A.C. Tooker, K. Hong, Y. Sun, et al., Quantification of airway diameters and 3D airway tree rendering from dynamic hyperpolarized 3He magnetic resonance imaging, Magn Reson Med. 53 (2005) 474–478. doi:10.1002/mrm.20349. G.R. Cutting, Cystic fibrosis genetics: from molecular understanding to clinical application, Nat. Rev. Genet. 16 (2015) 45–56. doi:10.1038/nrg3849. A.B. Aurora, E.N. Olson, Immune Modulation of Stem Cells and Regeneration, Cell Stem Cell. 15 (2014) 14–25. doi:10.1016/j.stem.2014.06.009. C. Kizil, N. Kyritsis, M. Brand, Effects of inflammation on stem cells: together they strive? EMBO Rep. 16 (2015) 416–426. doi:10.15252/embr.201439702. J.I. Pearl, L.S. Kean, M.M. Davis, J.C. Wu, Pluripotent stem cells: immune to the immune system? Sci Transl Med. 4 (2012) 164ps25–164ps25.

16

doi:10.1126/scitranslmed.3005090.

Figure 1

17

Figure 2

18

Figure 3

19

Figure 4

20

Figure 5

21

Figure Captions: Figure 1 HLOs seeded on PLG, PEG, and PCL scaffolds affects airway structure formation. A) Approximately 50 foregut spheroids were seeded onto PLG and PEG with or without RGD. Wholemount images were taken after 1 week in culture. B) Scaffolds were transplanted into the epididymal fat pad of immunocompromised mice and retrieved 8 weeks later. The PLG tHLO grew up to 2.5cm while the tissue remained within the PEG scaffolds with or without RGD. C) The histology of the PLG and PCL tHLOs had organized pseudostratified epithelium resembling native airway epithelium (shown by black arrows). The PEG tHLOs showed intact scaffold (shown by orange arrows) and both the PEG and the tissue within the pores of the PCL had no organized epithelium but remained as clusters of cells (shown by blue arrows). D) Some of the clusters of cells within the PEG scaffold were lung marker NKX2.1+ and human mitochondria (huMITO)+. Scale bars for A-B:1mm, C:200µm, and D:100µm. Figure 2 The innate immune profile for PEG, PLG, and PCL tHLOs. Foregut spheroids were cultured on scaffolds for 1 week in vitro, transplanted into the mouse epididymal fat pad, and then retrieved after 1 week to observe the innate immune response. A) The PEG tHLOs had 39% CD45+ Leukocytes (N=6) compared to 17% CD45+ cells in PLG tHLOs (N=7) and 19% CD45+ cells in PCL tHLOs (N=7) P<.005 B) PEG, PCL and PLG and similar percent of macrophages (CD11b+F4-80+). PEG tHLOs had significantly higher MDSCs (CD11b+GR1+Ly6c-) compared to PLG and PCL tHLOs, *P<.05. In contrast, PEG tHLO had significantly lower dendritic cells (CD11c+F4-80-) and monocytes (Ly6c+F480-) *P<.05. All error bars represent SEM.

22

Figure 3 HLOs fuse together to form airway structures. A) GFP H9 hESC line and H9 hESC derived HLOs were seeded onto 75:25 PLG and PCL scaffolds. Wholemount images of the scaffolds cultured for 1 week in vitro had GFP+ HLOs next to a pore of GFP- HLOs indicated by the red arrow head in both PLG and PCL conditions. B) After transplanted for four weeks, the GFP+ cells were mixing with the GFP- cells and both expressed early lung marker NKX2.1 (red). No organized epithelial structures were observed for either scaffold. C) After transplanted for 8 weeks, airway structures formed and were comprised of GFP+ and GFP- cells in both PLG and PCL tHLOs. The airway structures expressed NKX2.1 (red). The fused GFP+ and GFPHLOs in both PLG and PCL had multiciliated cells labelled by acetylated tubulin (ACTTUB, white) and basal cell marker P63 (red). Scale bars for A-B: 200µm, B-C: 100µm, and C: 50µm, 10µm. Figure 4 The degradation rate of the scaffold affected airway diameter. A) The average measurement was taken of the longest and shortest diameter for each cross section of an airway structure in a 8wk tHLO. The 85:15 PLG tHLOs (224µm) had the significantly shorter diameter compared to the 75:25 control PLG (333µm) *P<.05. The 50:50 PLG tHLO had the longest diameter at 415µm. The PCL control and large interconnected PCL tHLOs had similar diameter at 277µm and 299µm respectively compared to the 85:15 PLG tHLO (224µm). All error bars represent SEM. B-C) Histology sections of PLG (75:25), large interconnected PCL, 85:15 PLG and 50:50 PLG represent the quantified sections. Scale bars B) 200µm and C) 400µm. Figure 5 The degradation rate of the scaffold affected explant size of tHLO. A) Overall, the 75:25 PLG had the largest explant size (1.18cm, N=6) after an 8wk transplant compared to both fast (50:50 PLG, N=5) and slow degrading (85:15 and PCL, N=5) *P<.05. The fast degrading 50:50 PLG tHLO was significantly larger than the 85:15 PLG tHLO explant *P<.005. B) Ki67+ cells (red) were present within the epithelial airway structures labelled with ECAD (green) and the surrounding tissue both in PCL and 75:15 PLG. C) 75:25 PLG (19.6% ±1.8%) had significantly more Ki67+ cells than PCL (8.7% ±1.8%). D) The 75:25 PLG had significantly more Ki67+ cells in the ECAD+ and ECAD- areas (ECAD+: 6.7 ±0.9, ECAD-: 13.8% 1.7) compared to PCL tHLO (ECAD+: 3.009 ±1.0, ECAD-: 6.4% 1.1). Both for PCL and PLG tHLOs there was significantly more Ki67+ cells in the surrounding tissue (ECAD-) compared to the organized epithelium (ECAD+). **P>.005, *P>.05 All error bars represent SEM. Tables:

Table 1: Primary and secondary antibody information Primary Antibody Chicken anti-GFP Mouse anti-Acetylated Tubulin (ACTTUB) Mouse anti-E-Cadherin (ECAD) Mouse anti- Human Mitochondria (huMITO) Mouse anti-PLUNC Rabbit anti-Cytokeratin5 (CK5) Rabbit anti-NKX2.1

Source Aves Lab

Catalog # GFP-1020

Dilution 1:500

Clone polyclonal

Sigma-Aldrich

T7451

1:1000

6-11B-1

BD Transduction Laboratories

610181

1:500

36/E-Cadherin

Millipore

MAB1273

1:500

113-1

R&D Systems

MAP1897

1:200

monoclonal

Abcam

ab53121

1:500

polyclonal

Abcam

ab76013

1:200

EP1584Y

23

Rabbit anti-P63 Secondary Antibody Donkey anti-chicken 488 Donkey anti-mouse 488 Donkey anti-mouse 647 Donkey anti-mouse Cy3 Donkey anti-rabbit 488 Donkey anti-rabbit Cy3

Santa Cruz Biotechnology Source Jackson Immuno Jackson Immuno Jackson Immuno Jackson Immuno Jackson Immuno Jackson Immuno

sc-8344

1:200

H-129

Catalog # 703-545-155 715-545-150 415-605-350 715-165-150 711-545-152 711-165-102

Dilution 1:500 1:500 1:500 1:500 1:500 1:500

Table 2: Physico-chemical properties of microporous scaffolds that support HLO maturation into airway structures

Scaffold Material

Pore Size

Interconnected Pore Size

Degradation Rate

Spheroid Engraftment (Fig.1)

Airway Diameter Size (Fig.4)

Tissue Explant Size (Fig.5)

PLG 75:25

250425µm

10-100µm

Fast

Throughout Scaffold

333µm

1.18cm

PLG 85:15

250425µm

10-100µm

Medium

Throughout Scaffold

224µm

0.53cm

PLG 50:50

250425µm

10-100µm

Very Fast

Throughout Scaffold

415µm

0.81cm

PEG

250425µm

10-100µm

N/A

Within the Pores

No Airway Formation

N/A

PEG-RGD

250425µm

10-100µm

N/A

Within the Pores

No Airway Formation

N/A

PCL

250425µm

10-100µm

Slow

Within the Pores

277µm

0.65cm

Large Interconnected PCL

300µm

300µm

Slow

Within the Pores

299µm

0.5cm

24

Elsevier Editorial System(tm) for Biomaterials Manuscript Draft Manuscript Number: jbmt49472R1 Title: Human lung organoids develop into adult airway-like structures directed by physico-chemical biomaterial properties Article Type: FLA Original Research Section/Category: Biomaterials and Regenerative Medicine (BRM) Keywords: lung, organoids, scaffolds, stem cells Corresponding Author: Professor Lonnie D. Shea, Ph.D. Corresponding Author's Institution: University of Michigan First Author: Briana R Dye, Ph.D. Order of Authors: Briana R Dye, Ph.D.; Richard L Youngblood; Robert S Oakes, Ph.D.; Tadas Kasputis, Ph. D.; Daniel W Clough; Jason R Spence, Ph.D.; Lonnie D. Shea, Ph.D. Abstract: Tissues derived from human pluripotent stem cells (hPSCs) often represent early stages of fetal development, but mature at the molecular and structural level when transplanted into immunocompromised mice. hPSCderived lung organoids (HLOs) transplantation has been further enhanced with biomaterial scaffolds, where HLOs had improved tissue structure and cellular differentiation. Here, our goal was to define the physicochemical biomaterial properties that maximally enhanced transplant efficiency, including features such as the polymer type, degradation, and pore interconnectivity of the scaffolds. We found that transplantation of HLOs on microporous scaffolds formed from poly(ethylene glycol) (PEG) hydrogel scaffolds inhibit growth and maturation, and the transplanted HLOs possessed mostly immature lung progenitors. On the other hand, HLOs transplanted on poly(lactide-co-glycolide) (PLG) scaffolds or polycaprolactone (PCL) led to tube-like structures that resembled both the structure and cellular diversity of an adult airway. Our data suggests that scaffold pore interconnectivity and polymer degradation contributed to the maturation, and we found that the size of the airway structures and the total size of the transplanted tissue was influenced by the material degradation rate. Collectively, these biomaterial platforms provide a set of tools to promote maturation of the tissues and to control the size and structure of the organoids. Research Data Related to this Submission -------------------------------------------------There are no linked research data sets for this submission. The following reason is given: Data will be made available on request

