This article describes a protocol for aggregating and encapsulating spleen cells within a semi-solid basement membrane matrix. Basement membrane matrix constructs can be used in three-dimensional culture for studying organoid development, or for in vivo transplantation and tissue regeneration studies.
The spleen is an immune organ that plays a key role in blood-borne immune responses. The anatomical or functional loss of this tissue increases susceptibility to severe blood infections and sepsis. Auto-transplantation of spleen slices has been used clinically to replace lost tissue and restore immune function. However, the mechanism driving robust and immunologically functional spleen tissue regeneration has not been fully elucidated. Here, we aim to develop a method for aggregating and encapsulating spleen cells within a semi-solid matrix in order to investigate the cellular requirements for spleen tissue formation. Basement membrane matrix encapsulated cell constructs are amenable to both in vitro tissue culture of three-dimensional organoids as well as transplantation under the kidney capsule to directly assess in vivo tissue formation. By manipulating the input cells for aggregation and encapsulation, we demonstrate that graft-derived PDGFRβ+MAdCAM-1– neonatal stromal cells are required for spleen tissue regeneration under animal transplantation models.
Traumatic rupture of the spleen and the appearance of multiple splenic nodules in the body was one of the first indications that spleen tissue harbored regenerative capacity1,2. Spleen auto-transplantations were later introduced into the clinic to preserve spleen tissue in patients requiring emergency splenectomy3. Yet, despite being a part of clinical practice for decades, very little is known about how the spleen regenerates. Animal transplantation models have provided insight into multiple parameters of spleen regeneration and immune function4,5. In particular, experimental modifications to the graft preparation method have allowed tissue regeneration to be studied in greater detail at the cellular and molecular level.
Transplantations involving whole spleen slices undergo a phase of mass necrosis before a new spleen structure is rebuilt6. The initial phase of graft necrosis suggests that the bulk of transplanted tissue largely consists of red and white blood cells and is unnecessary for spleen regeneration. This was investigated experimentally by excluding hematopoietic cells from spleen grafts before transplant under the mouse kidney capsule. Here, the non-leukocyte/non-erythrocyte fraction of the spleen, which includes stromal and endothelial cells, was shown to be sufficient to induce de novo tissue formation7. Spleen stromal tissue could be further processed into a single-cell suspension, enabling the use of cell sorting technologies to manipulate cellular graft composition. By selectively removing candidate cell types, two CD45–TER-119– stromal cell populations were identified that were indispensable for graft development: an endothelial-like CD31+CD105+MAdCAM-1+ cell population and a more broadly defined PDGFRβ+ mesenchymal cell population8.
The construction of grafts from spleen cells varies in terms of support materials and cell-loading processes. Tissue-engineered spleens have previously been prepared by loading splenic units onto a polyglycolic acid/poly-L lactic acid polymer scaffold5,9. Interestingly, spleen stromal cells absorbed into a collagen sponge failed to engraft, whereas stromal cells aggregated and loaded over a collagen sheet facilitated spleen regeneration8. The resuspension of spleen stromal cells inside a Matrigel matrix has also been demonstrated to induce cell aggregation under three-dimensional culture conditions10. However, this method has not been tested for use in transplantation models. The overall goal of the current protocol is to forcibly aggregate and encapsulate spleen stromal cells directly within the basement membrane matrix, which subsequently can be transferred to a three-dimensional in vitro tissue culture system or used as a vehicle for animal model transplantations (Supplementary Figure 1).
All animal procedures were conducted according to experimental protocols approved by the University of Queensland Animal Ethics Committee (UQBR/079/19).
1. Tissue collection and stromal cell preparation
2. Matrix encapsulation of cell aggregates
3. Three-dimensional organoid culture
4. Kidney capsule transplantation
Cell aggregation is important for promoting cell-to-cell contact and signaling. Encasing cell aggregates inside the basement membrane matrix supported both 3-dimensional cultures for in vitro tissue organoid formation and facilitated the mechanical delivery of cells into the kidney capsule for graft transplantation. To establish these constructs, the basement membrane matrix was first maintained in a fluidic state under ice-cold conditions. Cell aggregation was subsequently achieved by layering a concentrated cell suspension above and using centrifugal force to push cells through the high-density matrix. Optimization of the centrifugation speed (200-2000 x g) was required to achieve a mid-layer cell positioning (Figure 1A), which was also dependent on the cell number. Higher centrifugal forces (>500 x g) propelled cells to the very tip of the pipette (Figure 1B), and caution is required for smaller cell numbers (i.e., <5 x 105) that may be lost during the removal of the flexible laboratory film covering. Conversely, cells may not travel adequately through the basement membrane matrix if G-forces are too low (≤200 x g) (Figure 1C). Centrifugation speed for cell numbers <1 x 105 or >2.5 x 106 may need further optimization. Following centrifugation, cells were set inside the basement membrane matrix by warming the construct at 37 °C for 15 min. This solidification process enabled the matrix "plug" to be ejected entirely (Figure 1D) using a thin wire plunger.
