Maternal milk cell components are uptaken by infant liver macrophages via extracellular vesicle mediated transport
These are the figures and data for the manuscript with this title above by Rose Doerfler et a..
Paper Abstract: Milk is a multifaceted biofluid that is essential for infant nutrition and development, yet its cellular and bioactive components, particularly maternal milk cells, remain understudied. Early research on milk cells indicated that they cross the infant’s intestinal barrier and accumulate within systemic organs. However, due to the absence of modern analytical techniques, these studies were limited in scope and mechanistic analysis. To overcome this knowledge gap, we have investigated the transintestinal transport of milk cells and components in pups over a 21-day period. Studies employed a mT/mG foster nursing model in which milk cells express a membrane-bound fluorophore, tdTomato. Using flow cytometry, we tracked the transport of milk cell-derived components across local and systemic tissues, including the intestines, blood, thymus, mesenteric lymph nodes, and liver. These experiments identified milk-derived fluorescent signal in intestinal epithelial and immune cells as well as liver macrophages in 7-day-old pups. However, the minute numbers of macrophages in mouse milk suggest that maternal cells are not systemically accumulating in the infant; instead, pup macrophages are consuming milk cell membrane components, such as apoptotic bodies or extracellular vesicles (EVs). Ex vivo experiments using primary macrophages support this hypothesis, showing that immune cells preferentially consumed EVs over milk cells. Together, these data suggest a more complex interplay between milk cells and the infant’s immune and digestive systems than previously recognized and highlight the need for future research on the role of milk cells on infant health.
Methods:
2.1 Fostering of Mice
All mouse experiments were approved by the Institutional Animal Care and Use Committee at Carnegie Mellon University under protocol no. PROTO201600017 and performed in accordance with all institutional, local, and federal regulations. Mice were housed under controlled temperature (25 °C) in 12-hour light-dark cycles. Animals were given ad libitum access to standard diet and water. mT/mG mice (Jackson Labs) expressing the fluorescent protein tdTomato on all cells and C57BL/6 mice (Charles River) were coordinately mated. Within 24 hours after birth, the C57BL/6 pups were transferred to the mT/mG dam. At time of transfer, the pups were rolled in the bedding of the foster dam’s cage, in order to mask the scent of the pups and reduce the risk of rejection by the foster mother. The rate of rejection was minimal. We weighed the fostered pups and found that they were healthy and grew at the same rate as non-fostered pups (Fig. S1). At time points up to 21 days of age, the age when mouse pups are typically weaned, the pups were euthanized by CO2 asphyxiation followed by decapitation. Blood was collected by cardiac puncture. Organs were collected for analysis by flow cytometry and imaging. Non-fostered C57BL/6 and mT/mG mouse pups were used as negative controls and positive controls, respectively.
2.2 Mouse Milk Collection
To collect milk directly from a lactating mouse, pups were separated from the dam for 2 hours prior to milking. The dam was then anesthetized using isoflurane, and 2 IU oxytocin was injected intraperitoneally to start milk letdown. We had the most success in obtaining mouse milk when two people worked together: one person to manually express milk from the mouse, and another person to collect the milk using a 10 μL pipette. Cells were isolated from the milk as follows: 100 μL of milk was combined with 1 mL phosphate buffered saline (PBS), and the milk was centrifuged for 10 minutes at 500 x g, 4°C. The cell pellet was transferred to a fresh 1.5 mL tube with 1 mL PBS and washed twice by centrifugation for 10 minutes at 500 x g, 4°C, to remove remaining fats and proteins.