Cover Letter Click here to download Cover Letter: Dye et al_CoverLetterBiomaterials.docx Lonnie D. Shea, Professor and Chair 1119 Carl A. Gerstacker, 2200 Bonisteel Blvd Ann Arbor, MI 48109-2099 (734) 764-7149, (F) 734-936-1905 [email protected]

November 15, 2019 Kam W. Leong, PhD Editor-in-Chief Biomaterials Dear Dr. Leong, My co-authors and I are pleased to submit our revised manuscript titled “Human lung organoids develop into adult airway-like structures directed by physico-chemical biomaterial properties” for your consideration in Biomaterials. This manuscript has been read and approved for resubmission by all authors, each of whom contributed in the experimental design and manuscript preparation. We further confirm that the order of authors listed in the manuscript has been approved by all authors and that there are no other persons who satisfied the criteria for authorship. In this manuscript, we demonstrate that biomaterial scaffolds can be used to investigate airway maturation, immune response, as well as overall explant and airway size in maturing human lung organoids derived from pluripotent stem cells. Our results indicate that these materials differentially support the formation of airway structure and size. We appreciate the time that the reviewers and the editor have committed to comment on the manuscript. The reviewers appeared highly supportive of the manuscript, yet had questions regarding the control conditions and the characterization of organoid maturation. A point by point response to the reviewer comments is available, which describes how the manuscript was revised to address clerical edits and provide additional context to the maturation of the organoids. We are grateful for the comments and believe that addressing the comments has enhanced the presentation of the data and impact of the manuscript. Collectively, these studies identify biomaterial design parameters that influence airway maturation and structure during organoid tissue formation, which can provide models for diseases such as asthma, chronic obstructive pulmonary disease (COPD), and cystic fibrosis with the potential to perform long-term studies. We confirm that we have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. In so doing, we have followed the regulations of our institutions concerning intellectual property. We further confirm that any aspect of the work covered in this manuscript that has involved experimental animals has been conducted with the ethical approval of all relevant bodies and that such approvals are acknowledged within the manuscript. This manuscript has not been published and is not under consideration for publication elsewhere. We confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome. We understand that the Corresponding Author is the sole contact for the Editorial process. He is responsible for communicating with the other authors about progress, submissions of revisions and final approval of proofs. We confirm that we have provided a current, correct email address which is accessible by the Corresponding Author and which has been configured to accept email from [email protected]. Thank you for your consideration! Sincerely,

Lonnie Shea

Response to reviewers Click here to download Response to reviewers: Dye et. al_Response to Reviewer Comments_Biomaterials.docx

Reviewer(s)' Comments to Author: Editorial Comments: We appreciate the feedback from the editor and have addressed the comments with the revisions detailed below. Reviewer #1 (Remarks to the Author): We thank Reviewer 1 for their thoughtful review and comments that have improved the manuscript. We have addressed the reviewer’s concerns as detailed below: 1. Introduction, line 36 - can the author better define or validate the term "indistinguishable", this seems a bit strong. This term was replaced to better convey the similar features our tHLOs had compared to a native model. On page 3, “After 8 weeks, the transplanted HLO (tHLO) had airway-like structures that resembled native adult airways including proper cellular organization, epithelial cellular ratios and airway cell types”. The similar features include airway cell markers for ciliated, club, goblet and basal cells along with the surrounding mesenchyme, smooth muscle and cartilage. In addition to cell types, the cellular organization and cellular ratio is similar to native adult airways. 2. Overall, the conclusion about interactions between biomaterial properties and tissue maturation seems correlative rather than causative. We agree and have clarified our conclusions. We modified our discussion on page 12 to convey this, “Multiple scaffolds maintained the ability to form mature airway structures, which did not form without the presence of a material, yet the explant size was a function of the scaffold properties.” 3. Why exactly do the authors think that pore size should influence cell maturation/differentiation? Our results suggest that the fusion of cells in adjacent pores contributes to the formation of airway structures, thus investigated if increased scaffold pore interconnectivity could enhance the ability for tissue development through cell fusion in adjacent pores. In order to investigate this effect, we increased the size of pore interconnections. We revised this section to convey this rationale more clearly on page 9, which now reads, “We next tested the hypothesis that an increase in the size of pore interconnections could increase HLO fusion while creating larger airway structure, as HLO fusion may be limited for slow degrading materials that led to smaller airway structures. We obtained PCL scaffolds constructed by 3D printing rods of PCL in a cross-hatch pattern within a 5mm wide, 2mm tall cylinder (same dimensions as the scaffold used in earlier experiments), which has large pore connections (300µm) relative to the PCL scaffolds described previously, which range from 10 – 100 µm [14]. In testing the large interconnected porous scaffolds, the size of the airway structures was similar between PCL scaffolds with varying pore interconnectivity (Figure 4A, B). These results suggest that while fusion of cells can be aided by adjacent pores and degradation to contribute to the formation of airway structures, pore interconnectivity does not seem to directly determine the size of airway-like structures.” 4. I don't believe that foregut endoderm can be appropriately termed "lung organoid" We agree. On page 6 of the results section, we clarify that “PLG and PEG scaffolds were seeded with foregut spheroids and cultured for 7 days in vitro, during which time the foregut spheroids grew to fill the pores (Figure 1A). Scaffolds were then transplanted into the epididymal fat pad of NSG mice. This highly vascularized implant site [21] is accessible by a minimally invasive surgery, has a large surface area [22] and the presence of pro-

angiogenesis cytokines [23], which has supported the use of this site for transplantation. After 8 weeks, tissue was found within and around the PLG scaffold, and histological examination revealed that the tHLO tissue contained airway-like structures.” This text describes how we transplant foregut spheroids and then they mature into lung organoids either in vitro (Dye et a. 2015) or in vivo (Dye et al. 2016). This transition was characterized in the previous publications (Dye et al. 2015 and Dye et al. 2016), but briefly the SOX2, ECAD expressing endoderm started expressing lung specific markers for instance NKX2.1, SFTPC, P63, and SOX9. 5. Why was the epididymal fat pad chosen as implantation site? We have added more rationale for choosing this site to address this question. On page 6, “Scaffolds were then transplanted into the epididymal fat pad of NSG mice. This highly vascularized implant site [21] is accessible by a minimally invasive surgery, has a large surface area [22] and the presence of pro-angiogenesis cytokines [23], which has supported the use of this site for transplantation.” 6. How are the authors defining "airway-like"? A ring of cell in a preformed circular space isn't really an airway. Can this be better analyzed in terms of cellular composition and function? Similarly, Nkx2.1 is not a lung/airway specific marker We previously characterized both the in vitro and in vivo lung organoids using lung specific markers and airway specific makers (Dye et al. 2015 and Dye et al. 2016). We added a sentence in the introduction on page 3, which now reads: “After 8 weeks, the transplanted HLO (tHLO) had airway-like structures that resembled native adult airways including proper cellular organization, epithelial cellular ratios and airway cell types. Airway-like structures were also surrounded by smooth muscle and possessed cartilage, as would be the case in the human airways [9].” We realize that NKX2.1 is a marker of thyroid and brain, but this is addressed in the first publication Dye at al. 2015. We also chose to use NKX2.1 as our lung marker because prior publications have shown NKX2.1 plays a central role in the regulation of lung morphogenesis before birth and is required for differentiation and gene regulation of diverse subsets of respiratory epithelial cells. In the normal human lung, NKX2-1 is selectively expressed in subsets of epithelial cells lining conducting airways and in alveolar type II cells in the lung periphery. We have added citations in on page 7 to reflect this, “huMITO+ cells co-expressed NKX2.1, an early lung epithelial–specific transcription factor[26,27], yet no mature cell types or airway-like structures were observed in the PEG explants (Figure 1D).” For airway markers, the transplanted HLOs were fully characterized in Dye et al. 2016 and shown to express airway markers for ciliated, club, basal, and goblet cells. We refer the reader to our prior characterization on page 7, “Further characterization of the lung tissue was previously described in Dye et al. [9].” 7. Why do the authors hypothesize the PEG to have higher immune cell recruitment relative to PLG? The overall goal for this study was to show that we found that the materials created different immune environments that may have contributed to organoid maturation. We hypothesize the increase in immune cell numbers in PEG scaffolds relative to PLG and PCL could be due in part to hydrogel swelling, which could create a larger volume that has the capacity to contain more immune cells. We have modified our discussion on page 13 to say, “The increase in immune cell recruitment in PEG scaffolds could be due in part to hydrogel swelling, which creates a larger volume that can contain more immune cells.” 8. Why do the authors hypothesize the proliferation rate in PCL is higher? Our results in Fig.5 indicate a higher proliferation rate in the PLG scaffolds, and we believe this is the condition to which the reviewer is referring to. In our discussion, we rationalize the increased proliferation rate by noting on page 12 that “polymer degradation would also

