Mechanisms of radiation resistance in Arthrospira sp. PCC 8005

Yadav Anu


Cuypers Ann, (UHasselt),

SCK•CEN Mentor

Janssen Paul
+32 14 33 21 29

SCK•CEN Co-mentor

Leys Natalie
+32 14 33 27 26

Expert group


PhD started


Short project description

Arthrospira are cyanobacteria with a high protein content and rich in carbohydrates, essential fatty acids, and a variety of minerals, vitamins, and nutritional pigments such as beta-carotene. Hence, Arthrospira species have a long history of human consumption (now being commercialised under the name ‘Spirulina’) and are also used worldwide as feed for fish, poultry, and farm animals. Besides their nutritional value Arthrospira species are known for their clinical relevance (e.g. as antioxidants and immunity modulators), their potential in bioremediation and biosequestration, and their applied use in CO2 mitigation and biofuel technology. The Arthrospira sp. PCC 8005 strain was selected by the European Space Agency as an oxygen producer as well as a nutritional end-product for the life support system MELiSSA.

When we investigated the genomic stability of irradiated PCC 8005 cells (since this strain would be used in space missions hence subjected to cosmic radiation) we were surprised to find that this strain could withstand up to 5,000 Gray of gamma radiation. To put this in perspective, the average dose for 50% lethality (LD50) in mammals is about a thousand times less, or 5 Gy. The classical dogma of radiobiology is that cellular damage by ionising radiation (IR) lays primarily at the genetic level i.e. either during or immediately after the transfer of radiation energy to genomic DNA. Indeed, IR causes DNA double stranded breaks in the narrow range of ~0.002-0.008 DSB/Gy/Mb. These DNA lesions are highly cytotoxic because replication cannot proceed. Hence, it was first thought that IR-resistant organisms would possess very efficient DNA repair mechanisms. Although this generally holds true, IR-resistant and IR-sensitive organisms suffer the same number of DNA DSBs for an equivalent dose of IR and often possess similar  DNA repair pathways. It now has been recognized that proteins are major targets of radiation damage through the action of reactive oxygen species (ROS). These ROS, which also cause oxidative damage to DNA, lipids, and other biomolecules, mainly result from the radiolysis of water since water is the most abundant chemical in living cells. Oxidative damage of a protein may render it dysfunctional with a potential detrimental effect on the cell depending on the its function and role. Particularly, an efficient protection of DNA repair enzymes against ROS attacks seems paramount to an enhanced radioresistance. This protection involves antioxidant or ROS-scavenging enzymes, pigments, metal ions (e.g. Mn[II]-complexes) or other ions, compatible solutes, as well as free aminoacids or small peptides.

In addition, IR-resistant bacteria seem to have acquired molecular adaptations to their DNA repair proteins, giving raise to novel enzymatic functions as well as to new substrate specificities. In fact, an IR-sensitive bacterium like E. coli can acquire a 10,000 fold increase of IR-resistance by undergoing iterative cycles of irradiation and outgrowth. Full-genome sequencing of isolates that evolved such extreme IR-resistance showed that mutations within these isolates were mainly situated within ten loci whose genes could be generally grouped in four functional catagories: (i) suppression of oxidative damage (rsxBD, gsiB, fnr), (ii) DNA repair and replication (recA, dnaBT/priAC, yfjK), (iii) basic respiration (glpD), and (iv) cell wall biogenesis (clpXP, wcaMKC, nanET).

The Arthrospira sp. PCC8005 genome was sequenced by us (Janssen et al., JBact 192(9): 2465–2466) (v1) and a Nimblegen tiling array ('Arthrospira HX12') holding >135,000 probes was designed. These probes were mapped to the improved sequence data (v5 - six ordered contigs; EMBL database acc. nr. GCA_000176895) of the genome, covering a total of 5,865 coding regions (CDS) and 3,141 intergenic regions. Detailed genome data were made available by us via the genome annotation platform MaGe (Genoscope, Evry, France – subject to account registration). These data indicate that Arthrospira sp. PCC 8005 makes use of the RecFOR pathway for ds-DNA repair and has normal sets of ss-DNA repair proteins (e.g., UvrABCD, MutST) while it lacks other proteins known for their involvement in DNA repair (e.g., the regulatory protein LexA, and RecBCD). On the level of protection against protein oxidation, a number of ROS-scavenging proteins and enzymes were identified, e.g. superoxide dismutase and at least two peroxiredoxins, but no catalase.

Using the above tiled array of probes we performed gene expression profiling of Arthrospira sp. PCC 8005 via the extraction of total RNA from cells exposed to 3,200 and 5,000 Gy of gamma radiation. The association analysis between the induction or repression of genes following irradiation and the underlying genetic mechanisms and biochemical pathways however proved to be difficult because most IR-affected genes were 'conserved hypothetical' (i.e. genes whose geneproduct show only a sequence match to database proteins of unkown function) or 'hypothetical' (i.e. genes that code for proteins with no match to the public protein databases). Nonetheless, a number of radiation-affected genes were assigned to photosynthesis, pigment biosynthesis, energy production, carbon fixation, oxidative stress response, heat-shock proteins, and DNA repair.