2.3 Preparation of Organs for Flow Cytometry
Intestines were collected from mice and placed into ice-cold PBS. Small intestines were flushed with cold PBS using a 25G syringe to wash out intestinal contents. A single suspension of cells was then prepared for flow cytometry according to a modified version of the protocol developed by Couter and Surana.(23) This protocol was written for adult mice, but some modifications are necessary when working with mouse pups. The authors recommend turning intestinal pieces inside out to expose the epithelial layer. The intestines of mouse pups are much smaller and more fragile than those of adult mice, so it is very difficult to invert the intestinal pieces. Instead, we used a scalpel or razor to cut along the length of the intestine to expose the mucosal surface and then cut the intestine into small pieces. The epithelial layer was removed by placing the intestinal pieces in a tube of 30 mL RPMI (Roswell Park Memorial Institute) medium containing 500 μL FBS (VWR), 60 μL of 0.5 M EDTA (Sigma), and 93 μL of a 50 mg/mL dithiothreitol stock solution (VWR). After incubating the tube with shaking at 37°C for 15 minutes, the epithelial cells were collected through a cell strainer, and the remaining intestinal pieces were considered to be the lamina propria. These pieces were washed with RPMI to remove EDTA, and then transferred to tubes containing the digest media. For the enzymatic digest of the lamina propria, the intestinal pieces were placed into a solution containing collagenase and dispase. We used lower concentrations of enzymes than those recommended by the protocol, because mouse pups have less connective tissue than adults do. For 25 mL of media, we used 20 mg of collagenase II (MP Biomedicals) and 10 mg of dispase II (Sigma). Intestinal pieces were digested in the media for 30 minutes at 37°C with shaking, and then vortexed to break up clumps of cells. Following digestion, the cells were strained through a 70 µm cell strainer (VWR). The intestinal cells from the epithelial layer and the lamina propria were pelleted by centrifugation for 10 minutes at 500 x g, 4°C. The pelleted cells were re-suspended in PBS to wash off remaining enzymes and debris, and then centrifuged again at 500 x g. From there, the single-cell suspensions were stained for flow cytometry. Spleens were collected from mice and placed into ice-cold PBS. A single-cell suspension was prepared by smashing the spleen through a 70 μm cell strainer using the plunger of a 3 mL syringe. Mechanical dissociation was used here instead of enzymatic because it is simpler, and studies have shown that it is appropriate for our chosen endpoints(24). Cells were pelleted for 10 minutes at 500 x g, then re-suspended in 5 mL of cold red blood cell lysis buffer (VWR). After 5 minutes of incubation in lysis buffer, cells were re-suspended in flow buffer and washed twice by centrifugation. For the thymus and mesenteric lymph nodes, there is no enzymatic digest necessary and no red blood cell lysis step. Again, mechanical dissociation was chosen for the same reasons as above (25, 26). The most important consideration is the dissection step. Mesenteric lymph nodes, found along the colon, were removed with forceps. Any mesenteric fat was carefully removed, because it can cause additional noise in the sample and reduce cell viability. These organs were smashed through a 70 μm cell strainer using the plunger of a 3 mL syringe, centrifuged for 5 minutes at 500 x g, then re-suspended in flow buffer and washed once more by centrifugation to prepare a single-cell suspension. Livers were digested using a gentleMACS dissociator and the mouse liver digestion kit (Miltenyi Biotech, Auburn, CA), as per the manufacturer's instructions. Blood was collected from mice by cardiac puncture. The blood was collected into 2 mL EDTA-coated tubes and placed on ice. Blood was combined with 2 mL red blood cell lysis buffer and incubated at room temperature for 5 minutes. Then, the cell pellet was quenched with 10 mL PBS and washed twice by centrifugation at 500 x g to remove lysed red blood cells and other debris. In mouse pups that have not yet begun to eat solid food, the milk in the stomach forms a soft curd. To collect milk from the stomachs of pups after euthanasia, a small incision was made in the stomach, and the milk curd was removed with forceps. The milk curd was smashed through a 70 μm cell strainer using the plunger of a 3 mL syringe, centrifuged for 10 minutes at 500 x g, 4°C, re-suspended in flow buffer, and washed twice by centrifugation.
