Improving embryonic stem cell expansion through the combination of perfusion and Bioprocess model design

Abstract

Background: High proliferative and differentiation capacity renders embryonic stem cells (ESCs) a promising cell source for tissue engineering and cell-based therapies. Harnessing their potential, however, requires well-designed, efficient and reproducible expansion and differentiation protocols as well as avoiding hazardous by-products, such as teratoma formation. Traditional, standard culture methodologies are fragmented and limited in their fed-batch feeding strategies that afford a sub-optimal environment for cellular metabolism. Herein, we investigate the impact of metabolic stress as a result of inefficient feeding utilizing a novel perfusion bioreactor and a mathematical model to achieve bioprocess improvement.

Methodology/principal findings: To characterize nutritional requirements, the expansion of undifferentiated murine ESCs (mESCs) encapsulated in hydrogels was performed in batch and perfusion cultures using bioreactors. Despite sufficient nutrient and growth factor provision, the accumulation of inhibitory metabolites resulted in the unscheduled differentiation of mESCs and a decline in their cell numbers in the batch cultures. In contrast, perfusion cultures maintained metabolite concentration below toxic levels, resulting in the robust expansion (>16-fold) of high quality 'naïve' mESCs within 4 days. A multi-scale mathematical model describing population segregated growth kinetics, metabolism and the expression of selected pluripotency ('stemness') genes was implemented to maximize information from available experimental data. A global sensitivity analysis (GSA) was employed that identified significant (6/29) model parameters and enabled model validation. Predicting the preferential propagation of undifferentiated ESCs in perfusion culture conditions demonstrates synchrony between theory and experiment.

Conclusions/significance: The limitations of batch culture highlight the importance of cellular metabolism in maintaining pluripotency, which necessitates the design of suitable ESC bioprocesses. We propose a novel investigational framework that integrates a novel perfusion culture platform (controlled metabolic conditions) with mathematical modeling (information maximization) to enhance ESC bioprocess productivity and facilitate bioprocess optimization.

Figure 1. Schematic of the experimental design.

mESCs were encapsulated in alginate hydrogels, as described previously , , and cultured within a batch operated HARV bioreactor and a custom-built perfusion bioreactor.

Figure 2. Batch culture growth kinetics, viability and metabolism.

a) Total number of viable ESCs; experimental (◊) and model predictions (–); b) Simulation results of the different ‘naïve’ (X_U) and ‘primed’ (X_D) mESC sub-populations; c) Micrographs of the mESCs encapsulated in the alginate hydrogels at day 9 (i, ii); Live (green)/dead (red) fluorescence micrographs showing live cells (iii) forming colonies of <200 µm in diameter as well as a high proportion of dead cells (iv) at day 9; d) experimental (symbols) and simulation (lines) glucose and lactate concentration profiles; e) experimental (symbols) and simulation (lines) glutamine and ammonia concentration profiles. Experimental values represent mean±SD, N = 3.

Figure 3. LIF concentration and associated gene expression in batch cultures.

a) LIF growth concentration levels over the 6 day culture period remain significantly higher than the half maximal and differentiation threshold levels. b) Gene expression levels of LIF-Stat3 signalling (Socs3, Stat3, Klf4) and BMP-ID signalling (Sox1, Id1 and Id3) for control (day 0) and day 6 batch culture values. *P<0.05, Student t-test. Experimental values represent mean±SD, N = 3.

Figure 4. Pluripotency-related gene expression in batch cultures.

Gene expression levels of a) Rex1, b) Oct3/4, c) Fgf5, d) Sox2, e) Dppa3 and f) Nanog. Model simulation results are predicted for Rex1, Fgf5 and Dppa3 (line). *P<0.05, one-way ANOVA. Experimental values represent mean±SD, N = 3.

Figure 5. Perfusion culture growth kinetics, viability and metabolism.

a) Total number of viable ESCs; experimental (□) and model predictions (–); b) Simulation results the cellular states of ‘naïve’ (X_U) and ‘primed’ (X_D) mESCs; c) Micrographs of the mESCs encapsulated in the alginate hydrogels at day 9 (i, ii); Live (green)/dead (red) fluorescence micrographs showing the live cells (iii) forming large colonies of 300–400 µm in diameter with minimal dead cells (iv) at day 9; d) experimental (symbols) and simulation (lines) glucose and lactate concentration profiles; e) experimental (symbols) and simulation (lines) glutamine and ammonia concentration profiles. Experimental values represent mean±SD, N = 3.

Figure 6. LIF concentration and associated gene expression in perfusion cultures.

a) LIF growth concentration levels remain high in both the perfusion and batch cultures throughout the culture period. No difference in LIF concentration between the two operation modes was observed. b) Gene expression levels of LIF-Stat3 signalling (Socs3, Stat3, Klf4) and BMP-ID signalling (Sox1, Id1 and Id3) for control (day 0) and day 6 batch culture values. *P<0.05, Student t-test. Experimental values represent mean±SD, N = 3.

Figure 7. Pluripotency-related gene expression in perfusion cultures.

Gene expression levels of a) Rex1, b) Oct3/4, c) Fgf5, d) Sox2, e) Dppa3 and f) Nanog. Model simulation results are predicted for Rex1, Fgf5 and Dppa3 (line). *P<0.05, one-way ANOVA. Experimental values represent mean±SD, N = 3.

Figure 8. Gene expression differences between batch and perfusion cultures.

The accumulation of metabolic stress within batch cultures (a) leads to the down-regulation of the expression levels of Rex1 and Dppa3 accompanied by the up-regulation of the Fgf5. Perfusion feeding (b) removes the metabolic stress resulting in the up-regulation of Rex1 and Dppa3 and the down-regulation of the differentiation marker Fgf5. Results were normalised with fresh (day 0) mESCs.