function to remove the polymer as a substrate for organoid development, which may also influence proliferation and maturation.” 9. The authors state that "airway structures" fuse and expand, as in development. And then bifurcates to continue tissue maturation. Was any branching or bifurcations noted in this model? We did not observe bifurcation, but did note this in the discussion on page 11 how this happens in native lung development, “Collectively, the HLOs could only form airway structures if they had the space to fuse and expand together into epithelial tubes. This need of expansion aligns with airway formation during lung development where the airways start as one epithelial bud that continually bifurcates and expands into the surrounding mesenchyme to ultimately form a network of airways [32,33].” 10. Was mucous production found in the "airway structures"? Yes, in the previous publication (Dye et al. 2016), we observed mucous production in the tHLOs. Reviewer #2 (Remarks to the Author): We thank Reviewer 2 for their thoughtful review and comments that will help improve this manuscript. We have addressed the reviewer concerns by making the following revisions: 1. Currently, various types of lung cells have been applied to make lung organoids and in vitro systems to mimic lung cancer and fibrosis as well as in the creation of bronchiolar or alveolar structures for specific proposes, including drug screening, tissue repair or therapy. In this study, however, the authors did not describe the purpose and rationale for only using hPSCs in organoid but without any procedure (or coculture) to induce the hPSCs differentiation to specific cell types of the airway or supporting tissue. Furthermore, while there are three main cell types in the human airway, including ciliated cells, goblet cells, and basal cells. However, in this study, authors only observed cilia cells (ACTTUB+) and basal cells (P63+) but not goblet cells in the newly forming tissue. The consequences of this missing cell type is substantial in the biology of lung. The authors should expand on why this cell is not derived and expand on the implications. Thank you for your comment. In prior publications, we describe the purpose of the hPSCderived culture method and characterized both the in vitro and in vivo lung organoids using lung specific markers and airway specific makers in Dye et al. 2015 and Dye et al. 2016. We added a sentence in our introduction on page 3 that references the most recent publication: “After 8 weeks, the transplanted HLO (tHLO) had airway-like structures that resembled native adult airways including proper cellular organization, epithelial cellular ratios and airway cell types. Airway-like structures were also surrounded by smooth muscle and possessed cartilage, as would be the case in the human airways [9].” For airway markers, the transplanted HLOs were fully characterized in Dye et al. 2016 and shown to express airway markers for ciliated, club, basal, and goblet cells. 2. In the Fig.2, authors used CD45 cell expression level to evaluate the scaffold's ability to induce innate immune responses. In mice, CD45 is expressed on T cells, B cells, dendritic cells, NK cells, Stem cells, macrophage/monocytes, and granulocytes. However, in NSG mice (immunodeficient mouse), macrophages and dendritic cells are defective, although they still exist in the mouse. In Fig. 2B, the expression level of macrophages and dendritic cells may not be a proper index to reflect the real immune-response of biomaterials in mice. Meanwhile, authors did not describe the detail of the animal experiment in the Materials & Methods part which leads to uncertainty regarding the results. We agree that investigating the immune response in a NSG mouse can limit our insights so we revised this section to acknowledge this on page 7, “Note that these studies were performed within NSG mice that lack an adaptive immune system in order to prevent

rejection, yet these mice retain innate immune cells that respond to transplantation of the construct,” and on page 8, “While the immune cell population at the graft may not fully recapitulate the immune-response of biomaterials in an immunocompetent environment, these data suggest that the immune response could be a contributing factor to the inhibition of HLO maturation in PEG scaffolds relative to PLG and PCL scaffolds.” We have also added to our Methods section more detail about the animal transplants to give more clarity on these studies. On page 4, “Scaffold transplantation: Scaffolds were implanted as previously described [9]. Briefly, mice were anesthetized and prepped as for omental transplants. The epididymal fat pads of male 7–10 week old NOD-scid IL2Rgnull (NSG) were exposed using a lower midline incision. Scaffolds were then placed along the epididymal blood vessels and covered with epididymal fat. An intraperitoneal flush of Zosyn (100 mg/kg; Pfizer Inc.) was administered after which the incision was closed in 2-layers using absorbable sutures. Mice were euthanized between 1-8 weeks post-transplant.” 3. In Fig. 4, the airway diameter seemingly has a positive correlation to PLG degrading rate. When PLG degrades faster, the diameter becomes wider. There are two concerns, first, why did the author use PCL but not PLG as a control? Second, the authors did not show or cite the reference about the degradation rate between PCL and PLG. For Figure 4, we use the PLG 75:25 condition as our PLG control. We have labeled this condition as a control to clarify this point. On page 7, we made a revision that now includes a citation and reads “Both PLG and PCL are polyester polymers that degrade predominantly by simple hydrolysis of the ester bond into acidic monomers. However, due to its higher molecular weight and higher hydrophobicity, PCL has a slower degradation rate than PLG [28,29]” 4.

Figure legends of Fig. 3, 4 and 5 are not consistent with the figures. Thank you for bringing this to our attention. The captions have been corrected so that they now are consistent with the figures.

5. Pg.6 Line 56 → Could use clearer terminology to describe varying pore sizes in initial experiments than, "Pore interconnectedness was varied through the initial fabrication and also through the degradation rate of the polymers" We have modified this sentence with clearer terminology so that, on page 3, it now reads “The interconnected pore size was varied, as well, through the initial scaffold fabrication and also through the degradation rate of the polymers.” a. Also a brief explanation of why the pore sizes were varied at this point and the value of this in the experiment. On page 9, we explain the purpose of increasing the size of the pore interconnections was to investigate if it could “increase HLO fusion while creating larger airway structure, as HLO fusion may be limited for slow degrading materials that led to smaller airway structures.” 6. Pg.7 Line 36: At this point in the article there has been no detailed description of what denotes an airway-like structure. The histology needs to better characterize in words. FIG.1 should have added arrows or lines to highlight relevant parts of the stain to help the reader We added a sentence in the Introduction section page 3 that details an airway-like structure: “After 8 weeks, the transplanted HLO (tHLO) had airway-like structures that resembled native adult airways including proper cellular organization, epithelial cellular ratios and airway cell types. Airway-like structures were also surrounded by smooth muscle and possessed cartilage, as would be the case in the human airways [9].” In Fig.1 of our results section, the focus of the H&E stains in Fig.1 is to convey where airway structures have developed within the scaffolds. We now characterize these stains using text on page 6 and 7 about how the images show “tissue was found within and around the PLG scaffold, and

histological examination revealed that the transplanted human lung organoid (tHLO) tissue contained airway-like structures (Figure 1C). The PEG hydrogels only had clusters of cells within the pores and did not possess tissue resembling airway structures (Figure 1C). No organized epithelial structures were observed within the pores of the PCL grafts but the scaffolds had clusters of cells similar to what is observed in the PEG transplants (Figure 1C). The PLG and PCL tHLOs had organized pseudostratified epithelium resembling native airway epithelium. The PEG tHLOs and the tissue within the pores of the PCL had no organized epithelium, but remained as clusters of cells.” We have also updated the histology figures with arrows that highlight the relevant parts of the stain as well. a. Pg.7 Line 40 → "PLG scaffold had degraded and the material was not detected in histological sections" No mention of assay or method determined for this claim. We did not carry out a degradation study as detecting the rate of scaffold degradation in vivo can be challenging. We have modified this statement to reference prior literature that indicates the PLG scaffold would be degrading by this time point. “The majority of the PLG scaffold should be degraded by this time point [24,25] and, accordingly, the material was not detected in histological sections (Figure 1C).” 7. Pg.8 Fig.1: Need some form of quantitative analysis on material degradation, is not enough to infer from images. Why is PLG the only one compared to PEG and not also PCL in A, B, and D → PCL in C seems random in this figure (even though it is the control in the following experiments, so it begs the question, why is it not relevant in the other sections of this figure?). Need native tissue/non-transplanted control to properly compare all results. Authors make claims that PLG is indistinguishable from native adult airways, so they need this control. In this study, the authors claimed that the degradation of the material affected airway size and overall explant size of the tHLO — however, Fig. 5A showed that 50:50 PLG has smaller explant size than 75:25 PLG and 85:15 PLG has smaller airway diameter than PCL control in Fig 4A. In Fig. 4 and 5, the degradation rate of PLG scaffolds did not remain consistent with the airway diameter and explant size. There is no solid result to support that the degradation is the critical factor to affect airway formation. Meanwhile, the authors did not present data or reference about the degradation rate of PLG used in the study. We have modified our statement about scaffold degradation to reference prior literature that suggests the PLG scaffold would be degrading by this time point. “The majority of the PLG scaffold should be degraded by this time point [24,25] and, accordingly, the material was not detected in histological sections (Figure 1C).” We previously characterized both the in vitro and in vivo lung organoids using lung specific markers and airway specific makers (Dye et al. 2015 and Dye et al. 2016). We compared all tissue to adult airway in Dye et al. 2016. We did not transplant this tissue however we did compare our tHLOs to adult airways. We added a sentence in our introduction referencing our prior publication with this information: “While many other studies have shown successful transplantation of hPSC-derived tissues under the kidney capsule or other vascular sites within the murine host, HLOs required the assistance of a PLG microporous polymer scaffold to support engraftment and vascularization following transplant into the epididymal fat pad of immunocompromised mice. After 8 weeks, the transplanted HLO (tHLO) had airway-like structures that resembled native adult airways including proper cellular organization, epithelial cellular ratios and airway cell types. Airway-like structures were also surrounded by smooth muscle and possessed cartilage, as would be the case in the human airways [9].” We have also modified our conclusion to reflect that the interactions between biomaterial properties and tissue maturation was more correlative rather than causative. In regards to PLG scaffold degradation, on page 9 of the results section, we made a revision that now includes a citation and reads “We also transplanted HLOs onto 85:15 PLG polymer which degrades at a rate intermediate of that between 75:25 PLG and PCL [30]. The 85:15 PLG tHLOs had significantly smaller airway diameter (224µm, p=0.049) than the 75:25 PLG (Figure 4A, C).

