Anti-Inflammatory Effects of Encapsulated Human Mesenchymal Stromal/Stem Cells and a Method to Scale-Up Cell Encapsulation

Abstract

Mesenchymal stem/stromal cells (MSC) promote recovery in a wide range of animal models of injury and disease. They can act in vivo by differentiating and integrating into tissues, secreting factors that promote cell growth and control inflammation, and interacting directly with host effector cells. We focus here on MSC secreted factors by encapsulating the cells in alginate microspheres, which restrict cells from migrating out while allowing diffusion of factors including cytokines across the capsules. One week after intrathecal lumbar injection of human bone marrow MSC encapsulated in alginate (eMSC), rat IL-10 expression was upregulated in distant rat spinal cord injury sites. Detection of human IL-10 protein in rostrally derived cerebrospinal fluid (CSF) indicated distribution of this human MSC-secreted cytokine throughout rat spinal cord CSF. Intraperitoneal (IP) injection of eMSC in a rat model for endotoxemia reduced serum levels of inflammatory cytokines within 5 h. Detection of human IL-6 in sera after injection of human eMSC indicates rapid systemic distribution of this human MSC-secreted cytokine. Despite proof of concept for eMSC in various disorders using animal models, translation of encapsulation technology has not been feasible primarily because methods for scale-up are not available. To scale-up production of eMSC, we developed a rapid, semi-continuous, capsule collection system coupled to an electrosprayer. This system can produce doses of encapsulated cells sufficient for use in clinical translation.

Keywords: alginate; encapsulation; human IL10; mesenchymal stem cells; spinal cord injury.

Figure 1

eMSC viability and biological activity in vitro and ex vivo. (A) In vitro eMSC live (green)/dead (red) staining same day (day 0) after preparation of eMSC; (B) Survival of eMSC in vitro (C) Survival of eMSC ex vivo after recovery from cauda equina of SCI rats; (cell number at day 0 = 86.0 ± 9.79, n = 6). On day 0, cells in eMSC were stained (number of cells/capsule = 56.6 ± 7.44, n = 5) after washing them from the needle hub post-injection. eMSC were incubated in vitro for 0, 7, and 42 days as noted, and eMSC from the same batches were injected in SCI rats and retrieved 7 and 42 days later (ex vivo), (B). The % of live cells was determined from confocal micrographs. One-way ANOVA followed by Tukey’s test was used to analyze the data (mean ± SEM). * p < 0.05 and *** p < 0.001. (D) PGE2 expression from eMSC in vitro and ex vivo after recovery from a 42-day incubation in vivo in the cauda equina as in (C). The eMSC recovered at 42 days were incubated for 24 h without LPS (−LPS) and after removing the conditioned medium it was replaced with fresh medium containing 1 µg/mL LPS (+LPS) for an additional 24 h. Student t-test was used for statistical analysis. ** p < 0.01.

Figure 2

Measurement of human IL-10 in CSF 7 days after injection of eMSC into the cauda equina of SCI rats. Levels of human IL-10 were measured by ELISA in CSF collected from the foramen magnum at the base of the brain of SCI rats injected with capsules, without (empty) or with eMSC (n = 2/group). Student t-test was used for statistical analysis. * p < 0.05.

Figure 3

Effect of human eMSC in rat endotoxemia. (A) Expression of rat TNF-α in sera of rats injected with LPS and measured by ELISA after injection of saline alone or capsules without (empty) or with eMSC (n = 3/group). (B) Multiplex assay of sera from cardiac puncture showed significant reductions in rat IL-1β, IL-6, IFN-γ, and TNF-α. (C) Human IL-6 levels in the rat were detected at significantly higher levels than empty capsules control. n = 5 for Empty and n = 3 for eMSC. One-Way ANOVA or student t-test was done to test significance between the groups. * p < 0.05 and ** p < 0.01.

Figure 4

Schematic of Encapsulator and RaCCS. (A) An air pump is used with constant pressure to pressurize the vessel containing the alginate cell suspension, which provides force to drive it towards the needle. Spherical droplets fall in the electric field into the collecting dish where they are crosslinked. (B) The RaCCS consists of a peristaltic pump that drives crosslinking buffer into the dish through an inlet tube and the capsules are collected through an outlet tube onto a filter (yellow) and washed.

Figure 5

Capsule properties using constant force with a syringe pump. (A) Different needles with varying outer radii were used and diameters of the resulting capsules were measured. Capsule diameter is a linear function of needle outer radius; f(x) = 2092x − 43, R-squared = 0.9965; (B) Number of capsules produced is modeled as a function of needle outer radius, f(x) = (0.2021x) − 3, R-squared = 0.9918.

Figure 6

Flow rates increase exponentially as a function of needle inner radius. The viscosity of 2.25% (w/v) alginate was measured to be 330 mPas and used to calculate the predicted curves. The flow rate is proportional to the 4th power of needle inner radius in agreement with the Hagen Poiseuille Equation (Appendix A).