Cryopreservation and re-culture of a 2.3 litre biomass for use in a bioartificial liver device

Abstract

For large and complex tissue engineered constructs to be available on demand, long term storage using methods, such as cryopreservation, are essential. This study optimised parameters such as excess media concentration and warming rates and used the findings to enable the successful cryopreservation of 2.3 litres of alginate encapsulated liver cell spheroids. This volume of biomass is typical of those required for successful treatment of Acute Liver Failure using our Bioartificial Liver Device. Adding a buffer of medium above the biomass, as well as slow (0.6°C/min) warming rates was found to give the best results, so long as the warming through the equilibrium melting temperature was rapid. After 72 h post thaw-culture, viable cell number, glucose consumption, lactate production, and alpha-fetoprotein production had recovered to pre-freeze values in the 2.3 litre biomass (1.00 ± 0.05, 1.19 ± 0.10, 1.23 ± 0.18, 2.03 ± 0.04 per ml biomass of the pre-cryopreservation values respectively). It was also shown that further improvements in warming rates of the biomass could reduce recovery time to < 48 h. This is the first example of a biomass of this volume being successfully cryopreserved in a single cassette and re-cultured. It demonstrates that a bioartificial liver device can be cryopreserved, and has wider applications to scale-up large volume cryopreservation.

Fig 1. The three experimental designs tested.

A–The cryovial set-up (CV). Samples were added to cryovials and cryopreserved as per the test conditions. B–The scale down process set-up (SDP). 6ml vials were filled with ELS. The vial was then insulated–indicated by the black outline–so that heat could escape only from the base of the vial–indicated by blue arrow. Samples in this set-up experienced the same conditions as a 2.9 litre volume on cooling, but were thawed rapidly. C–The large volume set-up (LV). 2.3 litres of biomass was added to the freezing chamber—indicated by the pink hemisphere–with 0.6 litre supernatant on top, indicated as red. A large air fraction existed above the biomass. The CV and SDP vials are shown to scale in C.

Fig 2. Warming profiles experienced during warming of the large volume cryopreservation cylindrical chamber.

Thermocouples were placed at the bottom of the biomass (black) and the top of the biomass (lightest grey), as well as three others equidistant apart between the bottom and top following a straight line through the deepest part of the sample (dark to lighter grey). Section A demarks warming in the -80°C freezer, section B thawing in the -30°C freezer, and section C -10°C in the Planer controlled rate freezer. Relatively little intra-sample temperature variation was observed, a maximum of 22°C difference was observed at the start of the warming process, diminishing to 9°C intra-sample variability after the 4 h warming process. Average warming rate was 0.6°C/min.

Fig 3. Viability (left) and viable cell number (right) of samples being warmed at 1°C a minute from -196°C, to the temperatures indicated, before being warmed rapidly to thaw in a water bath (in 2ml cryovials).

Data 24 h post-thaw. The rightmost sample (4°C), was warmed slowly until thaw. No significant difference was seen in viabilities. All samples warmed rapidly through the phase transition had significantly increased viable cell number over the samples warmed slowly (4°C sample), viable cell number was 16.3 ± 1.7 million cells/ml immediately prior to cryopreservation. N = 5 ± SD, significance defined as * = P<0.001, unpaired Student’s t-test.

Fig 4. Viability (left) and viable cell number (right) of samples cryopreserved with 20% of the volume excess medium (in 2ml cryovials).

24 h post-thaw no significant difference was observed in either viability or viable cell number between any sampling location. viable cell number was 18.2 ± 1.6 million cells/ml immediately prior to cryopreservation N = 5 ± SD.

Fig 5. Top; viability (left) and viable cell number (right) post-thaw of the large volume cryopreservation.

Base: Viability (left) and viable cell number (right) of cryopreserved cryovials (CV), scale-down process vials (SDP), and the large volume (LV) (light grey, dark grey, and black respectively). The viability was significantly higher in the scale-down process vials compared to the large volume cryopreservation at all measured timepoints, and was significantly higher in the cryovials over the large volume at 24 h post-thaw. Both the cryovial and scale down process vials samples had significantly higher viable cell number over the biomass at 24 and 48 h post-thaw, the scale-down process vials significantly better also at 72 h post-thaw. N = 5 ± SD, significance over large volume cryopreservation defined as * = P<0.001, + = P<0.005 unpaired Student’s t-test.

Fig 6. Glucose consumption (left) and lactate production (right) following cryopreservation.

Top–glucose consumption and lactate production per ml ELS as a fraction of unfrozen control 72 h post-thaw, comparing the cryovial cryopreservation (light grey), scale-down process cryopreservation (dark grey), and large volume cryopreservation (black). No significant difference was observed in glucose consumption. Lactate production was significantly higher in both the cryovial and scale-down process samples over the large volume cryopreservation. Centre–production at various timepoints in the large volume cryopreservation per ml ELS over an unfrozen control. Base–Total glucose and lactate measured in the bioreactor at set timepoints post large volume thaw. N = 5 ± SD, significance over large volume cryopreservation (top) or unfrozen control (centre) defined as * = P<0.001, unpaired Student’s t-test.

Fig 7. AFP production of large volume cryopreserved samples, and comparisons with cryovials and scale down process samples.

Top: production at timepoints post-thaw as a fraction of unfrozen control, either per ml ELS (left), or per million viable cells (right). Base: comparison between cryovials (light grey), scale down process samples (dark grey), and large volume samples (black) at 72 h post-thaw. Per ml ELS, both conditions produce significantly less AFP than the large volume sample, and per million cells the cryovial production is significantly lower compared with the large volume cryopreservation samples. N = 5 ± SD, significance compared to large volume cryopreservation (top) defined as * = P<0.001, + = P<0.005, unpaired Student’s t-test.

Fig 8. The proportion of extra-cellular ice melted at different temperatures during the warming phase.

Below the glass transition temperature (-120°C) all extra-cellular water is either ice or is vitrified in solute concentrated channels between the ice crystals. On warming the ice crystals melt due to the freezing point suppression of the high solute concentrations, a process which accelerates at high sub-zero temperatures until complete thaw occurs at -4.5°C. Data for a 12% v/v aqueous DMSO solution and adapted from Rasmussen and MacKenzie [48].