Skip to content

Synthetic Life Part II – The ‘Minimal Cell’

June 7, 2011

This post follows on from the previous, with respect to the theme of creating a truly synthetic artificial cell. An important question to consider is just how far one can ‘down-size’ a set of genes for a single-celled organism….

Stripping Down a Cell

If one has ambitions towards generating a wholly synthetic cell which can grow as an independent replicator, an eminently logical approach to start with is to render the problem as simple as possible. In order to do this, it follows that one should identify the genes of such a cell which are absolutely essential, and which do not provide merely an auxiliary function which might only be required in certain circumstances. Natural free-living organisms may encounter a range of different environmental challenges (including a variety of parasites), the evolved genetic responses against which are deployed only when necessary. (It is energetically costly to synthesize proteins or enzymatic products when they are not needed; therefore strict control mechanisms are typically arranged such that the relevant genes are expressed only at opportune times). Yet in the cosseted confines of the laboratory, defense mechanisms or systems for coping with environmental stresses are not essential, and can therefore be removed from a ‘parts list’ of essential genes for a ‘minimal cell’.

Reducing Genomes the Natural Way

It should be noted that functional genetic minimization has many natural precedents, where organisms adapt to environments where some of their ancestral genes become superfluous. A favorite example is the loss of eyes of cave-dwelling creatures whose lives are spent in total darkness. But even more striking cases are seen for single-celled organisms which are obligate intracellular parasites, such as Rickettsial bacteria, agents of some human diseases (noted in a previous post). And this trend of genomic simplification is taken to a much greater level in the case of eukaryotic organelles, as also previously visited. Although mitochondria and chloroplasts perform essential services for their host cells, these organelles have divested much of their functional needs to proteins encoded by the host nuclear genome. And the result of this is a great reduction in genomic size relative to their remote free-living ancestors. All this can be seen fairly simply in evolutionary terms: loss of a non-essential gene can provide a fitness advantage over competitors within the same species which retain the original gene complement. Consequently, over evolutionary time, there is a trend favoring such losses, which will occur even if the relevant fitness benefit is a small increment.

Defining the Minimal Genome

Nature thus shows us many examples of genomic reduction, which (if the organism in question is to survive) must stop short of an essential gene set required for replication itself. But for artificial purposes, how can one go about defining the minimal gene set required? In essence, three types of approach can be defined. Firstly, the large and constantly increasing genomic databases allow comparisons between closely-related and phylogenetically distant organisms. Genes which have homologs persisting across wide evolutionary gulfs are argued as having a high probability of being indispensable. When applied to prokaryotes, systematic application of this comparative method can give rise to estimates of the minimal essential gene set. While this is a convenient strategy when data is available, problems can occur, and it is far from fool-proof. For example, if broad evolutionary spans are compared, some essential genes may have been substituted by genes with non-homologous coding sequences. In such circumstances, a candidate gene will find no cross-species match, and thereby be spuriously classified as ‘non-essential’.

A second approach empirically evaluates the effect of systematically inactivating all genes of an organism, combined with assessing the effect of such ‘knock-outs’ on cellular viabilities. Where cells can still survive and grow, the specific genes in each case which have undergone transposon-mediated ablation are hence deemed as dispensable. If done on a genome-wide scale, a picture of the minimal gene complement can emerge. Among a range of alternative methods, this can be conducted by ‘global’ transposon mutagenesis, where a selectable transposable element inactivates genes by random insertion into genomic sites. This approach has been applied to Mycoplasma genitalium, the bearer of the smallest natural genome supporting independent growth in defined laboratory media. Yet this technique too can have pitfalls for the unwary. For example, if an essential metabolic function makes use of a set of redundant gene copies (or functionally similar gene variants), knockout of any single gene will be dispensable, but ablation of the whole set will be lethal. But uncritical use of transposon mutagenic data would conclude that all these genes were dispensable, and removable from the minimal-essential list. Rational application of accumulated high-throughput data can potentially be of great help in sorting the truly essential from the dispensable gene set. But owing to the often-complex inter-relationships between biochemical pathways, a full ‘minimal’ replicative and metabolic organismal definition requires a systems-level approach and hence falls within the domain of systems biology (touched upon in a previous post). Techniques for global analysis of the transcriptome, proteome, and metabolome for a small-genome Mycoplasmal species (M. pneumoniae) have been performed, which will be valuable in helping define the essential gene sets of related mycoplasmal species.

The third approach has an inherently different aspect to the first two. Both of these begin with existing genomes, and attempt to take them to their ‘bare bones’ state. As such, these have been termed ‘top-down’ strategies. The alternative (perhaps not surprisingly) is called a ‘bottom-up’ pathway, where one tries to deduce all required parts for a minimal cell, and accordingly construct and deploy them. This aim thus requires a biochemical definition of all the functional processes which must take place in order to allow cell maintenance, growth, and replication. If this is accomplished, the corresponding genes which encode the necessary proteins and RNAs mediating these essential functions can be likewise defined. This more fundamental approach has led Forster and Church (in 2006) to the conclusion that only 151 specific genes would enable a minimal cell. This estimate does not include biosynthetic genes (such a cell would require full nutritional supplementations), nor genes for lipid or polysaccharide metabolism. (It was assumed that appropriate lipid vesicles could self-assemble and auto-catalytically propagate themselves, based on other experimental data).

