Team:Groningen/Brainstorm/Buoyant Bacteria




About 150 species of prokaryotes (well studied examples are cyanobacteria and halophilic archaea) in aquatic habitats have been shown to contain gas vesicles. These gas vesicles provide cells with bouyancy, there function is either to positioning the bacterium (in water) in order to get the right amount of oxygen or the right amount of light. Gas vesicles are made exclusively of proteins and contain gas. When gas vesicles are present in a cell, the overall density of that cell is lowered and the cell becomes bouyant.

We want to utilize this bouyancy for an application like for example separation of specific molecules or specific cells.

Gas vesicles are hollow proteinous organelles made of gvpA and gvpC (in cyanobacteria) and are permeable to gases and fills by diffusion. It is impermeable to water because of its hydrophobic inside. GvpA is a small 70AA long protein which forms a linear crystalline array of ribs to form the cylindrical shell and conical ends. GvpC is usually on the outside of the gas vesicle to make it stronger and stabilizes the structure.

In B. megaterium a gas vesicle gene-cluster was found, which contained 14 gvp genes (gas vescicle polycitonic genes) which were functionally expressed in E. coli by Ning Li and Maura Cannon (Li and Cannon, 1995). The best bouyant fenotype was found when gvpA, gvpP, gvpQ and ORF-1 were excluded. It was suggested that gvpB on the gvp-cluster of B. megaterium is an homolog of gvpA. 900px

Gas vesicles in iGEM

In iGEM 2007, Melbourne created a biobrick for gas vesicle formation. In iGEM 2008 Kyoto had the idea to lift the titanic from the bottom of the sea with the help of bouyant bacteria.

Melbourne iGEM 2007 team constructed a bouyant E. coli and also they added a BioBrick of the short version of the gvp-cluster (without gvpA,P,Q, ORF-1 and AraC). This BioBrick is available and also in the microtiter plate send by HQ to us, so we can use this. Melbourne changed the gvp-cluster by cleaning it from 3 EcoRI sites and 1 PstI site, this leaded to a accidental addition of a 10x repeat of "TCTGCAAATTA". They mention that they added the BioBrick prefix and suffix to the BioBrick, though the restriction sites of these additions cannot be found by CloneManager in the sequence available on the Partregistry. Hopefully the part is available on a standard vector, which has the pre- and suffix for BioBricks. For cloning this part we can use the optimized protocol (restriction on part with XbaI and SpeI, on the vector with SpeI) for ligation of the gvp-cluster and a vector BBa_J61035 (3539bp, copy nr??), this unluckaly leads to ligation of the gvp-cluster in an unspecific direction. This can of course be tested by restriction, PCR or sequencing, but it takes more time as another step will be introduced. The vector has two selection markers: Ampicillin and Gentamycin.

Alternative cloning strategy

Another possibility is to use eg. XbaI and SpeI and also cut a vector with these enzymes, this would lead to ligation of the gvp-cluster in one direction. A possible vector for this strategy on the partsregistry is: BBa_J23018 (2298 bp).

Missing information

  • Used promotor for expression of the gvp-cluster:
    • Inducible (may be used for proof-of-principle)
    • Constituative (may be used for proof-of-principle)
    • Metal sensitive
  • What kind of vector was used by Li and Cannon (1995) or Melbourne (2007)? Is there a negative effect of high copy number?
  • What is the maximum amount of pressure gas vesicles can handle? At which depth would this be, how can one put this kind of pressure on a water column?
  • What is the density of gas vesicles in cells (normally or in case of over-expression)
  • Modelling parameters:

-Cyanobacteria (Bowen and Jensen, 1965): gas vacuoles made up of gas vesicles (75 nm in diameter and up to 1.0 ,um in length, single wall layer only 2 nm thick) 0,7MPa gives irreversible loss of buoyancy fenotype, but found in next generation.


Walsby, A.E. 1994. Gas Vesicles. Microbiological reviews. Vol. 58, No. 1, p. 94-144.

Li, N., Cannon, M.C. 1998. Gas Vesicle Genes Identified in Bacillus megaterium and Functional Expression in Escherichia coli. Journal of Bacteriology. Vol. 180, No. 9, p. 2450–2458.

Astrid C. Sivertsen et al. 2009 Solid-State NMR Evidence for Inequivalent GvpA Subunits in Gas Vesicles. Journal of Molecular Biology 387, p1032–1039.

Lewinson O., Lee A.T., Rees D.C. 2009. A P-type ATPase importer that discriminates between essential and toxic transition metals. PNAS. vol. 106, no. 12, p. 4677-4682.