New class of therapeutic smart artificial cells

therapeutic artificial cells

Date: 22nd September 2020

Artificial cells are engineered particles that mimic one or many functions of a biological cells.  They can be used to sense changes in the body or the environment and respond in a variety of ways such as releasing drugs, killing pathogens, or removing pollutants.  They are part of a new generation of platforms for synthesising therapeutic proteins on-demand in diseased tissues.  Now chemists have developed therapeutic artificial cells that can chemically communicate with neurons and promote the differentiation of neural stem cells.

The ultimate goal of a therapeutic artificial cell is to communicate and interact with host cells.  However, many attempts at building these biomimics is focused on synthesising biological-like activity in the laboratory in an absence of other living cells.  Although, a few labs have started to incorporate artificial cells with living ones, the ability of these cells to integrate within the host’s cellular community may provide opportunities to engineer advance drug delivery system.  Yet to date artificial cells that interact with eukaryotic cells under physiological conditions have not been demonstrated.

Now chemists led by Sheref Mansy, from the University of Alberta, Canada, in collaboration with researchers at the University of Trento, Italy, have developed a new class of genetically controlled, stimuli-responsive artificial cells that are formulated to be functional under physiological conditions, and can chemically communicate with mammalian cells.

The team started by synthesising an artificial cellular chassis that was designed to integrate with eukaryotic cells.  The artificial cells were housed within a phospholipid vesicle and contained transcription-translation machinery and DNA templates that coded for brain-derived neurotrophic factor (BDNF), the N-3-oxohexanoyl homoserine lactone (3OC6 HSL)–responsive transcriptional repressor LuxR, and perfringolysin O (PFO) – forming the genetic control of the cell.  PFO expression was controlled by a genetic AND-gate that required both LuxR and its chemical trigger 3OC6 HSL for gene expression.  Monomers of PFO then assembled into pores which allowed the release of BDNF from the cell.  Whilst, BDNF was always synthesised it was only released from the artificial cell in the presence of 3OC6 HSL.

The team wanted to test whether the artificial cells were capable of influencing the differentiation and maturation of mouse embryonic stem cell–derived neural stem (mNS) cells.  Indeed, over a 19 day experiment the artificial cells drove the mNS cells towards differentiating into neurons and inhibited apoptosis only when exposed to 3OC6 HSL.  The team also showed that BDNF-responsive signalling pathways were activated in the mNS cells suggesting the artificial cells had guided the differentiation of neural stem cells into mature neurons in response to an environmental signal.


Next the team wanted to further test the functionality of the artificial cells in the context of human cells.  To do this they engineered a BDNF-responsive human cell line (HEK293T) that upon activation expressed the reporter gene, GFP (green fluorescent protein).  Here, the 3OC6 HSL-activated artificial cells drove expression of GFP in the human cells increasing the number of GFP expressing cell by ~45% over the control samples.  This demonstrated to the team that their system could drive a desired phenotypic change through the controlled release of BDNF.

But were the synthetic cells functional or releasing their cargo due to degradation?  Well importantly the team also showed that both the intravesicular production of BDNF and the membrane pores were functional, suggesting degradation of the cell was not a concern.

Whilst the data so far was encouraging it was important that system could remain stable under physiological conditions. Indeed this was the case, the integrity and function of the cells remained viable at a range of temperatures and osmolality, furthermore there was no toxicity observed suggesting the system was viable in variety of physiological conditions.

Conclusions and future applications

The team here have generated artificial cells that respond to the presence of a small molecule in the environment which stimulated the synthesis and release of a potent protein signal, brain-derived neurotrophic factor. This allowed chemical communication with neurons and promoted the differentiation of neural stem cells.

The work brings the field closer to the construction of artificial cells that advances what it typically envisioned for smart drug delivery systems and opens up new possibilities in therapeutics. The data suggest that artificial cells are a versatile chassis for the in situ synthesis and on-demand release of chemical signals that elicit desired phenotypic changes of eukaryotic cells.  However, whilst this is an important proof-of-concept the team see this as the beginning of the journey for this system.

There are many improvements that could be made; they would like to develop more advanced sensing capabilities of the engineered system for example creating artificial cells that could respond to the physiological changes of the host such as changes in cation or neurotransmitter concentration.  They could be targeted to specific regions or cells in the body such as cancerous cells or damaged regions – engineered to aid healing through stem cell differentiation such as axon or heart regeneration, or to deliver therapeutic drugs to eliminate tumours.  We have recently reported that cell-free exosomes can improve myocardial recovery in infarcted pigs it is not a stretch to envisage artificial cells playing a similar role.

CRISPR-based therapeutics are emerging at a phenomenal rate, with several clinical trials underway, and the first in vivo gene edited patient was treated by Allergan and Editas Medicine earlier this year.  In vivo repair of a single mutation in mice has also shown to partially restore hearing in mice, however targeted delivery of the gene editing machinery in general remains an issue. However, smart delivery systems could transform the field, targeted release of gene editing machinery in response to distinct physiological conditions from artificial cells could be an exciting development.

Whilst, this platform is still far from clinical translation, the work has highlighted its potential and opened up the opportunities for this new class of genetically controlled, stimuli-responsive artificial cells.  Whilst there are still many hurdles still to jump, the next step – transforming the platform for in vivo use – will be a crucial milestone for this technology.

For more information please see the press release from the University of Alberta

Toparlak, Ö. D., J. Zasso, S. Bridi, M. D. Serra, P. Macchi, L. Conti, M.-L. Baudet and S. S. Mansy (2020). “Artificial cells drive neural differentiation.” Science Advances 6(38): eabb4920.