In addition, the 85:15 PLG tHLO had a similar phenotype to the PCL tHLO, with the airwaylike structures present adjacent to the scaffold and the tissue within the pores remaining as cell clusters (Figure 4A, C). We then fabricated a faster degrading polymer scaffold consisting of 50:50 PLG to further investigate whether degradation could support larger airway-like structures [30].” 8. Fig 3-5 captions are mixed between figures and need to be corrected (made staining very hard to interpret) Thank you for bringing this to our attention. The captions have been corrected so that they now are consistent with the figures. 9. Pg. 11 Fig. 3 → 4 week timepoint should also have a higher magnification similar to 8 weeks to enhance the argument and properly visualize timeline between 4 and 8 (C → D) The main purpose of showing our 4wk time point was to convey to the reader that, while tissue integration was occurring, the epithelial structure had not formed at this early time point. As a result, we felt the low magnification images were sufficient to support our claim. We provide higher magnification images at this 8 week time point to better characterize the airway that formed. 10. Pg. 12 Line 20 → "PCL was utilized instead of PLG because PCL has a lower melting point and is feasible to 3D print. In testing the larger pore scaffolds, the size of the airway structures was similar between both types of PCL scaffolds" a. All data up to this point have not indicated the importance of doing this experiment on PCL alone and just give the reader the impression that PLG is no longer relevant as it is not modular and conducive for translation b. This section only made remote sense after reading the "Scaffold degradation affects the HLO derived airway size" so needs to either be reworked or presented in another order i. Line 44 from next section that helped make argument clearer: Our aim for this study was to investigate the relationship between pore interconnectivity and tissue growth. We investigated this effect in a model that initially showed minimal tissue growth for ease of comparison. Thus, we used a PCL scaffold, which was a model that could easily be fabricated with a 100% pore interconnectivity using a 3D printer. We have revised this section to clarify this point and removed the translatability comparison to the PLG scaffold. This section on page 9 now reads, “We next tested the hypothesis that an increase in the size of pore interconnections could increase HLO fusion while creating larger airway structure, as HLO fusion may be limited for slow degrading materials that led to smaller airway structures. We obtained PCL scaffolds constructed by 3D printing rods of PCL in a cross-hatch pattern within a 5mm wide, 2mm tall cylinder (same dimensions as the scaffold used in earlier experiments), which has large pore connections (300µm) relative to the PCL scaffolds described previously, which range from 10 – 100 µm [14]. In testing the large interconnected porous scaffolds, the size of the airway structures was similar between PCL scaffolds with varying pore interconnectivity (Figure 4A, B). These results suggest that while fusion of cells can be aided by adjacent pores and degradation to contribute to the formation of airway structures, pore interconnectivity does not seem to directly determine the size of airway-like structures.” 11. "The airway diameters in the PCL tHLOs trended towards being slightly smaller (276 µm) compared to 75:25 PLG tHLOs" a. Page 12. Line 28 → "pore size does not seem to determine the size of airway-like structures." I'm not sure if I agree with this statement, since something else is obviously contributing to the lack of formation in PCL and you were unable to test PLG. The data presented at this point in the article do not necessarily support this claim.

This statement has been revised and moved to the “Scaffold degradation affects the HLO derived airway size” results section to present more data that can support this claim. We now suggest that degradation can play a role in air-way formation based on Fig.4 data then further investigate if increasing the interconnected pore size can increase HLO fusion while creating larger airway structure, as HLO fusion may be limited for slow degrading materials that led to smaller airway structures. We make the claim on page 9, “These results suggest that while fusion of cells can be aided by adjacent pores and degradation to contribute to the formation of airway structures, pore interconnectivity does not seem to directly determine the size of airway-like structures.” Thus, the article now claims that slow degradation could be playing a greater role in contributing to the lack of formation of structure in PCL rather than the interconnected pore size. 12. Need a table that compares the different materials, pore size, degradation rate, and a citation of which figures they are used in. Currently all the different biomaterials are confusing when you are trying to remember what their pore size and degradation rate are in relation to one another. Thank you for this suggestion. We have added a summarized results table that is now referenced in our discussion that should make these relationships between scaffold material and organoid development more clear. On page 11, the text now reads that “in this report, we have demonstrated that the type of material and degradation of the microporous scaffold can affect lung airway formation, airway size, and explant size derived from transplanted HLOs (Table 2).” 13.

Fig. 5 A, D → Current presentation of data is complicated

We have revised this section to improve the flow of information. We now asses all the PLG conditions, then compare them to the PCL conditions and, for Fig. 5D, we gave more rationale for our analysis. This section on page 10 now reads, “After 8 weeks, we found the size of the explant was a function of the polymer type and design. The size of the explant for HLOs transplanted on PLG scaffolds reached diameters up to 2.5 cm, which is five times the original scaffold diameter (5mm). Relative to 75:25 PLG tHLOs, the other tHLO conditions had a significant reduction in explant size. The explant sizes in fast degrading 50:50 PLG, slow degrading 85:15, and large interconnected PCL scaffolds were 0.81 cm, 0.53 cm, and 0.65 cm respectively. The PCL tHLO explant size was significantly smaller than 75:25 PLG tHLOs, with the PCL explants having diameters in the range of 0.5 to 0.6 cm (Figure 5A). However, the 50:50 PLG tHLOs were significantly larger than the 85:15 PLG tHLO. All together, these data suggested that both the slow and fast degrading PLG caused a reduction in explant size, but the size reduction was more significant in the slow degrading polymers, 85:15 PLG and PCL. Both the PCL control and PCL large interconnected pores were similar in size (Figure 5A). Thus, the size of pore interconnections does not appear to significantly impact the explant size. Proliferation within the explant was subsequently investigated as a contributing factor to the explant size. A significant two-fold change in proliferation (Ki67+cells) in 75:25 PLG tHLO (19.5%) was observed relative to PCL tHLO (8.7%) (Figure 5B-C). During native lung development, both the branching airway epithelium and surrounding mesenchyme contain proliferating cells [19,31]. ECAD+ was used to differentiate between the mesenchyme and epithelium structures. The increase in proliferation in PLG tHLOs was observed in both the clusters of epithelial cells marked by E-Cadherin (ECAD+) within airway-like structures, and the surrounding tissue (ECAD-). Interestingly, both in PLG and PCL tHLOs, proliferation was significantly greater in the tissue adjacent to the scaffold relative to the ECAD+ airway-like structures (Figure 5D).” In addition, we have added a table in the discussion section that summarizes these results, which we believe clarifies the presentation of data.

14. Pg. 15 Line 29 → Why is this the first time we are hearing E-cad as an airway-like marker. E-cad is not an airway-like marker that is accepted in lung biology. Thank you for your comment. We agree that ECAD is an epithelial marker. In this manuscript, ECAD labels the epithelium, for which prior staining with Nkx2.1 had indicated the tissue was lung (Dye et al. 2016, 2015). We and others (Chen et al 2013, Rawlins et al 2017) have employed this combination of Nkx2.1 and ECAD effectively labels lung epithelium. We have modified this paragraph to more accurately describe the use of ECAD. This paragraph on page 10 now reads, “The increase in proliferation in PLG tHLOs was observed in both the clusters of epithelial cells marked by E-Cadherin (ECAD+) within airway-like structures, and the surrounding tissue (ECAD-).” 15. More cell lines should be used in the current study not just mentioned in the discussion (CTFR) a. Pg. 17 Paragraph stating on Line 8 seems random in comparison to rest of discussion Since our focus was to understand the influence of varying biomaterial properties on tHLO development and maturation we did not incorporate additional cell lines in these studies. We used multiple cell lines in Dye et al 2015 and Dye et al 2016 that determined that all the cell lines behaved in a similar manner. With this information and our focus on biomaterials, we decided to use one cell line with multiple material platforms that could address the biomaterial design parameters. This analysis of multiple material platforms directed our submission to the journal Biomaterials. We do agree that using additional cell lines could provide more insight now that we have established the foundation in this manuscript. We modified our discussion section on page 12 to indicate potential future studies, and the text now reads, “A more controlled growth of the organoid would allow for longer studies to be performed, since the explant growth will not impede the mouse health. For instance, future studies could use HLOs generated from patient specific hPSCs lines that have Cystic Fibrosis (CF).”

Manuscript Click here to download Manuscript: Dye et al tHLO_Scaffold_Manuscript.docx

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Click here to view linked References

Human lung organoids develop into adult airway-like structures directed by physicochemical biomaterial properties Briana R. Dye1, Richard L. Youngblood1, Robert S. Oakes1, Tadas Kasputis1, Daniel W. Clough1, Jason R. Spence2, Lonnie D. Shea1 1. Department of Biomedical Engineering; 2. Department of Internal Medicine, University of Michigan, Ann Arbor, MI

1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Abstract Tissues derived from human pluripotent stem cells (hPSCs) often represent early stages of fetal development, but mature at the molecular and structural level when transplanted into immunocompromised mice. hPSC-derived lung organoids (HLOs) transplantation has been further enhanced with biomaterial scaffolds, where HLOs had improved tissue structure and cellular differentiation. Here, our goal was to define the physico-chemical biomaterial properties that maximally enhanced transplant efficiency, including features such as the polymer type, degradation, and pore interconnectivity of the scaffolds. We found that transplantation of HLOs on microporous scaffolds formed from poly(ethylene glycol) (PEG) hydrogel scaffolds inhibit growth and maturation, and the transplanted HLOs possessed mostly immature lung progenitors. On the other hand, HLOs transplanted on poly(lactide-co-glycolide) (PLG) scaffolds or polycaprolactone (PCL) led to tube-like structures that resembled both the structure and cellular diversity of an adult airway. Our data suggests that scaffold pore interconnectivity and polymer degradation contributed to the maturation, and we found that the size of the airway structures and the total size of the transplanted tissue was influenced by the material degradation rate. Collectively, these biomaterial platforms provide a set of tools to promote maturation of the tissues and to control the size and structure of the organoids.