In vitro spleen organoid formation
Basement membrane matrix plugs ejected into an appropriate tissue culture vessel reproducibly formed 3-dimensional organoid-like structures which could be maintained following standard tissue culture techniques and conditions (37 °C, 5% CO2, 95% humidity)11. In culture, the supporting basement membrane matrix substrate gradually dissipated between days 0 and 14 (Figure 2A, i-iii) and could no longer be observed by day 21, leaving an intact organoid-like spherical cell mass (Figure 2A, vi) measuring approximately 545 µm (s.d. = 120 µm, n = 6)11 in diameter. These organoid structures were supported in culture beyond 30 days but did not increase in size over time (Figure 2B). Spleen organoids were composed of CD45–TER-119– stromal, CD45–TER-119+ erythrocyte, and CD45+TER-119– lymphoid cells (Figure 2C), however, the frequency of each population was variable across individual organoids (n = 4; Table 1). To assess the general structure of spleen organoids, 50 µm thick tissue cryosections were prepared and visualized. Organoids were comprised of a non-hollow structure, with cells present across the entire diameter of the tissue (Figure 2D). Assessment of organoid sections at 7 µm single cell layer thickness revealed that cells were arranged in cord-like structures without a clear spatial orientation (Figure 2D), with areas void of nuclei between strings of cells. CD45 antibody staining verified the presence of nucleated hematopoietic cells which densely surrounded peripheral regions of the organoid (Figure 2E). In addition, CD45+ cells of distinct, spindle-shaped morphology were observed in more central organoid regions. A general absence of CD90.2 (T cell) and CD19 (B cell) antibody staining, but positive CD11b staining, demonstrated the specific presence of myeloid cells (Figure 2F). Non-hematopoietic CD105+ and CD31+ endothelial cells were also detected in multiple organoids11.
In vivo spleen graft transplantation
The encapsulation of aggregated spleen stromal cells within a semi-solid matrix facilitates animal transplantation studies. This technique was used to embed unfractionated or CD45–TER-119– FACS sorted neonatal spleen stromal cell preparations inside a basement membrane matrix plug, serving as a vehicle for transplantation under the mouse kidney capsule. In line with similar cell-aggregation protocols8, basement membrane matrix encapsulated cell constructs (MECCs) successfully regenerated spleen tissue in 4/5 independent animal transplantations, thereby confirming the viability of this graft construction technique. To test the utility of MECCs in defining cell types required for spleen regeneration, MECC grafts were constructed from neonatal spleen cells which were specifically depleted of PDGFRβ+MAdCAM-1+ or PDGFRβ+MAdCAM-1– stromal (CD45–) cells11. Grafts lacking PDGFRβ+MAdCAM-1+ cells retained the capacity for gross tissue regeneration (4/4 grafts; Figure 3A)11. Three out of four regenerated tissues exhibited normal spleen cell composition and structure, displaying central arterioles, white pulp follicles, segregated T and B cell compartments, follicular dendritic cells, marginal zone reticular cells, and marginal metallophilic macrophages, red pulp sinusoids, myeloid cells and macrophages (Figure 3B)11. In contrast, grafts that lacked PDGFRβ+MAdCAM-1– cells largely failed to regenerate spleen tissue (1/4 grafts)11. These data support the importance of graft-derived PDGFRβ+ cells in spleen tissue regeneration8, and pinpoint a specific requirement for PDGFRβ+MAdCAM-1– stromal cells.
Figure 1. Basement membrane matrix encapsulation of aggregated spleen stromal cells. A 200 µL pipette tip is used to facilitate cell aggregation inside a basement membrane matrix. A fluidic matrix layer is first established by aspirating 2 µL of ice-cold basement membrane matrix into a pre-chilled pipette tip. A cell suspension is deposited above the matrix layer, and centrifugal force is applied to mobilize and aggregate cells within the basement membrane matrix. (A) Settings for centrifuge speed are optimized to position cell aggregates midway along the matrix layer. (B) Excessive centrifugal speed results in cells that aggregate at the end of the pipette. (C) Insufficient speed prohibits cell movement through the matrix. (D) A semi-solid matrix plug is ejected from the pipette tip following incubation at 37 °C for 15 minutes. Arrowheads indicate the placement of cell aggregates within the matrix. Scale Bar, 100 µm. Please click here to view a larger version of this figure.
