Identification of UV irradiation-resistant microbes.
Initially, samples from several spacecraft surfaces and associated assembly facilities were subjected to a heat shock protocol (2) to isolate sporulating bacteria. Spores of 43 strains (Table 1) were screened for UV254 resistance using an Hg lamp. Nineteen strains exhibited growth after receiving a dose of 1,000 J m–2. Phylogenetic analyses were performed for all the strains tested, and the strains were unambiguously determined to be low-G+C-content gram-positive Firmicutes based on 16S rRNA gene sequence analysis. The 16S rRNA gene sequences of all isolates were compared, and a bootstrap analysis (500 replicates) was performed to avoid sampling artifacts. The analyses indicated that the strains tested exhibited close phylogenic relationships with Bacillus species. Neighbor-joining, parsimony, and maximum-likelihood analyses were performed with this subset of bacteria using several subdomains of the 16S rRNA genes. A maximum-likelihood phylogenetic tree based on 16S rRNA gene sequences of several Bacillus species is shown in Fig. 2. The branching order of this tree showed that there were two distinct clusters. The top clade consisted of Bacillus rRNA group 1 species, and the other clade was formed by species belonging to rRNA group 2, as well as the rRNA group 1 Bacillus species that produce exosporium-like structures (38).
The levels of 16S rRNA gene nucleotide sequence similarity between the strains tested in this study and closely related Bacillus
species, recognized by GenBank "BLAST" searches, were between 91.6 and 99.8%. Sequence variation of B. subtilis
, B. mojavensis
, B. atrophaeus
, B. licheniformis
, and B. pumilus
). The strains of B. subtilis
and B. pumilus
sequenced in this study were tightly bound phylogenetically, as all of them exhibited levels of 16S rRNA gene sequence similarity of >97.5%. Additionally, a high level of 16S rRNA gene sequence variation (
8%) was observed between members of rRNA group 1 and members of rRNA group 2; however, a high degree of dissimilarity within a well-described genus is not uncommon. Based on 16S rRNA analysis, five of the environmental strains tested, B. pumilus
FO-33, FO-36b, SAFN-036, and SAFR-032 and B. subtilis
42HS-1, were identified as members of rRNA group 1, and two strains, B. odysseyi
and B. psychrodurans
VSE1-06, were identified as members of rRNA group 2 (Table 1
Martian UV irradiation simulation.
The JPL and KSC lab spectra were very similar to the models in terms of shape and level, and the KSC spectrum was more representative (Fig. 1). The JPL spectrum was deficient in 250-nm energy and exhibited a significant dip at 330 nm. This resulted in lower energies in the UVC and UVA bands and very similar levels for the UVB band. The 330-nm dip seen in the JPL spectrum was most likely due to lens coatings in the solar simulator but was not deemed to be critical because most of the absorption band was in the UVA region of the solar simulation. Our results (see below) indicated that UVA had a minimal effect on the survival of bacterial spores. The JPL and KSC simulations of Mars solar environments were calibrated to fall within a range of UV irradiation encountered under clear-sky conditions, with an optical depth of angles.
Survival of endospores in aqueous solution under simulated Martian UV irradiation conditions (JPL simulation).
Of the 19 strains that exhibited UVC resistance, seven spacecraft-associated isolates were chosen for further study due to elevated LD90 of UV irradiation (data not shown) or other traits. The bacterial strains chosen and the determining factors used for selection for exposure to simulated Mars UV irradiation were as follows: B. odysseyi was selected because of its morphological novelty (25); B. psychrodurans VSE1-06 was selected because of its low-temperature tolerance; four strains of B. pumilus were selected because of their predominant occurrence; and B. subtilis 42HS-1 was selected because it is a close relative of the well-studied reference strain B. subtilis 168 (6, 14). In addition to the environmental strains mentioned above, B. subtilis 168 was selected as a control since this strain has been used in numerous other resistance studies that have been described previously (13, 14, 16, 17, 33, 46). Similarly, B. megaterium ATCC 14581 was also chosen for further study due to its high UV resistance compared to the UV resistance of other reference strains used in the current study. Most of the B. pumilus environmental isolates were obtained from the JPL spacecraft assembly facility class 100K cleanrooms; the only exception was B. pumilus 015342-2 ISS, which was isolated from surfaces of the International Space Station. Both B. subtilis 42HS-1 and B. odysseyi were cultured from the surface of the Mars Odyssey, and B. psychrodurans VSE1-06 was recovered from air samples collected in the assembly facility for the Mars Exploration Rovers, Payload Hazardous Servicing Facility, KSC (Table 1).
The results of exposing Bacillus spores in aqueous solution to UVA, UVA+B, and total UV are shown in Table 2. Compared to the exposure time for the full UV spectrum, 2- to 25-fold increases in the time of exposure to UVA or UVA+B were required to reduce the viable spore counts by 50%. Likewise, 90% reductions in viable spore numbers required 35- and 140-fold-greater times of exposure to UVA+B and UVA, respectively, than to the full UV spectrum. As shown in Table 2, none of the Bacillus species tested was completely eradicated even after 30 min of exposure to UVA+B irradiation or UVA (Table 2). The LD50s, LD90s, and LD100s of the nine types of bacterial spores tested with various UV spectra in the Mars solar simulation showed that UVA+B irradiation was significantly less lethal than full-spectrum UV irradiation; therefore, as expected, the 200- to 280-nm range is more damaging than longer wavelengths. Although all spores tested exhibited sensitivity to UVA+B, most damage by UVA+B might be attributed to UVB. This was further confirmed by the observation that all the spores tested except the B. pumilus FO-033 spores were resistant to UVA, with growth observed even after 30 min of exposure (Table 1).
Based on the Bacillus
species tested, resistance to UV A, UVA+B, or full-spectrum UV irradiation was found to be strain specific. However, for Mars full UV spectra, three of four B. pumilus
strains tested in this study exhibited LD50
s of 40 to 80 s and LD90
s of 100 to 270 s. Except for B. pumilus
SAFR-032 and B. megaterium
, spores of the Bacillus
species tested exhibited an LD50
, and LD100
of s, respectively, for Mars full UV spectra. Furthermore, B. pumilus
SAFR-032 spores that showed resistance to 2,000 J m–2
) were not completely killed after 30 min when they were exposed to the full UV spectrum under simulated Mars intensities. In addition, SAFR-032 spores showed greater resistance (LD50
of 84 s and LD90
of 270 s) than the spores of the other strains to exposure to the full UV spectrum under the simulated Mars solar UV irradiation conditions. Reference strain B. subtilis
168 exhibited LD50
, and LD100
for the full UV spectrum of 24 s, 42 s, and 72 s, respectively.
Plots of the survival rates for B. pumilus SAFR-032, B. megaterium, and B. subtilis 168 at various times for full Mars UV irradiation are shown in Fig. 3. B. pumilus SAFR-032 spores exhibited classical inactivation kinetics, with a characteristic "shoulder" extending to 2 min followed by strict exponential inactivation. However, B. subtilis 168 and B. megaterium spores exhibited a sharp decline in viability immediately after 30 s of UV exposure. After 10 min of UV exposure, the cultivability of B. megaterium spores was completely lost, while a portion of B. pumilus SAFR-032 spores survived.
Effects of various UV spectra under Mars solar UV irradiation conditions on the survival of B. pumilus
SAFR-032 spores are shown in Fig. 4
. Neither UVA nor UVA+B radiation was effective in inactivating SAFR-032 spores, even after 30 min of exposure, whereas the full simulated spectrum reduced SAFR-032 spore viability by more than 3 orders of magnitude after 30 min of irradiation. When the initial concentration of spores was reduced from 106
and to 104
before UV exposure, the spores of B. pumilus
SAFR-032 were completely killed in 8 and 2 min, respectively (data not shown).
The effects of viable and heat-killed SAFR-032 spores on the UV-sensitive B. subtilis 168 spores were tested (Fig. 5). Equal portions of viable spores of B. pumilus SAFR-032 and B. subtilis 168 were mixed to obtain a final density of approximately 106 spores ml–1 (5 x 105 spores ml–1 of each strain) and exposed to full simulated Martian UV irradiation. The time required to eradicate all B. subtilis 168 spores when they were mixed with an equal number of B. pumilus SAFR-032 spores was 17 min, compared to 1.2 and 5 min when B. subtilis 168 spores at concentrations of 5 x 105 and 1 x 106 spores ml–1, respectively, were exposed alone (Fig. 3 and Table 2). Heat-killed B. pumilus SAFR-032 spores provided no protection to B. subtilis 168 (data not shown).
Vegetative cells of B. pumilus
SAFR-032, B. subtilis
168, and Acinetobacter radioresistens
50 v1, isolated from Mars Odyssey, were exposed to UV254
at a rate of 1 J m–2
and recovered under lighted conditions (Fig. 6
). The results indicate that the three strains exhibited similar inactivation rates up to a total dose of 50 J m–2
, after which cells of B. pumilus
SAFR-032 and A. radioresistens
survived a 200-J m–2
dose (a 3-log reduction), while B. subtilis
168 cells were unrecoverable after a 50-J m–2
Survival of endospores in aqueous solutions under simulated Martian UV irradiation conditions (KSC simulation).
Since several publications have shown that there are variations in Martian UV simulations (7, 21, 46), we UV irradiated spores of B. pumilus SAFR-032 and B. pumilus 168 using an alternate Martian UV simulation system developed at KSC (32). Spores of these organisms were exposed to Martian UV, UVA, UVA+B, and full-spectrum irradiation conditions. Aliquots of UV-exposed spores were plated onto TSA after UV exposure for 0, 1, 2, 5, 10, and 20 min, and the surviving spore-forming cells were counted under appropriate cultural conditions. Figure 7A, B, and C show the inactivation curves for the spores of SAFR-032 and 168 exposed to full Mars UV, UVA+B, and UVA, respectively. The LD90 for B. subtilis 168 spores was full Martian UV irradiation, and SAFR-032 spores exhibited at least threefold-higher resistance than 168 spores exhibited (Fig. 7A). Under the total UV spectrum the cultivability of B. subtilis 168 spores was lost after approximately 90 s, while B. pumilus SAFR-032 exhibited two- to threefold-greater resistance, remaining cultivable after 5 min (Fig. 7A). With UVA+B exposure, the survival of B. pumilus SAFR-032 dramatically increased, with 90% of the spores remaining cultivable, while B. subtilis 168 was almost noncultivable after 20 min of exposure (Fig. 7B). Similar results were observed with UVA irradiance; however, B. subtilis 168 remained cultivable after 20 min (Fig. 7C). It should be noted that it was not possible to tune the KSC xenon arc lamp in real time; therefore, when filters were placed in front of the light path, a 20 to 40% decrease in UV fluence occurred (57). This decrease in fluence partially explains the enhanced resistance to the two wavelengths during the experiment. However, even at the decreased fluences B. pumilus SAFR-032 remained more resistant than B. subtilis 168.