2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Introduction Human lung organoid models facilitate the study of cell fate decisions during development and for modeling diseases such as cystic fibrosis and goblet cell metaplasia, and infections such as respiratory syncytial virus [1-8]. We have previously demonstrated that human pluripotent stem cell (hPSC)-derived human lung organoids (HLOs) possess a complex tissue structure in vitro, which includes both the epithelium and supporting tissue (cartilage, smooth muscle, fibroblasts) [1,9]. Notably, in vitro HLO cultures reflect the fetal airway, with adult airway-like structures generated only after in vivo transplantation [9]. Maturation following in vivo transplantation of HLOs reflects observations with numerous other organoid and hPSC-based systems [1,10-13],. While many other studies have shown successful transplantation of hPSC-derived tissues under the kidney capsule or other vascular sites within the murine host, HLOs required the assistance of a PLG microporous polymer scaffold to support engraftment and vascularization following transplant into the epididymal fat pad of immunocompromised mice. After 8 weeks, the transplanted HLO (tHLO) had airway-like structures that resembled native adult airways including proper cellular organization, epithelial cellular ratios and airway cell types. Airway-like structures were also surrounded by smooth muscle and possessed cartilage, as would be the case in the human airways [9]. However, these previous studies did not identify the polymer scaffolds design parameters that conferred an engraftment and growth advantage for HLOs. In this report, we investigated the physico-chemical properties of microporous scaffolds that support HLO maturation into airway structures. Polymers have different degradation rates and may have distinct interactions with the host, so microporous scaffold support of transplanted HLO

were

tested using

diverse

materials

including

poly(lactide-co-glycolide)

(PLG),

polycaprolactone (PCL), and poly(ethylene glycol) (PEG). The interconnected pore size was varied, as well, through the initial scaffold fabrication and also through the degradation rate of

3

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

the polymers. For these material platforms, we investigated airway maturation, immune response, as well as overall explant and airway size. Identifying the biomaterial design parameters that influence airway maturation and structure will enable the development of platforms that can direct the structure to better model airway homeostasis and disease environments. Methods Maintenance of hESCs, generation of HLOs, and seeding on scaffolds H1 human embryonic stem cell (hESC) line (NIH registry #0043) and H9 (NIH registry #0062) was obtained from the WiCell Research Institute. H1 hESC line was used to derive all HLOs for these experiments except for Figure 2 where H9 hESC and H9 GFP hESC lines were used to derive HLOs. H9 GFP hESC line was generated by infecting H9 hESCs with pLenti PGK GFP Puro virus generated from the plasmid purchased from AddGene (Cat#: 19070)[14]. H9 GFP hESC clonal line was generated by puromycin selection flow cytometry analysis sorting (FACS) for GFP high expressing cells. All hESC lines were approved by the University of Michigan Human Pluripotent Stem Cell Research Oversight Committee. hESCs were maintained as previously described [15]. HLOs were derived as previously described [1]. Foregut spheroids, which grow into HLOs, were seeded on scaffolds as previously described [9]. Scaffold fabrication 75:25 PLG scaffolds were fabricated as previously described [9,16]. 85:15 (Resomer® RG 858 S, Poly(D,L-lactide-co-glycolide), Sigma, Cat#: 739979-1G) and 50:50 PLG (Resomer® RG 505, Poly(D,L-lactide-co-glycolide) ester terminated, MW: 54,000-69,000, Sigma,Cat#: 739960) were fabricated the same as 75:25 PLG scaffolds. 20% (w/v) 4-arm PEG maleimide, 20,000MW (JenChem, Cat#: A7029-1) hydrogels were fabricated as previously described[17]. PCL scaffolds were fabricated as previously described[18]. Large interconnected PCL were commercially bought (National Institute of Standards and Technology, Cat#:8394). Scaffold transplantation

4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Scaffolds were implanted as previously described [9]. Briefly, mice were anesthetized and prepped as for omental transplants. The epididymal fat pads of male 7–10 week old NOD-scid IL2Rgnull (NSG) were exposed using a lower midline incision. Scaffolds were then placed along the epididymal blood vessels and covered with epididymal fat. An intraperitoneal flush of Zosyn (100 mg/kg; Pfizer Inc.) was administered after which the incision was closed in 2-layers using absorbable suture. Mice were euthanized between 1-8 weeks post-transplant. Immunohistochemistry, hematoxylin and eosin stain (H&E), and imaging Immunostaining and H&E were carried out as previously described [19]. Antibody information and dilutions can be found in Table 1. All images and videos were taken on a Nikon A1 confocal microscope or the Zeiss Axio Observer.Z1. Flow Cytometry Cell disassociation and flow cytometry was previously described[18]. Antibodies used are as follow: Fluor® 700 anti-CD45 (1:125, clone 30-F11, Biolegend), V500 anti-CD11b (1:100, clone M1/70, BD Biosciences), FITC anti-Ly6C (1:100, clone HK.14, Biolegend), PE-Cy7 anti-F4/80 (1:80, clone BM8, Biolegend), APC anti-CD11c (1:80, clone N418, Biolegend), and Pacific Blue™ anti- Ly-6G/Ly-6C (Gr-1) (1:70, clone RB6-8C5, Biolegend). Quantification Airway diameters were measured using ImageJ software. The longest and shortest diameter was measured per airway structure and then averaged together. Explant size was measured using a ruler by placing the longest side of the tHLO on the ruler. Ki67+ cells and ECAD+, Ki67+ cells were quantified using a program developed in lab by Kevin Rychel, previously described[20]. Experimental replicates and statistics All experiments were done on at least three (N=3) independent biological samples for each experiment. All error bars represented SEM while the long bar represented the average. Statistical differences were assessed with Prism software using unpaired t-test.

5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Results PLG, PEG and PCL have varying extents of HLO derived airway maturation Microporous

scaffolds

with

similar

architectures

and

formed

from

either

75:25

(lactide:glycolide) PLG or 20% (w/v) 4-arm PEG-maleimide microporous scaffold were tested for their ability to support transplantation of foregut spheroids. Both scaffolds had pores ranging in size from 225 µm to 450µm, and were cylinder shaped with a diameter of 5mm diameter and a thickness of 2mm. PLG is a degradable, hydrophobic polyester that will adsorb proteins while PEG scaffolds are non-degradable hydrogels and were formed with or without the adhesion peptide RGD. PLG and PEG scaffolds were seeded with foregut spheroids and cultured for 7 days in vitro, during which time the foregut spheroids grew to fill the pores (Figure 1A). Scaffolds were then transplanted into the epididymal fat pad of NSG mice. This highly vascularized implant site [21] is accessible by a minimally invasive surgery, has a large surface area [22] and the presence of pro-angiogenesis cytokines [23], which has supported the use of this site for transplantation. After 8 weeks, tissue was found within and around the PLG scaffold, and histological examination revealed that the tHLO tissue contained airway-like structures. The majority of the PLG scaffold should be degraded by this time point [24,25] and, accordingly, the material was not detected in histological sections (Figure 1C). Growth of the transplanted PLG scaffolds contrasted with the PEG-seeded scaffolds. 8 weeks after transplantation, PEG scaffolds appeared intact and the spheroids remained within the pores independent of whether the scaffold was modified with RGD (Figure 1B). We hypothesized that the individual HLOs would grow together on the surface of the PEG scaffold and form airway-like structures. However, the transplanted spheroids remained within the separate pores. We investigated the possibility that smaller airways formed within the pores of the PEG scaffold;

6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

however, the PEG hydrogels only had clusters of cells within the pores and did not possess tissue resembling airway structures (Figure 1C). The tissue within the PEG pores was derived from both spheroids and host (murine) cells, as we observed cells expressing the humanspecific mitochondria marker (huMITO) along with huMITO-negative cells. huMITO+ cells coexpressed NKX2.1, an early lung epithelial–specific transcription factor[26,27], yet no mature cell types or airway-like structures were observed in the PEG explants (Figure 1D). In comparison, the organized airway structures observed in the PLG explants expressed the lung marker NKX2.1 and were huMITO+. Further characterization of the lung tissue was previously described in Dye et al. [9]. We next tested the ability of PCL scaffolds to support transplantation and lung organoid growth in vivo. Both PLG and PCL are polyester polymers that degrade predominantly by simple hydrolysis of the ester bond into acidic monomers. However, due to its higher molecular weight and higher hydrophobicity, PCL has a slower degradation rate than PLG [28,29]. PCL scaffolds were seeded with spheroids and transplanted for 8 weeks. Similar to the PEG, PCL scaffolds were still intact after 8 weeks in vivo (Figure 1C). No organized epithelial structures were observed within the pores of the PCL grafts but the scaffolds had clusters of cells similar to what is observed in the PEG transplants (Figure 1C). On the other hand, there were airway-like structures that formed on the outside of the scaffolds where tHLOs had expanded. Collectively, these results suggest that the polyester polymers (PLG, PCL) supported the spheroid engraftment and development of airway structures after 8 weeks, but that failure of the scaffold to degrade prevented the growth and development of spheroids into airway-like structures. Initial immune response at microporous scaffolds may contribute to HLO responses: The initial immune response within the microporous scaffolds was investigated as a potential mechanism underlying the differential maturation on the various materials. Note that these studies were performed within NSG mice that lack an adaptive immune system in order to prevent rejection, yet these mice retain innate immune cells that respond to transplantation of