Figure 2. In vitro culture of basement membrane matrix-encapsulated spleen aggregates and the formation of 3-dimensional organoid structures. (A) Spleen organoid development over a 35-day time course. Panels (i-vi) correspond to Days 0, 7, 14, 21, 28, and 35 in culture. Images were captured on a Nikon TS2 Inverted Phase Contrast Microscope. Scale bar, 500 µm. (B) Spleen organoid diameter over extended culture periods. Each line represents an individual organoid. (C) Flow cytometry characterization of stromal (CD45–TER-119–), lymphoid (CD45+TER-119–) and erythroid (CD45–TER-119+) cell populations after 53 days in culture. Upper panel: Control tissue prepared from neonatal spleen stroma. Lower panel: Spleen organoid, Day 52. (D) Organoid tissue gross morphology at 50 µm (upper panels) and 7 µm (lower panels) section thicknesses after 22 days in culture. (E) Localization and morphology of CD45+ hematopoietic cells after 34 days in culture. Magnified region is shown in right panels. Sections are 7 µm. (F) Assessment of myeloid (CD11b), T (CD90.2) and B (CD19) cells after 34 days in culture. Magnified region is shown in inset. Sections are 7 µm. Images were captured using a Nikon Eclipse Ti2-E Live Cell Microscope. Scale bars (D, E, F), 100 µm. Please click here to view a larger version of this figure.
Figure 3. Basement membrane matrix graft tissue analysis. Grafts were constructed from neonatal spleen stroma depleted of PDGFRβ+MAdCAM-1+ cells (n = 4). Additional data available online11. (A) Macroscopic appearance of a regenerated spleen graft at 4 weeks post-transplantation under the kidney capsule. The boxed yellow area indicates regenerated spleen tissue. The image was captured using a Leica M60 Stereomicroscope equipped with a Snap zoom adapter and an Apple iPhone 6S. Scale bar, 1000 µm. (B) Composite multi-color immunofluorescence images of 30 µm-thickness graft tissues cryosectioned on the coronal plane. Antibody staining was performed with indicated markers to visualize spleen tissue micro-architecture. Images were captured using a Nikon Eclipse Ti2-E Live Cell Microscope. Boxed areas show regions of higher magnification. Scale bar, 1000 µm. Please click here to view a larger version of this figure.
Organoid No. | Stromal cells | Erythrocytes | Lymphocytes |
1 | 62 | 7.7 | 31 |
2 | 22 | 24 | 55 |
3 | 3.6 | 0.1 | 96 |
4 | 10 | 76 | 12 |
Table 1. Percentage of stromal, erythroid and lymphoid cell populations amongst individual spleen organoids.
Supplementary figure 1. Schematic overview of the protocol highlighting the major steps in generating matrix embedded cell constructs for in vitro and in vivo studies. Please click here to download this File.
The aggregation of neonatal spleen cells inside a semi-solid medium represents a viable method for generating spleen constructs. Similar basement membrane matrix-based protocols have been used to initiate three-dimensional spleen cultures10. Here, we demonstrate that spleen constructs are equally amenable to in vitro organoid culture systems as well as to in vivo transplantation models. Of note, the transplantation of in vitro cultured spleen organoids has not yet been tested but it would be of interest to determine the capacity for full or partial spleen tissue regeneration. This protocol also inherits cell aggregation techniques described in previous spleen transplantation studies8. These studies involved loading cell aggregates over a collagen sheet followed by transplantation under the kidney capsule. Encasement of cell aggregates inside the basement membrane matrix offers a distinct advantage in terms of graft orientation, where variation associated with collagen sheet-side down versus cell aggregate-side down transplantation is eliminated.
The methodology for preparing basement membrane matrix constructs is relatively straightforward. The use of high-concentration basement membrane matrix ensures that cells are initially supported in an aggregated state, before the matrix dissipates over several days. In our experience, there is minimal mixing and dilution of the basement membrane matrix following careful layering (Step 2.6) and centrifugation (Step 2.8) of a PBS-suspended cell solution. Most optimizations center around positioning the cell pellet within the matrix plug (Figure 1). This can be addressed by testing several variables such as centrifugation speed, time, basement membrane matrix volume composition, PBS-suspended cell solution volume, and pipette tip size. Cell number and characteristics (size, density) may also influence settings, and each cell preparation should be optimized individually.