Genome Minimization for Fun and Profit

The systematic reduction of genomes aims to define to the smallest set of genes which continues to permit the host cell to exist as an independent replicator, and is thereby an important stepping stone in the noble quest towards the artificial generation of a cell which is wholly synthetic. This would unquestionably constitute ‘synthetic life’.

Yet this venture is certainly more than a starry-eyed example of the idealistic pursuit of scientific knowledge, as real benefits can ensue. Minimization can offer considerable advantages for biotechnologists. Not surprisingly, one of the first targets for this sort of enterprise was E. coli, the long-standing work-horse of molecular biologists. The same group which first sequenced the common K-12 strain of this familiar bacterium also became involved in a corresponding genomic minimization project. They achieved a 15% reduction in genomic size, which involved not just the removal of genetic redundancies or non-essential genes, but also potentially parasitic insertion sequences. The absence of the latter reduces the ‘evolvability’ of the organism, or the capacity of the organism to undergo evolutionary adaptations in the face of environmental changes. In turn, this reduction in evolvability is associated with the ablation of insertion sequence-mediated mutagenesis. While compromised adaptability may not be a very good thing for free-living organisms, biotechnologists prize stability of engineered microbial factories, and therefore this becomes an attractive feature. Another group has combined known E. coli deletions, and found that an entire megabase can be removed from the E. coli chromosome, down to 3.6 Mb (a 22% reduction). Moreover, this reduced-genome cell had a superior growth rate to the wild-type strain.

In the terms of the relevance to humans of bacteria as chemical factories, the genus Streptomyces has long been an exceptionally productive source of low-molecular ‘secondary metabolites’. As the ‘secondary’ tag implies, synthesis of these compounds is not essential to the operation of Streptomyces host cells themselves. For engineering of these cells to make novel metabolites, however, new sets of biosynthetic genes may be necessary. In such circumstances, genes specifying synthesis of endogenous metabolites might potentially lower efficiency, through production of similar but unwanted products, or competition for required basic building block molecules. Accordingly, removal of endogenous metabolite production from Streptomyces by a genomic reduction process would be biotechnologically useful, and this has been achieved. Where bioengineering involves transfer of large multigene blocks specifying entire genetic pathways to a modified host cell, we can be reminded of a comment in the previous post, to the effect that it is not necessary to synthesize an entire genome in order to conduct research and product development in metabolic engineering.

But in any case, a subset of the larger field of genomic engineering, it is clear that genome reduction to a functionally useful minimal state can be a profitable enterprise, and certain commercial interests are involved in this area.

And on that note….

It may be quite easy to show

Which genomic regions can go

As genomes thus shrink

It’s striking to think

At the same time, profits may grow.

References & Details

(In Order of Citation)

A favorite example is the loss of eyes of cave-dwelling creatures whose lives are spent in total darkness.’ Among other examples, we can consider loss of ability to synthesize Vitamin C by primates (since this vitamin is normally ubiquitous in the primate diet), and loss of sweet taste receptors by cats.

‘…..a fully synthetic cell…’ ‘Fully’ is added here to help distinguish a completely synthetic cellular replicator from a Venter-group version of a so-called ‘synthetic cell’ bearing a synthetic genome. See the previous post for a discussion of related issues.

‘….three types of approach can be defined.’ These have been put forward and discussed in more detail by Forster & Church, 2006.

‘…..this can give rise to estimates of the minimal essential gene set.’ A number of these kinds of studies have been published. For example, See Mushegian & Koonin, 1996. They used the sequence of Mycoplasma genitalium (as the smallest known genome of an independent organism, as noted in the previous post), and deduced by comparative genomics that its initial 468 protein-coding genes could be reduced to a core of 256 essential genes.

‘… conducted with transposon mutagenesis….’ This is a very large field in its own right, since all cellular organisms are subject to invasion by adventitious mobile genetic elements. As an example of a screening strategy combined with a specific E. coli transposon mutagenesis, see Winterberg et al. 2005.

This approach has been applied to Mycoplasma genitalium…….’ See Glass et al. 2006, which refined an earlier transposon mutagenic study in concluding that 382 of a total of 482 M. genitalium genes were essential. Interestingly, 28% of the essential genes were of unknown function. Subsequent proteomic analysis in the closely related M. pneumoniae has given some information regarding possible functions for some of these genes, but more work is required.

Techniques for global analysis …… for a small-genome Mycoplasmal species….’ See Güell et al. 2009 (transcriptome); Kühner et al. 2009 (proteome); and Yus et al. 2009 (metabolome).

‘….termed ‘top-down’ …… ‘bottom-up’…..’ For a more detailed discussion of this, see Jewett & Forster 2010.

‘….has led Forster and Church (in 2006) to the conclusion that only 151 specific genes would enable a minimal cell. ….’ See Forster & Church, 2006.

‘….which first sequenced the common K-12 strain of this familiar bacterium…’ This was done by Blattner et al. 1997, as noted in the previous post.

‘…….achieved a 15% reduction in genomic size…..’ See Posfai et al. 2006. (More will be said about some aspects of this in the next post). ‘….reduces the ‘evolvability’ of the organism….’ See Umenhoffer et al. 2010.

‘…..Another group has combined known E. coli deletions…..’ See Mizoguchi et al. 2008.

‘…..removal of endogenous metabolite production from Streptomyces…’ See Komatsu et al. 2010.

‘……certain commercial interests are involved in this area.’ As a case in point, Scarab Genomics markets a reduced-genome version of E. coli.

Next Post: Two weeks from now.


No comments yet

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s