7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

the construct. Lung organoids were transplanted and collected after 1 week in vivo, with analysis of the innate immune response. PEG tHLOs had a significantly greater percentage of leukocytes (CD45+) than PLG and PCL tHLOs (Figure 2A), which indicates greater cell recruitment from the host tissue. From the CD45+ population, the PEG tHLOs had significantly higher percent of CD11b+GR1+Ly6c- cells, often referred to as myeloid derived suppressor cells, and significantly less CD11c+F4-80- (dendritic) cells and Ly6c+F4-80- (monocyte) cells compared to PLG and PCL tHLOs (Figure 2B). No differences were observed between the PLG and PCL scaffolds. While the immune cell population at the graft may not fully recapitulate the immune-response of biomaterials in an immunocompetent environment, these data suggest that the immune response could be a contributing factor to the inhibition of HLO maturation in PEG scaffolds relative to PLG and PCL scaffolds. HLO fusion during formation of airway-like structures We subsequently investigated the contribution of HLO interaction in adjacent pores to the formation of airway structures, which was motivated by the observations of airway structures forming on the surface of slow degrading PCL scaffolds, but not within the pores. In order to analyze the fusion of multiple organoids into an airway structure, scaffolds were seeded with +

-

HLOs constitutively expressing GFP (GFP ) and GFP HLOs. Following culture of HLOs in PLG and PCL scaffolds for 1 week in vitro, we observed pores seeded with GFP+ HLOs that were adjacent to pores containing GFP- HLOs (Figure 3A). Scaffold were transplanted and retrieved after 4 and 8 weeks in vivo. After 4 weeks in vivo, airway structures were not present, yet there were populations expressing the lung marker NKX2.1 that were either GFP+ or GFP- indicating both HLO populations survived and successfully generated lung progenitors (Figure 3B). After 8 weeks, airway structures were observed in PLG and on the outside of PCL tHLOs, and these structures contained airway-like structures that possessed multiciliated cells (ACTTUB+) and basal cells (P63+). These airway structures had both GFP+ and GFP- cells both in PCL and PLG

8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

tHLOs (Figure 3C), indicating that the airway structures form by the HLOs fusing from adjacent pores in both the surface pores of PCL and in the degrading pores of the PLG tHLOs. Scaffold degradation affects the HLO derived airway size We next investigated the contribution of polymer degradation to the number of airway-like structures with the hypothesis that faster degradation would permit more HLOs fusing in adjacent pores, thus, support larger airway-like structures. Previously, the 75:25 PLG scaffolds were used to transplant HLOs and were thoroughly characterized in Dye et al. 2016 [9]. Yet, now with the use of multiple types of polymers with varying degrees of degradation, we quantified the impact on airway diameter size. We first investigated polymers with faster and slower degradation rates than 75:25 PLG. The airway diameters in the PCL tHLOs trended towards being slightly smaller (276 µm) compared to 75:25 PLG tHLOs (333 µm, Figure 4A). We also transplanted HLOs onto 85:15 PLG polymer which degrades at a rate intermediate of that between 75:25 PLG and PCL[30]. The 85:15 PLG tHLOs had significantly smaller airway diameter (224µm, p=0.049) than the 75:25 PLG (Figure 4A, C). In addition, the 85:15 PLG tHLO had a similar phenotype to the PCL tHLO, with the airway-like structures present adjacent to the scaffold and the tissue within the pores remaining as cell clusters (Figure 4A, C). We then fabricated a faster degrading polymer scaffold consisting of 50:50 PLG to further investigate whether degradation could support larger airway-like structures [30]. The 50:50 PLG tHLOs trended towards larger airways (414µm) relative to the 75:25 PLG (Figure 4A,C), though the difference was not significant. Collectively, the size of the airway structures was influenced by the degradation rate of the scaffold, with slower degrading polymers leading to smaller airwaylike structures. We next tested the hypothesis that an increase in the size of pore interconnections could increase HLO fusion while creating larger airway structure, as HLO fusion may be limited for slow degrading materials that led to smaller airway structures. We obtained PCL scaffolds constructed by 3D printing rods of PCL in a cross-hatch pattern within a 5mm wide, 2mm tall

9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

cylinder (same dimensions as the scaffold used in earlier experiments), which has large pore connections (300µm) relative to the PCL scaffolds described previously, which range from 10 – 100 µm [14]. In testing the large interconnected porous scaffolds, the size of the airway structures was similar between PCL scaffolds with varying pore interconnectivity (Figure 4A, B). These results suggest that while fusion of cells can be aided by adjacent pores and degradation to contribute to the formation of airway structures, pore interconnectivity does not seem to directly determine the size of airway-like structures. Controlling the tHLO explant size After 8 weeks, we found the size of the explant was a function of the polymer type and design. The size of the explant for HLOs transplanted on PLG scaffolds reached diameters up to 2.5 cm, which is five times the original scaffold diameter (5 mm). Relative to 75:25 PLG tHLOs, the other tHLO conditions had a significant reduction in explant size. The explant sizes in fast degrading 50:50 PLG, slow degrading 85:15, and large interconnected PCL scaffolds were 0.81 cm, 0.53 cm, and 0.65 cm respectively. The PCL tHLO explant size was significantly smaller than 75:25 PLG tHLOs, with the PCL explants having diameters in the range of 0.5 to 0.6 cm (Figure 5A). However, the 50:50 PLG tHLOs were significantly larger than the 85:15 PLG tHLO. All together, these data suggested that both the slow and fast degrading PLG caused a reduction in explant size, but the size reduction was more significant in the slow degrading polymers, 85:15 PLG and PCL. Both the PCL control and PCL large interconnected pores were similar in size (Figure 5A). Thus, the size of pore interconnections does not appear to significantly impact the explant size. Proliferation within the explant was subsequently investigated as a contributing factor to the explant size. A significant two-fold change in proliferation (Ki67+cells) in 75:25 PLG tHLO (19.5%) was observed relative to PCL tHLO (8.7%) (Figure 5B-C). During native lung development, both the branching airway epithelium and surrounding mesenchyme contain proliferating cells [19,31]. ECAD+ was used to differentiate between the mesenchyme and

10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

epithelium structures. The increase in proliferation in PLG tHLOs was observed in both the clusters of epithelial cells marked by E-Cadherin (ECAD+) within airway-like structures, and the surrounding tissue (ECAD-). Interestingly, both in PLG and PCL tHLOs, proliferation was significantly greater in the tissue adjacent to the scaffold relative to the ECAD+ airway-like structures (Figure 5D).

Discussion In this report, we have demonstrated that the type of material and degradation of the microporous scaffold can affect lung airway formation, airway size, and explant size derived from transplanted HLOs (Table 2). Previously, 75:25 PLG microporous scaffolds were used to transplant HLOs into the epididymal fat pad [9]. Since no maturation occurred when HLOs were placed into the kidney capsule or sewn in the omentum of an immunocompromised mouse, we hypothesized that the tHLOs needed a surface to grow and expand on in order to mature in airway structures. We then tested this hypothesis and found that the HLOs did not require the support of the pores to form airway-like structures, but instead needed specific material properties to allow for the development of airway structures. More specifically, the degradation of the material affected airway size and overall explant size of the tHLO. As the airway structures formed, the individual HLOs fused together to form these structures evident by the GFP+ and GFP- HLOs forming into one airway. Interestingly, the buds forming together occurred when the scaffold held its shape (PCL) or degraded (75:25 PLG) over the 8 weeks in vivo. Thus, the degradation was not necessary for the HLOs to fuse together to form the airway structures; however, there were no structures within the pores of the scaffolds that held the scaffold structure during the 8 weeks (PCL and 85:15 PLG). The fusion of the HLOs took place either on the surface of the PCL scaffold or as the PLG scaffold degraded. After incorporating a larger interconnected porous design to the PCL scaffold, we observed airway

11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

formation within the scaffold. Collectively, the HLOs could only form airway structures if they had the space to fuse and expand together into epithelial tubes. This need of expansion aligns with airway formation during lung development where the airways start as one epithelial bud that continually bifurcates and expands into the surrounding mesenchyme to ultimately form a network of airways [32,33]. The degradation of the material contributed to multiple properties of the organoids: airway size and overall explant size. Deriving variations in airway size will allow the study of airway diseases such as COPD and asthma in both large and small airway models [34]. An airway ranging in size from 200-350µm represents a 5th generation airway in a native human lung while the 4th generation ranges from 400-600µm [35]. Here by changing the degradation we were able to represent two types of airways, 4th and 5th generation size that are observed in the native adult lung. The data with PLG and PCL indicated that the airway size was maximal for 75:25 PLG, with slightly smaller structures for 50:50 PLG and 85:15 PLG, suggesting that degradation plays a role and that maximal size occurs at an intermediate rate of degradation. One mechanism by which degradation can influence airway size is through the fusion of organoids from adjacent pores. Polymer degradation would influence fusion by increasing the size of the interconnections between pores over time, which would allow for greater connectivity. Polymer degradation would also function to remove the polymer as a substrate for organoid development, which may also influence proliferation and maturation. Collectively, these results are consistent with the general idea that the polymer degradation should be matched with the rate of tissue formation. Multiple scaffolds maintained the ability to form mature airway structures, which did not form without the presence of a material, yet the explant size was a function of the scaffold properties. PCL scaffolds allowed for the formation of structures at the surface of the material, yet not within the pores of the scaffold. The increased size resulted from an increase of proliferation from the supporting tissue including the mesenchyme with a lesser extent increase of proliferation from

12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

the organized epithelium. However, the increased size of the explant did not influence the size of the airway-like structures that formed. The scaffold properties influenced the proliferation of the progenitor cells, that subsequently influenced the overall size of the explant. A more controlled growth of the organoid would allow for longer studies to be performed, since the explant growth will not impede the mouse health. For instance, future studies could use HLOs generated from patient specific hPSCs lines that have Cystic Fibrosis (CF). Patients with CF have mutations in Cystic Fibrosis Transmembrane conductance Regulator (CFTR), which causes excess of mucus within the airways which leads to chronic infection and inflammation of the lung epithelium[36]. With this model, HLOs could be generated from CF patient specific hPCS and studied in the tHLO model that provides a human airway model. Since the explant size is smaller, longer studies can be conducted in order to understand the short and long-term effects of CF on the airway epithelium and surrounding tissue including smooth muscle, cartilage and vasculature. Microporous scaffolds composed of PEG did not support maturation over the 8 weeks and the HLOs remained as NKX2.1 progenitors. HLOs were seeded onto PEG hydrogels with and without modification with RGD, a fibronectin binding peptide, in order to investigate if ECM signaling may be a signal directing maturation. The presence of RGD peptide did not impact maturation, suggesting that either adhesion is not a limiting factor in HLO maturation or that the RGD is insufficient to trigger the necessary signaling cascades. The innate immune response, which is active in immunocompromised mice, differed significantly between PEG versus PLG and PCL. The increase in immune cell recruitment in PEG scaffolds could be due in part to hydrogel swelling, which creates a larger volume that can contain more immune cells. This difference in scaffold environments could be a contributing factor that influenced HLOs maturation since it is known that the immune system affects hPSCs and tissue regeneration including adult stem cells [37-39].

