Here, we present a protocol to engraft human brain organoids at multiple maturation stages into the chick chorioallantoic membrane (CAM). Brain organoids were grown following unguided standardized protocols.
Engrafting organoids into vascularized tissues in model animals, such as the immunodeficient mouse or chick embryo chorioallantoic membrane (CAM), has proven efficient for neovascularization modeling. The CAM is a richly vascularized extraembryonic membrane, which shows limited immunoreactivity, thus becoming an excellent hosting model for human origin cell transplants.
This paper describes the strategy to engraft human brain organoids differentiated at multiple maturation stages into the CAM. The cellular composition of brain organoids changes with time, reflecting the milestones of human brain development. We grafted brain organoids at relevant maturation stages: neuroepithelial expansion (18 DIV), early neurogenesis (60 DIV), and early gliogenesis (180 DIV) into the CAM of embryonic day (E)7 chicken embryos. Engrafted brain organoids were harvested 5 days later and their histological features were analyzed.
No histological signs of neovascularization in the grafted organoids or abnormal blood vessels adjacent to the graftings were detected. Moreover, remarkable changes were observed in the cellular composition of the grafted organoids, namely, an increase in the number of glial fibrillary acidic protein-positive-reactive astrocytes. However, the cytoarchitectural changes were dependent on the organoid maturation stage. Altogether, these results suggest that brain organoids can grow in the CAM, and they show differences in the cytoarchitecture depending on their maturation stage at grafting.
Human brain organoids are an emerging technique that allows us to recapitulate the early development of the human brain in vitro1,2,3. Nevertheless, one of the major limitations of this model is the lack of vascularization, which plays indispensable roles not only in brain homeostasis but also in brain development4. In addition to the delivery of oxygen and nutrients, accumulating evidence suggests that the vascular system of the brain regulates neural differentiation, migration, and synaptogenesis during development5,6. Therefore, there is an urgent need to establish reliable models that can provide the missing vascular signaling and structure to brain organoids, enhancing the complexity of human brain organoid generation7.
Among the proposed methods for vascularization, two main streamlines can be considered: organoid engrafting into a living organism and purely in vitro technologies co-culturing endothelial cells and neural cells8,9,10,11,12. Intracerebral transplantation in mice is costly and time-consuming, making other technologies relevant for simpler models. The chick chorioallantoic membrane (CAM) assay has been used extensively to study angiogenesis13,14,15. In the last decade, several groups have successfully engrafted different types of organoids, including kidney16,17, cardiac18, and tumor organoids19,20, into CAMs. Nevertheless, little is known about the efficacy, toxicity/rejection, physiological effect, and methods to engraft human brain organoids into the CAM. Another interesting and yet unexplored aspect is the formation of a chimeric blood-brain barrier (BBB) between the CAM and the organoid astrocytic interface. Previous pioneering work suggested the putative feasibility of generating a BBB in the CAM by transplanting astrocytes and astrocyte-conditioned medium21,22,23. However, mature astrocytes seem to be unable to achieve this24,25. Thus, the astrocyte-induced formation of the BBB remains debatable, and transplanting human brain organoids would allow us to shed light on this controversy.
This video article describes a protocol for an in ovo human brain organoid transplant into CAM that promotes growth, improvement, and vascularization, resulting in organoids that encompass histologically compatible BBB elements. Here, we present a protocol ensuring the survival of the chicken embryo and report on the permissivity of the CAM to sustain brain organoid growth.
The White Leghorn chicken (Gallus gallus) embryos were treated by following the Guide for the Care and Use of Laboratory Animals from the Institute of Laboratory Animals Resources, Commission of Life Sciences, National Research Council, USA, and the experiments were approved by the Council for Care and Use of Experimental Animals from the University of Barcelona.
1. Non-guided brain organoid preparation
2. Egg maintenance, development activation, and eggshell puncture
3. Brain organoid grafting
4. Transplanted organoid harvesting
5. Immunofluorescence
6. Hematoxylin and eosin (H&E) staining
NOTE: To prove that the grafting worked, perform H&E staining.