Several limitations are important to highlight. Basement membrane matrix can be a temperature-sensitive product, and care should be taken to work fast to keep reagents cold. This prevents premature matrix solidification and ensures proper encapsulation of the cells. While all relevant materials are stored, prepared, or maintained in a cold environment, manual handling steps (e.g., ejecting the pipette tip, sealing with flexible laboratory film [Step 2.5]) may variably introduce external heat from the user, changing the matrix viscosity and influencing cell migration characteristics during centrifugation. Reproducibility within and between experiments may, therefore, be difficult to achieve consistently.
The second limitation is the requirement for user experience and proficiency. Certain manual techniques may require extended training to perform efficiently and successfully. For example, sealing the pipette tip with laboratory film (Step 2.5) may initially result in low success rates, but this will inevitably improve with practice. Competence with animal micro-surgical techniques is also obtained through repetition. For transplantations, care must be taken, especially when inserting the pipette tip underneath the kidney capsule (Step 4.2.10), inserting the wire plunger to eject the plug (Step 4.2.11), and withdrawing the tip from the kidney. It is important that the plug is ejected gently and steadily whilst the pipette tip is simultaneously withdrawn. Excess movement may result in the kidney capsule becoming perforated or the parenchyma becoming damaged, leading to excessive bleeding.
In summary, a process for encapsulating cells within a semi-solid matrix has been described here with a focus on understanding spleen tissue formation. This system could equally be applied to a wide range of cell types with the potential for multiple in vitro or in vivo downstream applications.
The authors have nothing to disclose.
This research was supported by the National Health and Medical Research Council of Australia (#GNT1078247).
96 Well Polypropylene 1.2 mL Cluster Tubes | Corning | CLS4401 | For placing inside a 14 ml conical tube |
B-mercaptoethanol | Gibco | 21985023 | Stock 55 mM, use at 50 uM |
Collagenase D | Roche | 11088858001 | |
Collagenase IV | Sigma-Aldrich | C5138 | From Clostridium histolyticum |
Deoxyribonuclease I (DNase I) | Sigma-Aldrich | D4513 | Deoxyribonuclease I from bovine pancreas,Type II-S, lyophilized powder, Protein ≥80 %, ≥2,000 units/mg protein |
Dulbecco’s Modified Eagle’s Medium (DMEM) | Sigma-Aldrich | D5796 | 4500 mg/L Glucose, L-Glutamine, and Sodium Bicarbonate, without Sodium Pyruvate, Liquid. Sterile Filtered. |
Dulbecco's Phosphate Buffered Saline (PBS) | Sigma-Aldrich | D8537 | Without calcium chloride and magnesium chloride, sterile-filtered |
Eclipse 200 μl Pipette Tips | Labcon | 1030-260-000 | Bevel Point |
Fetal Bovine Serum (FBS) | Gibco | 26140-079 | Lot# 1382243 |
GlutaMAX | Gibco | 35050061 | Stock 100X, use at 1X |
Matrigel | Corning | 354263 | Matrigel matrix basement membrane High Concentration, Lot# 7330186 |
MEM Non-essential Amino Acids | Gibco | 11140076 | Stock 100X, use at 1X |
Penicillin/Streptomycin | Gibco | 15140122 | Stock 10,000 units/ml Penicillin, 10,000 ug/ml Streptomycin |
Reversible Cell Strainer | STEMCELL Technologies | 27216 | 70 μm |
Ring Tweezers | NAPOX | A-26 | Ring size: 3 mm |
Rock Inhibitor (Y-27632) | MedChemExpres | HY-10071 | |
Thermofisher Heraeus Megafuge 40R Centrifuge | Thermofisher | Acceleration and deceleration speeds were set to 8 | |
Ultra Fine Tweezers | EMS | 78340-51S | Style 51S. Antimagnetic/anti-acid SA low carbon austenitic steel tweezers are corrosion resistant. Anti-glare satin finish. |
Vicryl 5/0 Suture Ligapak Reel | Ethicon | J283G | |
Wiretrol II Long Wire Plunger | Drummond | 5-000-2002-L | Stainless Steel Plunger, 25 & 50 μL/WRTL II, Long |
Wound Clip Applier | MikRon | 427630 | |
Wound Clips | MikRon | 427631 | 9 mm |