13

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Overall, HLO maturation was supported by multiple microporous scaffolds that resulted from fusion of organoid clusters in adjacent pores. Our studies show the physico-chemical properties of the scaffold can be manipulated to influence the properties of explant, such as the number and size of airways structures and the size of the explant. The biomaterials, thus, provide a tool that may be capable of directing tissue formation from organoids for the purpose of modeling normal development, and also for modeling disease states. Specific to airways, controlling airway and total explant size will allow for new models for airway diseases such as asthma, COPD, and CF with the potential to perform long-term studies.

Data Availability The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.

References [1]

[2]

[3]

[4]

[5]

[6]

[7]

B.R. Dye, D.R. Hill, M.A. Ferguson, Y.-H. Tsai, M.S. Nagy, R. Dyal, et al., In vitro generation of human pluripotent stem cell derived lung organoids, Elife. 4 (2015). doi:10.7554/eLife.05098. A.J. Miller, D.R. Hill, M.S. Nagy, Y. Aoki, B.R. Dye, A.M. Chin, et al., In Vitro Induction and In Vivo Engraftment of Lung Bud Tip Progenitor Cells Derived from Human Pluripotent Stem Cells, Stem Cell Reports. (2017). doi:10.1016/j.stemcr.2017.11.012. Y.-W. Chen, S.X. Huang, A.L.R.T. de Carvalho, S.-H. Ho, M.N. Islam, S. Volpi, et al., A three-dimensional model of human lung development and disease from pluripotent stem cells, Nat. Cell Biol. 372 (2017) 3. doi:10.1038/ncb3510. T. Tadokoro, Y. Wang, L.S. Barak, Y. Bai, S.H. Randell, B.L.M. Hogan, IL-6/STAT3 promotes regeneration of airway ciliated cells from basal stem cells, Proceedings of the National Academy of Sciences. 111 (2014) E3641–9. doi:10.1073/pnas.1409781111. H. Danahay, A.D. Pessotti, J. Coote, B.E. Montgomery, D. Xia, A. Wilson, et al., Notch2 is required for inflammatory cytokine-driven goblet cell metaplasia in the lung, Cell Rep. 10 (2015) 239–252. doi:10.1016/j.celrep.2014.12.017. J.R. Rock, M.W. Onaitis, E.L. Rawlins, Y. Lu, C.P. Clark, Y. Xue, et al., Basal cells as stem cells of the mouse trachea and human airway epithelium, Proceedings of the National Academy of Sciences. 106 (2009) 12771–12775. doi:10.1073/pnas.0906850106. K.B. McCauley, F. Hawkins, M. Serra, D.C. Thomas, A. Jacob, D.N. Kotton,

14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

Efficient Derivation of Functional Human Airway Epithelium from Pluripotent Stem Cells via Temporal Regulation of Wnt Signaling, Cell Stem Cell. 20 (2017) 844– 857.e6. doi:10.1016/j.stem.2017.03.001. A.J. Miller, Q. Yu, M. Czerwinski, Y.H. Tsai, R.C. bioRxiv, 2018, Basal stem cell fate specification is mediated by SMAD signaling in the developing human lung, Biorxiv.org. (n.d.). B.R. Dye, P.H. Dedhia, A.J. Miller, M.S. Nagy, E.S. White, L.D. Shea, et al., A bioengineered niche promotes in vivo engraftment and maturation of pluripotent stem cell derived human lung organoids, Elife. 5 (2016) 3025. doi:10.7554/eLife.19732. S.R. Finkbeiner, D.R. Hill, C.H. Altheim, P.H. Dedhia, M.J. Taylor, Y.-H. Tsai, et al., Transcriptome-wide Analysis Reveals Hallmarks of Human Intestine Development and Maturation In Vitro and In Vivo, Stem Cell Reports. (2015). doi:10.1016/j.stemcr.2015.04.010. J.G. Camp, F. Badsha, M. Florio, S. Kanton, T. Gerber, M. Wilsch-Bräuninger, et al., Human cerebral organoids recapitulate gene expression programs of fetal neocortex development, Proceedings of the National Academy of Sciences. 112 (2015) 15672–15677. doi:10.1073/pnas.1520760112. M. Takasato, P.X. Er, H.S. Chiu, B. Maier, G.J. Baillie, C. Ferguson, et al., Kidney organoids from human iPS cells contain multiple lineages and model human nephrogenesis, Nature. 526 (2015) 564–568. doi:10.1038/nature15695. S. Hrvatin, C.W. O'Donnell, F. Deng, J.R. Millman, F.W. Pagliuca, P. DiIorio, et al., Differentiated human stem cells resemble fetal, not adult, β cells, Proceedings of the National Academy of Sciences. 111 (2014) 3038–3043. doi:10.1073/pnas.1400709111. E. Campeau, V.E. Ruhl, F. Rodier, C.L. Smith, B.L. Rahmberg, J.O. Fuss, et al., A versatile viral system for expression and depletion of proteins in mammalian cells, PLoS ONE. 4 (2009) e6529. doi:10.1371/journal.pone.0006529. J.R. Spence, C.N. Mayhew, S.A. Rankin, M.F. Kuhar, J.E. Vallance, K. Tolle, et al., Directed differentiation of human pluripotent stem cells into intestinal tissue in vitro, Nature. 470 (2011) 105–109. doi:10.1038/nature09691. H. Blomeier, X. Zhang, C. Rives, M. Brissova, E. Hughes, M. Baker, et al., Polymer scaffolds as synthetic microenvironments for extrahepatic islet transplantation, Transplantation. 82 (2006) 452–459. doi:10.1097/01.tp.0000231708.19937.21. P.D. Rios, M. Skoumal, J. Liu, R. Youngblood, E. Kniazeva, A.J. García, et al., Evaluation of encapsulating and microporous nondegradable hydrogel scaffold designs on islet engraftment in rodent models of diabetes, Biotechnol. Bioeng. 115 (2018) 2356–2364. doi:10.1002/bit.26741. S.S. Rao, G.G. Bushnell, S.M. Azarin, G. Spicer, B.A. Aguado, J.R. Stoehr, et al., Enhanced Survival with Implantable Scaffolds That Capture Metastatic Breast Cancer Cells In Vivo, Cancer Research. 76 (2016) 5209–5218. doi:10.1158/00085472.CAN-15-2106. B.E. Rockich, S.M. Hrycaj, H.P. Shih, M.S. Nagy, M.A.H. Ferguson, J.L. Kopp, et al., Sox9 plays multiple roles in the lung epithelium during branching morphogenesis, Proceedings of the National Academy of Sciences. 110 (2013) E4456–64. doi:10.1073/pnas.1311847110. T. Kasputis, D. Clough, F. Noto, K. Rychel, B. Dye, L.D. Shea, Microporous Polymer Scaffolds for the Transplantation of Embryonic Stem Cell Derived Pancreatic Progenitors to a Clinically Translatable Site for the Treatment of Type I Diabetes, ACS Biomater Sci Eng. 4 (2018) 1770–1778. doi:10.1021/acsbiomaterials.7b00912.

15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

[21]

[22] [23]

[24]

[25]

[26]

[27]

[28]

[29]

[30] [31]

[32] [33] [34]

[35]

[36] [37] [38] [39]