2.4 Fostered Mice Flow Cytometry
Cells were stained with antibodies at a 1:100 dilution on ice for 30 minutes. All flow cytometry was conducted using an ACEA Novocyte 3000 cytometer. Data analysis was conducted in NovoExpress software. Analyzing intestinal cell populations by flow cytometry, especially cells from mouse pups, presents unique challenges. Dramatic physiological changes occur in the mouse’s digestive system during the first few weeks of life. At 7 days old, a mouse pup’s diet consists entirely of milk, but by the time of weaning at 21 days, the pup eats a diet of solid food, and the intestines more closely resemble adult intestines (Fig. S2A). These changes in diet and physiology affect the background fluorescence in flow cytometry studies (Fig. S2B). Therefore, it is necessary to use non-fostered mice of the same ages as the fostered mice for control groups. Next, it is important to draw flow cytometry gates carefully. Intestines are complex and contain debris and dead cells, which are autofluorescent and easily mistaken for fluorescent milk cells. Therefore, we used a gating strategy that excludes autofluorescent cells and debris by plotting tdTomato against another unused fluorophore (Fig. S3). Representative cell-specific gating strategies for the lamina propria, mesenteric lymph nodes, blood, spleen, and thymus are shown in Figs. S4-S8. When quantifying a rare cell population by flow cytometry, noise from background fluorescence and nonspecific antibody staining present complications. Therefore, careful controls and gating strategies are essential. Organs have varying levels of background fluorescence, so, for each organ, non-fostered C57BL/6 mice were used as fluorescence minus one (FMO) controls, and mT/mG mice were used as positive controls. Gates were drawn to include all the cells in the positive control mouse organs and exclude the cells in the FMO controls.
2.5 Imaging
Intestinal pieces were flushed with PBS, and then flash frozen in liquid nitrogen. The frozen tissue was embedded in OCT and then cryosectioned in 7 µm slices. The slides were then washed with PBS to remove OCT, and tissue was stained with 1:500 Hoechst 33342 and 1:50 AF647-phalloidin (Cell Signaling Technologies) for 1 hour. Slides were imaged on a Zeiss 880 laser scanning confocal microscope. Non-fostered C57BL/6 mice were used as negative controls, and mT/mG mice were used as positive controls. Image analysis was conducted using ZEN software and ImageJ.
2.6 Extracellular Vesicle Isolation and Characterization
Extracellular vesicles from milk were isolated by a combination of ultracentrifugation and size exclusion chromatography.(27–29) First, mouse milk was centrifuged at 2000×g for 10 min at 4°C to separate off the fat layer, which was then removed, yielding skim milk. Briefly, 50 μL of fat separated mouse this skim milk was then centrifuged at 2,000 ×g for 10 min at 4°C and then at 10,000 ×g for 30 min at 4°C. The supernatant was then ultracentrifuged at 100,000 ×g for 3 hours (TL-100 benchtop ultracentrifuge, Beckman-Coulter). The obtained crude EV pellet was washed in PBS once at 100,000×g for 3 hours. The washed pellet was resuspended in 1 ml of PBS and EVs were purified by mini-size exclusion chromatography (mini-SEC) using 1.5 cm x 12 cm mini-columns (Bio-Rad, Hercules, CA, USA; Econo-Pac columns) packed with 10 mL of Sepharose 2B (Millipore-Sigma, St. Louis, MO). Crude EVs (1.0 ml) obtained from ultracentrifugation were loaded onto the column and five 1 ml fractions corresponding to the void volume peak were collected in PBS(27) . Fraction four was collected and used for subsequent experiments as the ‘EV’ fraction. The fourth fraction was confirmed to contain milk EVs using three independent techniquse as per MISEV2018 guidelines.(30) : (1) transmission electron microscopy (TEM) to observe classic EV-like structures, (2) Western blotting to detect EV-enriched markers (TSG101 and CD9), and (3) nanoparticle tracking analysis (NTA) to determine a size distribution profile with an average size of approximately 100 nm.Given that SEC is a size-dependent assay, we anticipate that the EVs obtained using this approach contain a heterogeneous mixture of exomeres, EVs, and microvesicles in the size range of 30-200 nm.(31) The EVs were characterized by nanoparticle tracking analysis (NTA) and Western blotting as per MISEV2018 guidelines.(30) The concentration and size distribution of EVs were measured by NTA using NanoSight 300 (Malvern, UK). First, the vesicles were diluted in ddH2O, and then the video image was captured at the camera level of 14. The captured videos were analyzed using NTA software, maintaining the screen gain and the detection threshold at 1 and 5, respectively. To determine mean particle size/concentration in each sample, three consecutive measurements were obtained and averaged. Given that SEC is a size-dependent assay, we anticipate that the EVs obtained using this approach contain a heterogeneous mixture of exomeres, EVs, and microvesicles in the size range of 30-200 nm.(31)
2.7 On-bead Extracellular Vesicle Flow Cytometry
EV membrane stability in presence of simulated gastric fluid (SGF) was assessed using on-bead flow cytometry as previously described.(32, 33) Calcein Deep Red (AAT Bioquest Inc., Sunnyvale, CA) was solubilized in DMSO and diluted with 1× PBS to a final concentration of 10 μM. EVs were isolated and labeled incubated with Calcein Deep Red prior to treatment with simulated gastric fluid (SGF) for 30 minutes. The calcein reagent becomes fluorescent only within intact EVs. (32) The SGF formulation used was modified to reflect infant stomach conditions.(34–36) The buffer consisted of 34 mM sodium chloride and 19 mM potassium chloride, adjusted to pH 4 with hydrochloric acid, and then spiked with 80 U/mL porcine pepsin (Sigma). Non-treated EVs and sonicated EVs were used as controls. CD63-conjugated magnetic beads (ExoCap, MBL International, Woburn, MA) were prepared as previously described.(37) Briefly, monoclonal anti-CD63 antibody (MA5–24169, Invitrogen, Carlsbad, CA) was biotinylated using a one-step antibody biotinylation kit purchased from Miltenyi Biotec (Auburn, CA) as recommended by the manufacturer. Biotinylated CD63 antibody (5 μg) was incubated with thoroughly washed 0.5 ml of streptavidin coated magnetic beads (1 × 108 beads/ml) for 1 h at 23 °C under constant agitation. 10 μg of EVs (controls/SGF treated for 30 min at 37°C) were each incubated with 100 μl of CD63-conjugated magnetic beads for 18 h (overnight) at 4 °C under constant agitation. To assess membrane integrity, EVs were captured on CD63 magnetic beads, washed three times with PBS and 100,000 events/group were assessed using NovoCyte 3000 flow cytometer (Agilent Technologies, Santa Clara, CA). Data was analyzed using NovoExpress software (Agilent Technologies, Santa Clara, CA).
2.8 Bone Marrow Derived Macrophages
Bone marrow derived macrophages were isolated and cultured as previously described.(38) Briefly, C57BL/6 mice of at least 8 weeks of age were euthanized via CO2 asphyxiation followed by cervical dislocation. The hindlimbs were disinfected with 70% isopropanol. The hindlimb skin was excised to expose the muscle. Tibiae and femurs were removed and cleaned of muscle. The bones were washed in media and then both ends of each bone were removed using sterile scissors. Media was flushed through each bone to remove the cells. Cells were centrifuged and then resuspended in fresh media then strained using a 70 μm filter. Cells were counted and then seeded in 96 well plates at 200,000 cells per well. Bone marrow cells were cultured in DMEM/10% FBS/10% L929 conditioned media/1% Pen/Strep/2% NEAA/1% HEPES for 7 days, changing the media every 2 days. To polarize macrophages to M1 (pro-inflammatory) and M2 (anti-inflammatory), macrophages were treated with 20 ng/mL IFN-γ and 100 ng/mL LPS for M1 or 20 ng/mL IL-4 M2 for 24 hours. Following polarization, the media was replaced by normal media. Macrophages were treated with whole milk, milk cells or milk EVs (all derived from mT/mG mice) for 24 hours.
2.9 Macrophage flow cytometry
To characterize the uptake of milk cells or EVs, as well as the effect on macrophage polarization, macrophages were analyzed via flow cytometry following treatment. Macrophages were washed with PBS, then removed from the plates using Accutase. Following detachment, macrophages were transferred to round bottom plates and centrifuged. Cells were fixed using fixation buffer (R&D Biosystems) for 10 minutes then quenched in flow buffer (5% FBS/1X PBS). Cells were Fc blocked using 1:200 anti-mouse Fc Block (Biolegend) for 10 minutes. Macrophages were then incubated in antibody solution (1:100 Pacific Blue anti-arginase (ThermoFisher), 1:100 APC anti-iNOS (ThermoFisher) in Permeabilization/Wash buffer (R&D Biosystems) for 45 minutes. Cells were washed in flow buffer and then run on a NovoCyte flow cytometer.
2.10 Statistics and Data Analysis
All statistical analysis was conducted using Graphpad Prism 8 software. A significant difference was defined as p < 0.05. All error bars on graphs show mean and standard deviation.