Selecting the embryo maturation schedule for the transplant
The experiment begins at D0 when fertilized eggs are incubated at 38 °C and 60% relative humidity. The chorioallantoic membrane (CAM) is a highly vascularized extraembryonic membrane that develops after egg incubation. It is formed by the fusion of the allantois and chorion. At D1, after 24 h of incubation, the air chamber is punctured to prevent the CAM from attaching to the inner shell membrane. Puncturing the air chamber at D1 improves the quality of the air chamber compared to puncture at later stages (D4). The CAM continues to grow until D12 when it envelops the entire egg content and adheres firmly to the inner shell membrane (Figure 1). The hole made at D7 is enlarged to facilitate sample harvest. D7 is the optimal time for grafting when the CAM reaches maturity and covers the yolk sac and the embryo surface. At this stage, the air chamber hole is enlarged, and the organoids are transferred into the CAM with the help of an automatic pipette.
Brain organoid grafting
Throughout the inherent maturation of brain organoids in vitro, their cellular composition changes from neural progenitors to mature neurons and glial cells. Neurovasculature and neurogenesis are intermingled and interdependent developmental processes in vertebrates, and the signaling leading to this crosstalk is highly dependent on the maturation stage of the neural counterpart29. Neural progenitors are the major source of proangiogenic factors in the developing brain29,30. On the contrary, neurons and astrocytes support neoangiogenesis at mature stages and adulthood through the secretion of vascular endothelial growth factor31,32. These changes in the brain cytoarchitecture, which are reflected in brain organoids too, might lead to differential changes in the integration and response to CAM engrafting, depending on the maturation stage of brain organoids. Therefore, we grafted brain organoids at differentiation day 13 in vitro (DIV), when neuroepithelial cells and radial glia expand; at DIV 37, at the onset of neurogenesis; at DIV 60, when neurogenesis is robust and neurons of different identities are terminally differentiated; and at DIV 120, when a plethora of mature neurons is populating the organoid, generating a well-developed neuronal network and astrogenesis is also widespread.
In all cases, the samples were collected after 5 days of in ovo (DIO) incubation. To avoid biases due to batch effect in brain organoids, all comparisons between grafted and non-grafted organoids were performed in the same batch in a pairwise manner. Organoids were derived from at least two independent experiments per time point and a total of six independent differentiations. Grafting of multiple maturation stages was performed in parallel in each grafting experiment. Upon grafting, organoids continued the maturation process adjacent to the CAM vessels without any evidence of infiltration of blood vessels into the organoid (Figure 2).
Organoid survival upon grafting ranges from 80% in the youngest and drops to ~66% in mature organoids after 5 days post grafting, suggesting that permissivity and organoid grafting plasticity might change over time (Table 1). The organoid viability was determined by the presence of pyknotic nuclei in H&E stainings. The number of pyknotic nuclei was similar between transplanted and the non-transplanted control organoids (Figure 3), suggesting that CAM is permissive to sustain the live organoid upon transplantation. Moreover, no histological or hemogenic release was detected at the CAM, and the embryos remain alive, suggesting that neither the grafting procedure nor the organoid disrupt the integrity of the CAM or significantly affect embryo development.
Cell composition in grafted brain organoids
Unguided brain organoids have a marked inter- and intraorganoid cellular variability. However, they are predominantly neural, generating waves of progenitors, neurons, and astrocytes at a sequential pace2,26. Next, we evaluated whether grafting alters the neural composition in the brain organoids. Younger organoids (13 DIV + 5 DIO: 18 days) show no differences from their control non-grafted organoids in the presence of neurons (TUBB3+). At later maturation stages (60 DIV +5 DIO: 65 days and 120 DIV + 5 DIO: 125 days), the proportion of neurons (TUBB3+) is dramatically decreased compared to that in non-grafted organoids (Figure 4). In non-grafted organoids matured for 60 DIV, neurons are widely spread all around the organoids.
Notably, the number and localization of glial fibrillary acidic protein-positive (GFAP+) astrocytes, evidenced an increase compared to those observed in the non-grafted controls. Moreover, these GFAP+ astrocytes acquired an unexpected localization in the outer side of the organoid. In 18 DIV, free organoids (non-grafted), few astrocytes can be seen and no structural distribution could be detected; on the contrary, they appear homogeneously spread throughout the organoid (Figure 3).
Figure 1: Timeline of CAM assay setup for organoid transplantation. (A) Surgical paper adhesive bandage application on top of the egg. (B) Making a hole in the shell with a needle over the adhesive bandage. (C) Open the eggshell's window, view from the top, ready for the transplantation of a (D) human brain organoid (day 60). Scale bar = 100 µm. (E) Visualizing the engrafted organoid at the CAM after 5 days of in ovo incubation. (F) Hallmarks of brain organoid development from Day 0 to Day 120. The arrow indicates the location of the organoid. Abbreviation: CAM = chorioallantoic membrane. Please click here to view a larger version of this figure.
Figure 2: Hematoxylin and eosin staining of grafted organoids in the chorioallantoic membrane. (A) 13 DIV + 5 DIO organoid; (B) 37 DIV + 5 DIO organoid; (C) 60 DIV + 5 DIO organoid; and (D) 120 DIV + 5 DIO organoid; red squares mark the organoid. Scale bar = 500 µm. Abbreviations: DIV = day in vitro; DIO = day in ovo. Please click here to view a larger version of this figure.
Figure 3: Immunocytochemistry of organoids at different stages of development. GFAP was used for staining and DAPI for the nuclei. Scale bar = 100 µm. (A,B) Images of the same 13 DIV + 5 DIO grafted organoid where astrocytes are seen around the edge of the organoid. (C) Image of 18 DIV free organoid where astrocytes are barely visible. The difference between free and grafted organoids is very noticeable. (D,E) Images of 37 DIV + 5 DIO grafted organoid where few astrocytes can be seen. (F) Image of 42 DIV free organoid. Yellow arrows indicate the location of the astrocytes. Abbreviations: DIV = day in vitro; DIO = day in ovo; DAPI = 4',6-diamidino-2-phenylindole; GFAP = glial fibrillary acidic protein. Please click here to view a larger version of this figure.
Figure 4: Immunocytochemistry of 120 DIV organoids. TUBB3 was used for staining the neurons and DAPI for the nuclei. Scale bar = 100 µm; images were taken at 20x. (A) A 120 DIV +5 DIO grafted organoid where few neurons can be seen. (B) A 125 DIV, free organoid with plenty of neurons. The asterisk indicates the location of the organoid while the dashed line represents the limit between the organoid and the CAM (determined by histological analysis). Abbreviations: DIV = day in vitro; DIO = day in ovo; DAPI = 4',6-diamidino-2-phenylindole; CAM = chorioallantoic membrane. Please click here to view a larger version of this figure.
Organoids | 13 DIV | 37 DIV | 60 DIV | 120 DIV |
Alive | 6 (75%) | 10 (83%) | 6 (66%) | 6 (66%) |
Dead | 2 (25%) | 2 (16%) | 3 (34%) | 3 (34%) |
Table 1: Embryonic survival rates after CAM organoid grafting. Abbreviations: DIV = day in vitro; DIO = day in ovo; CAM = chorioallantoic membrane.
In this study, we describe a detailed protocol with numerous key steps that provide favorable growth and development of human brain organoids upon grafting without perturbing the survival of the chicken embryos. We recommended the use of sterile needles to puncture the air chamber of the egg after 24 h of incubation (day 1). Additionally, we also tried to make the puncture at day 4 (after checking through the eggshell by light to test the development of the vasculature to be sure that we were working only with healthy embryos). However, this resulted in a decline in embryo viability due to the proximity of the CAM vessels to the eggshell. It should be noted that applying too much force can result in breaking the eggshell with the subsequent damage of the CAM vasculature. The use of a light source to detect thinned areas of the eggshell was helpful for making blunt holes while simultaneously preventing overpenetration and passing through the egg, leading to invasion of the CAM.