Y. Yasunami, P.E. Lacy, E.H. Finke, A new site for islet transplantation--a peritoneal-omental pouch, Transplantation. 36 (1983) 181–182. doi:10.1097/00007890-198308000-00014. C. Schmidt, Pancreatic islets find a new transplant home in the omentum, Nature Biotechnology. 35 (2017) 8–8. doi:10.1038/nbt0117-8. N.O. Litbarg, K.P. Gudehithlu, P. Sethupathi, J.A.L. Arruda, G. Dunea, A.K. Singh, Activated omentum becomes rich in factors that promote healing and tissue regeneration, Cell Tissue Res. 328 (2007) 487–497. doi:10.1007/s00441-006-03564. Y. Park, Y. Chen, L. Ordovas, C.M. Verfaillie, Hepatic differentiation of human embryonic stem cells on microcarriers, J. Biotechnol. 174 (2014) 39–48. doi:10.1016/j.jbiotec.2014.01.025. H.E. Williams, J. Huxley, M. Claybourn, J.B.D.A. stability, 2006, The effect of γirradiation and polymer composition on the stability of PLG polymer and microspheres, Elsevier. (n.d.). E.E. Morrisey, B.L.M. Hogan, Preparing for the First Breath: Genetic and Cellular Mechanisms in Lung Development, Dev. Cell. 18 (2010) 8–23. doi:10.1016/j.devcel.2009.12.010. A.J. Miller, J.R. Spence, In Vitro Models to Study Human Lung Development, Disease and Homeostasis, Physiology (Bethesda). 32 (2017) 246–260. doi:10.1152/physiol.00041.2016. H.-J. Sung, C. Meredith, C. Johnson, Z.S. Galis, The effect of scaffold degradation rate on three-dimensional cell growth and angiogenesis, Biomaterials. 25 (2004) 5735–5742. doi:10.1016/j.biomaterials.2004.01.066. D. Mondal, M.G.I.J. of, 2016, Polycaprolactone-based biomaterials for tissue engineering and drug delivery: Current scenario and challenges, Taylor & Francis. (n.d.). H.K. Makadia, S.S. Polymers, 2011, Poly lactic-co-glycolic acid (PLGA) as biodegradable controlled drug delivery carrier, Mdpi.com. (n.d.). B.R. Dye, A.J. Miller, J.R. Spence, How to Grow a Lung: Applying Principles of Developmental Biology to Generate Lung Lineages from Human Pluripotent Stem Cells, Curr Pathobiol Rep. 4 (2016) 47–57. doi:10.1007/s40139-016-0102-x. R.J. Metzger, O.D. Klein, G.R. Martin, M.A. Krasnow, The branching programme of mouse lung development, Nature. 453 (2008) 745–750. doi:10.1038/nature07005. E.L. Rawlins, The building blocks of mammalian lung development, Dev. Dyn. 240 (2010) 463–476. doi:10.1002/dvdy.22482. R. Athanazio, Airway disease: similarities and differences between asthma, COPD and bronchiectasis, Clinics (Sao Paulo). 67 (2012) 1335–1343. doi:10.6061/clinics/2012(11)19. T.A. Lewis, Y.-S. Tzeng, E.L. McKinstry, A.C. Tooker, K. Hong, Y. Sun, et al., Quantification of airway diameters and 3D airway tree rendering from dynamic hyperpolarized 3He magnetic resonance imaging, Magn Reson Med. 53 (2005) 474–478. doi:10.1002/mrm.20349. G.R. Cutting, Cystic fibrosis genetics: from molecular understanding to clinical application, Nat. Rev. Genet. 16 (2015) 45–56. doi:10.1038/nrg3849. A.B. Aurora, E.N. Olson, Immune Modulation of Stem Cells and Regeneration, Cell Stem Cell. 15 (2014) 14–25. doi:10.1016/j.stem.2014.06.009. C. Kizil, N. Kyritsis, M. Brand, Effects of inflammation on stem cells: together they strive? EMBO Rep. 16 (2015) 416–426. doi:10.15252/embr.201439702. J.I. Pearl, L.S. Kean, M.M. Davis, J.C. Wu, Pluripotent stem cells: immune to the immune system? Sci Transl Med. 4 (2012) 164ps25–164ps25.

16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

doi:10.1126/scitranslmed.3005090.

Figure 1

17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Figure 2

18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Figure 3

19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Figure 4

20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Figure 5

21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Figure Captions: Figure 1 HLOs seeded on PLG, PEG, and PCL scaffolds affects airway structure formation. A) Approximately 50 foregut spheroids were seeded onto PLG and PEG with or without RGD. Wholemount images were taken after 1 week in culture. B) Scaffolds were transplanted into the epididymal fat pad of immunocompromised mice and retrieved 8 weeks later. The PLG tHLO grew up to 2.5cm while the tissue remained within the PEG scaffolds with or without RGD. C) The histology of the PLG and PCL tHLOs had organized pseudostratified epithelium resembling native airway epithelium (shown by black arrows). The PEG tHLOs showed intact scaffold (shown by orange arrows) and both the PEG and the tissue within the pores of the PCL had no organized epithelium but remained as clusters of cells (shown by blue arrows). D) Some of the clusters of cells within the PEG scaffold were lung marker NKX2.1 + and human mitochondria (huMITO)+. Scale bars for A-B:1mm, C:200µm, and D:100µm. Figure 2 The innate immune profile for PEG, PLG, and PCL tHLOs. Foregut spheroids were cultured on scaffolds for 1 week in vitro, transplanted into the mouse epididymal fat pad, and then retrieved after 1 week to observe the innate immune response. A) The PEG tHLOs had 39% CD45+ Leukocytes (N=6) compared to 17% CD45+ cells in PLG tHLOs (N=7) and 19% CD45+ cells in PCL tHLOs (N=7) P<.005 B) PEG, PCL and PLG and similar percent of macrophages (CD11b+F4-80+). PEG tHLOs had significantly higher MDSCs (CD11b+GR1+Ly6c-) compared to PLG and PCL tHLOs, *P<.05. In contrast, PEG tHLO had significantly lower dendritic cells (CD11c+F4-80-) and monocytes (Ly6c+F480-) *P<.05. All error bars represent SEM.

22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Figure 3 HLOs fuse together to form airway structures. A) GFP H9 hESC line and H9 hESC derived HLOs were seeded onto 75:25 PLG and PCL scaffolds. Wholemount images of the scaffolds cultured for 1 week in vitro had GFP+ HLOs next to a pore of GFP- HLOs indicated by the red arrow head in both PLG and PCL conditions. B) After transplanted for four weeks, the GFP+ cells were mixing with the GFP- cells and both expressed early lung marker NKX2.1 (red). No organized epithelial structures were observed for either scaffold. C) After transplanted for 8 weeks, airway structures formed and were comprised of GFP+ and GFP- cells in both PLG and PCL tHLOs. The airway structures expressed NKX2.1 (red). The fused GFP+ and GFPHLOs in both PLG and PCL had multiciliated cells labelled by acetylated tubulin (ACTTUB, white) and basal cell marker P63 (red). Scale bars for A-B: 200µm, B-C: 100µm, and C: 50µm, 10µm. Figure 4 The degradation rate of the scaffold affected airway diameter. A) The average measurement was taken of the longest and shortest diameter for each cross section of an airway structure in a 8wk tHLO. The 85:15 PLG tHLOs (224µm) had the significantly shorter diameter compared to the 75:25 control PLG (333µm) *P<.05. The 50:50 PLG tHLO had the longest diameter at 415µm. The PCL control and large interconnected PCL tHLOs had similar diameter at 277µm and 299µm respectively compared to the 85:15 PLG tHLO (224µm). All error bars represent SEM. B-C) Histology sections of PLG (75:25), large interconnected PCL, 85:15 PLG and 50:50 PLG represent the quantified sections. Scale bars B) 200µm and C) 400µm. Figure 5 The degradation rate of the scaffold affected explant size of tHLO. A) Overall, the 75:25 PLG had the largest explant size (1.18cm, N=6) after an 8wk transplant compared to both fast (50:50 PLG, N=5) and slow degrading (85:15 and PCL, N=5) *P<.05. The fast degrading 50:50 PLG tHLO was significantly larger than the 85:15 PLG tHLO explant *P<.005. B) Ki67+ cells (red) were present within the epithelial airway structures labelled with ECAD (green) and the surrounding tissue both in PCL and 75:15 PLG. C) 75:25 PLG (19.6% 1.8%) had significantly more Ki67+ cells than PCL (8.7% 1.8%). D) The 75:25 PLG had significantly more Ki67+ cells in the ECAD+ and ECAD- areas (ECAD+: 6.7 0.9, ECAD-: 13.8% 1.7) compared to PCL tHLO (ECAD+: 3.009 1.0, ECAD-: 6.4% 1.1). Both for PCL and PLG tHLOs there was significantly more Ki67+ cells in the surrounding tissue (ECAD-) compared to the organized epithelium (ECAD+). **P>.005, *P>.05 All error bars represent SEM. Tables:

Table 1: Primary and secondary antibody information Primary Antibody Chicken anti-GFP Mouse anti-Acetylated Tubulin (ACTTUB) Mouse anti-E-Cadherin (ECAD) Mouse anti- Human Mitochondria (huMITO) Mouse anti-PLUNC Rabbit anti-Cytokeratin5 (CK5) Rabbit anti-NKX2.1

Source Aves Lab

Catalog # GFP-1020

Dilution 1:500

Clone polyclonal

Sigma-Aldrich

T7451

1:1000

6-11B-1

BD Transduction Laboratories

610181

1:500

36/E-Cadherin

Millipore

MAB1273

1:500

113-1

R&D Systems

MAP1897

1:200

monoclonal

Abcam

ab53121

1:500

polyclonal

Abcam

ab76013

1:200

EP1584Y

23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Rabbit anti-P63 Secondary Antibody Donkey anti-chicken 488 Donkey anti-mouse 488 Donkey anti-mouse 647 Donkey anti-mouse Cy3 Donkey anti-rabbit 488 Donkey anti-rabbit Cy3

Santa Cruz Biotechnology Source Jackson Immuno Jackson Immuno Jackson Immuno Jackson Immuno Jackson Immuno Jackson Immuno

sc-8344

1:200

H-129

Catalog # 703-545-155 715-545-150 415-605-350 715-165-150 711-545-152 711-165-102

Dilution 1:500 1:500 1:500 1:500 1:500 1:500

Table 2: Physico-chemical properties of microporous scaffolds that support HLO maturation into airway structures

Scaffold Material

Pore Size

Interconnected Pore Size

Degradation Rate

Spheroid Engraftment (Fig.1)

Airway Diameter Size (Fig.4)

Tissue Explant Size (Fig.5)

PLG 75:25

250425µm

10-100µm

Fast

Throughout Scaffold

333µm

1.18cm

PLG 85:15

250425µm

10-100µm

Medium

Throughout Scaffold

224µm

0.53cm

PLG 50:50

250425µm

10-100µm

Very Fast

Throughout Scaffold

415µm

0.81cm

PEG

250425µm

10-100µm

N/A

Within the Pores

No Airway Formation

N/A

PEG-RGD

250425µm

10-100µm

N/A

Within the Pores

No Airway Formation

N/A

PCL

250425µm

10-100µm

Slow

Within the Pores

277µm

0.65cm

Large Interconnected PCL

300µm

300µm

Slow

Within the Pores

299µm

0.5cm

24

*Declaration of Interest Statement

Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: