A number of DNA lesions are generated in each cell every day, among which
double-stranded breaks (DSBs) constitute one of the most detrimental types of DNA
damage. DSBs lead to genome instability, cell death, or even tumorigenesis in human,
if not repaired timely. Two main pathways are known for DSB repair, homologous
recombination repair (HRR) and Non-homologous end joint (NHEJ). HR repairs
DSBs using a homologous DNA molecule as a template resulting in error free DNA
repair, whereas NHEJ promotes direct re-ligation of the broken DNA ends in an
error-prone manner. In eukaryotes DSBs occurred in the S/G2 phase of the cell cycle
are preferentially repaired by HRR pathway, while NHEJ is the favorate pathway to
repair DSBs in the G1 phase. Bacteria encode multiple pathways for DSB repair,
including RecBCD, the primary HR pathway, SbcC-SbcD, and one backup system,
RecFOR. In eukaryotes, the HRR pathway is mediated by Mre11-Rad50, homologs of
bacterial SbcD-SbcC. However, numerous proteins and multiple layers of regulation
exist to ensure these repair pathways are accurate and restricted to the appropriate
cellular contexts, making many important mechanistic details poorly understood in
eukaryotes. As a third domain of life, archaea is considered as a chimaera between
bacteria and eukaryotes. Its metabolic pathways and cell structures resemble those of
bacteria, whereas the information processing is of the eukaryal type or more similar to
their eukaryal counterparts. A number of archaea live in harsh conditions, such as high
temperature, high pressure, extreme pH, or strong radiation, which introduce more
DNA damages in genomes than normal environments. But the stability of archaeal
genomes is comparable with that in other two domains of life, suggesting that archaea
could harbor more efficient DNA repair systems. Study on archaeal DNA repair will
provide important clues for that on eukaryotes. The archaeal homologs of
Mre11/SbcD-Rad50/SbcC, but not RecBCD or RecFOR, have been identified,
indicating the existence of a Mre11-Rad50-mediated HRR pathway in archaea.
Eukaryotic Mre11-Rad50 complex exhibits ATPase activity, 3’-5’ double-stranded
DNA (dsDNA) exonuclease activity and single-stranded DNA (ssDNA) endonuclease
activity. The MRX/MRN complex (MR complexed with the third protein Xrs2
(Saccharomyces cerevisiae)/Nbs1 (higher eukaryotes)) initially processes broken
DNA ends in conjunction with Sae2/CtIP. The subsequent extensive processing is
carried out by the two parallel pathways, Exo1/EXO1 or Dna2/DNA2-Sgs1/BLM,
forming a long 3’-tail of ssDNA to be utilized in Rad51-dependent strand exchange in
HRR. The activities of archaeal MR complex are similar to that of eukaryotes. The
RecQ-like helicase Hjm and the 5’-flap endonuclease which exihibited both
endonuclease and 5’-3’ exonuclease activities have been identified in archaea;
however, it is unclear whether they are involved in dsDNA end resection. Intriguingly,
two other genes, encoding ATPase/helicase HerA and nuclease NurA, respectively, are
implicated in HR by their genetic association with mre11 and rad50 in thermophilic
archaea. This has been supported by biochemical characterization of the encoded
proteins that Mre11, Rad50, HerA, and NurA are capable of working in concert to
process dsDNA to from a 3’-overhang in vitro. So far, very few genetic studies have
been reported, especially for HerA and NurA. In this study, we investigated the
functions of HerA and NurA using the well developed genetic system of the
hyperthermophile Sulfolobus islandicus, combined with biochemical characterization,
cytologic, and transcriptomics analyses, in order to reveal their in vivo roles and
mechanism of these proeins.
In the previous study, it has been shown null mutants were not obtainable for
mre11, rad50, herA, and nurA in S. islandicus, suggesting that all of them could be
essential for cell viability. Here, their essentiality was further investigated and
confirmed by mutant propagation assay. Given the previous results that neither radA,
mre11, rad50, herA, and nurA mutant in Thermococcus kodakaraensis, another
hyperthermophile in Euryarchaea, or radA and hjm in S. islandicus could be isolated,
we speculated that HRR may be essential in the thermophilic archaea.