Even though egg rotation seems mandatory for the proper development of the chicken embryos during incubation, it is necessary to avoid it after puncturing the air chamber. Removal of albumin was not warranted for this model nor was the addition of PBS to avoid dehydration of the egg. Furthermore, the use of 70% ethanol solution should be done wisely, with extra care to dry any remaining ethanol solution on the shell to preserve embryonic viability. Another critical point is the size of the eggshell hole used for the engraftment, which requires balancing the smallest possible hole that allows the grafting manipulation comfort. The selection of the embryonic stage for graftings (E7) was driven by the maturation stage of the CAM and its relative size, favoring the integration of the transplant. However, in the eventual case that the transplanted organoids require longer in ovo maturation, the grafting can be performed at day E5.5/6. Brain organoid variability is controlled to ensure experimental viability. Organoids of the same age exhibit similar sizes, and it is important to note that, regardless of the age of the organoids, they were all incubated for 5 days after transplantation. Furthermore, for consistency, the control organoids were meticulously chosen to match the exact size of their engrafted counterparts. Another technical aspect, relevant to supporting the survival of both the grafted organoids and the embryo, is closing the grafting hole with a square piece of parafilm, to maintain a controlled in ovo environment.
Contrary to our initial hypotheses, young organoids survived better after the grafting process and CAM integration than the older ones. We hypothesized that younger organoids are more plastic due to their enriched presence in early-stage progenitors33, while mature neurons require a permissive neuronal environment for their maintenance. Moreover, cytoarchitectural changes in the early (D18) brain organoid graftings were detected in the distribution of astrocytes. Notably, we detected that astrocytes differentiated at this early stage only when grafted and they distributed themselves to generate a pseudo-capsule underneath the organoid surface. This structure could be a reminiscence of the potential of the organoid to generate a neurovascular interface that could mimic the BBB. Alternatively, the increased reactivity of astrocytes, positioned in a barrier-like structure, could be indicative of an immunogenic response of the organoid to secreted signaling from the CAM. However, further work is needed to demonstrate this hypothesis. In summary, this paper describes the steps to engraft organoids of multiple developmental stages in the chicken CAM, being a more permissive situation for early development organoids, and the cellular effect of these grafting in neurons and GFAP.
The authors have nothing to disclose.
We thank Dr. Alcántara and Dr. Ortega from UB and the rest of the members in Dr. Acosta's lab for the insightful discussions. S.A. is Serra-Hunter fellow assistant professor from the Generalitat de Catalunya at Universitat de Barcelona.
Anti-TUBB3 [Tuj1], mouse | BioLegend | 801201 | 1:1,000 |
Anti-GFAP, rabbit | GeneTex | GTX108711 | 1:500 |
Anti-rabbit AlexaFluor 488, goat. | Invitrogen | A-21206 | 1:1,000 |
Anti-mouse AlexaFluor 594, goat | Jackson ImmunoResearch | 715-585-150 | 1:500 |
Fertilized White Leghorn chicken (Gallus gallus) eggs | Granja Gibert (Cambrils, Spain) | ||
DAPI | Invitrogen | D1306 | 1:10,000 |
DPX | Sigma | 100579 | xylene-based mounting medium |
Gentle Dissociation Solution | CreativeBiolabs | ITS-0622-YT187 | cell dissociation solution |
Matrigel | BD Biosciences | 356234 | |
Mowiol 4-88 mounting media | Merk | 81381 | |
Paper towel, lab-grade | Sigma-Aldrich | Z188956 | |
ROCK inhibitor Y27632 | Millipore | SCM075 | 10 nM |
Sharp-Point Surgical Scissors | VWR | 470106-340 | |
Superfrost Plus Adhesion Microscope Slides | Epredia | J1800AMNZ |