This project is aimed at providing a better insight into the genetic mechanisms and biochemical pathways required for IR-resistance in Arthrospira sp. PCC 8005. A major obstacle for genetical research on Arthrospira sp. PCC 8005 is that Arthrospira species cannot be transformed by DNA i.e. no reliable genetic system exists to date. In addition, the generation of stable mutants is cumbersome although novel random mutagenesis methods have been developed and could be tested. Because genetic research is not immediately feasible, the project will rely initially on  bioinformatic analyses of existing and newly obtained proteomic and transcriptomic data.

(A) for a wide range of IR-resistant prokaryotes, a full catalogue of genes involved in all aspects of radioresistance, must be made, including genes that are, according to literature, involved in ROS-avoidance and –detoxification, general antioxidant functions,  metal ion homeostasis, oxygen transport, redox sensing, cell wall synthesis, cell division, SOS-response, DNA-repair, and other cellular processes reportedly affected by IR. Based on this catalogue (which would entail building a small database of protein sequences for all selected genes) a search will be undertaken for Arthrospira orthologs by performing a BLASTp operation of this database against the four Arthrospira proteomes known sofar (Arthrospira platensis C1 and NIES-39, Arthrospira maxima CS328, and strain PCC 8005). In parallel, protein sequences corresponding to IR-affected PCC 8005 genes (fold change between 2 and 0.5) will be collected to form a protein sequence set that will be compared (by BLASTp) to the proteomes from a range of IR-sensitive and IR-resistant organisms. It is expected that hese bioinformatic analyses will give first clues on the distribution of particular genes and their products across multiple organisms and may result in a broad correlation between the presence and absence of certain genes, or the differential expression of certain genes, in respect to IR-resistant and IR–sensitive phenotypes.

(B) because over 80% of the IR-affected PCC 8005 genes are hypothetical or conserved hypothetical, renewed efforts must be undertaken to improve the functional annotation of these genes. This could entail structural genomics, analysis of domain architecture, position-specific iterative BLAST, gene network analysis, pathway analysis, phylogenetics, and regulatory sequence analysis (of intergenic regions).

(C) as mentioned in the introduction, acuired IR-resistance in E.coli strongly relates to mutations in only a handful of genes.  The PCC 8005 genome contains five copies of clpP, two copies of clpX, and one copy each of nanE, priA, glpD, and recA. It may be worthwhile to analyse the corresponding protein sequences for particular domains and alterations of catalytic residues in respect to mutant and wildtype sequences in E. coli and orthologous sequences in IR-resistant organisms, in the hope to find clues for the Arthrospira sp. PCC 8005 IR-resistant phenotype on the molecular level. This line of investigation may seem a long shot but it cannot be excluded.

(D) it is unknown how many DSBs occur in the PCC 8005 genome when cells are exposed to discrete doses and types of radiation. An effort should be made to establish reliable DSB yield assays, testing a range of experimental conditions. Likewise, accurate antioxidant assays should be developed. With these assays up and working, it will become possible to (i) test Arthrospira IR-sensitive phenotypes (of natural or mutant strains, see below) for DSB yields and (ii) investigate whether extracts (or purified compounds after gene cloning and heterologous expression) of PCC 8005 have antioxidant properties and whether these Arthrospira cellular components can provide radiation protection to other cells.

(E) it is also unkown whether the IR-resistant phenotype of strain PCC 8005 is strain-specific or whether it is a common trait of Arthrospira. Therefor a range of additional Arthrospira strains (including the three above mentioned strains C1, NIES-39, and CS328 for which genome data are available) need to be tested for their capacity to withstand high doses of IR. Strains with an IR-sensitive phenotype, if they exist, need to be analysed at the molecular level. In case no genome data are available for such IR-sensitive Arthrospira, full genome sequencing must be considered as to generate appropriate genome data, or at least genes and their products potentially involved in IR-resistance (as evidenced from the above mentioned approaches) from these strains need to be characterized and further analysed at sequence level.

(F) bioinformatic approaches have only a predictive function. To identify PCC 8005 genes directly responsible for the IR-resistant phenotype it is necessary to generate mutants and screen for an IR-sensitive phenotype. However, random chemical or physical mutagenesis does not allow to readily locate the introduced mutation(s) while site directed mutagenesis or gene-specific mutagenesis by cloning and homologous recombination is currently impossible owing to the multiple defense mechanisms in Arthrospira against foreign DNA transfer. Although success is uncertain, a special challenge would be to develop an Arthrospira genetic system. This would not only allow gene-specific mutagenesis but also generally give Arthrospira research a tremendous boost.