To further characterize the essentiality of the archaea-specific proteins, HerA and
NurA, mutant genes coding for proteins containing point mutations at the conserved
amino acid residues were constructed and used for genetic complementation in S.
islandicus. For HerA, among the six mutants, K154R, D176E, D176N, E356D,
E356Q, and R381K, only D176E complemented the deficiency of the wild type HerA,
indicating that the Walker A (K154), Walker B (E356) and Arginine finger motif
(R381) as well as the conserved residue D176 of HerA are indispensable for its in vivo
functions. For NurA, two central residues related to its nuclease activity, D58 and
K202, and two hydrophobic residues involved in the interaction with HerA, I295 and
F300, were chosen for mutation. We showed that neither of D58E, D58A, K202R, nor
K202A was able to complement the wild type NurA, suggesting that the nuclease
activity of NurA is essential for cell viability. Furthermore, I295L and F300Y
substitutions were found to be able to achieve complementation, whereas I295E and
F300E failed to do so. The interaction between the HerA and NurA mutants I295L,
I295E, F300Y, and F300E of S. islandicus were further examined by gel filtration,
which revealed that I295L and F300Y maintained the interaction with HerA while
I295E and F300E did not. Taken together, these results indicate that the interaction
between NurA and HerA is essential for cell viability.
To figure out what activities of mutant proteins that were required for the genetic
complementation and to reveal the in vivo roles of the proteins, the wild type and
mutant genes of HerA and NurA were cloned and the proteins were expressed in, and
purified from, E. coli. Biochemical characterization showed that the ATPase activity
of HerA(K154R), D176N, E356Q, and R381K, which failed to exert genetic
complementation, were very low or undetectable, suggesting that the ATPase activity
is essential for the in vivo functions of HerA. And D176E and E356D contained about
1/7 and 1/5, respectively, of the wild type ATPase activity. However only D176E
could complement the deficiency of chromosomal herA, suggesting that the ATPase
activity of HerA was not sufficient for its cellular function. Due to failure to detect
HerA helicase activity, the DNA degradation activity of the HerA-NurA complex was
analyzed and showed that the 5’-3’ exonuclease activity of E356D-NurA reduced to
less than 50% of wild type HerA-NurA while D176E-NurA maintained this activity as
high as that of the wild-type. This suggests that efficient 5’-3’ exonuclease activity is
indispensable for cell viability, which is essential to produce 3’-overhang for HRR
and represents the in vivo activity of HerA-NurA in the cell.
Further, using protein-specfic antibodies and immunofluorescence microscopy,
we examined foci formation of HRR proteins in S. islandicus cells. Under the
physiological growth conditions, a majority of cells harbored one or two HerA foci.
The number of cells with more than two HerA foci increased after UV-irradiation,
suggesting that HerA could be involved in the repair of UV-induced DNA damage.
The pattern of NurA foci was similar to that of HerA, while the numbers of RadA foci
in most cells were 0 or 1, and did not increase apparently after UV-treatment,
indicating that RadA may work differently from HerA-NurA.
To better understand other putative functions of HerA in vivo, this protein was
overexpressed in S. islandicus cells. We found that HerA overexpression reduced cell
viability and produced abnormal cells with enlarged size and increased DNA contents,
as shown by microscopy and flow cytometry. DNA damaging agent assay showed that
this strain is as sensitive as the wild type strain to methyl methanesulfonate (MMS)
and cisplatin, while it exhibited higher sensitivity to hydroxyurea (HU), an agent
revealed to cause G2 arrest in S. islandicus cells, compared with the wild type strain.
Microarray analysis showed that genes involved in cell division were down-regulated
while the transcription of the genes implicated in chromosome resolution/segregation
were also changed in the strain, suggesting that HerA overexpression impair DNA
metabolism and resulted in mis-regulation of cell cycle.
Finally, we construted a S. islandicus strain chromosomally encoding an
N-terminal His-tagged HerA by introducing a his-tag-coding sequence at the 5’ end of
herA gene. The His-tagged HerA and its putative interaction proteins were purified
from S. islandicus cells. NurA as well as two other proteins probably involved in HRR,
ATPase (SiRe_1432) and Holliday junction resolvease Hjc (SiRe_1431), were
identified in the fractions. The interactions between HerA and ATPase and Hjc were
confirmed by in vitro pull-down assay. This result provided clues for further
investigation into the mechanism of the pathway(s) in which herA proteins are
involved.