Journal of Medical Microbiology (2006), 55, 1271–1275 DOI 10.1099/jmm.0.

46488-0

Effect of gamma irradiation on viability and DNA of

Staphylococcus epidermidis and Escherichia coli
Andrej Trampuz,1 Kerryl E. Piper,1 James M. Steckelberg1
and Robin Patel1,2
Correspondence Division of Infectious Diseases, Department of Internal Medicine1 and Division of Clinical
Robin Patel Microbiology, Department of Laboratory Medicine and Pathology2, Mayo Clinic College of
patel.robin@mayo.edu Medicine, Rochester, MN 55905, USA

Gamma irradiation is widely used for sterilization; however, its effect on elimination of
amplifiable DNA, an issue of relevance to molecular diagnostic approaches, has not been well
studied. The effect of gamma irradiation on the viability of Staphylococcus epidermidis and
Escherichia coli (using quantitative cultures) and on their DNA (using quantitative 16S rRNA gene
PCR) was evaluated. Viability was abrogated at 2?8 and 3?6 kGy for S. epidermidis and E. coli,
respectively. The radiation dose required to reduce viable bacteria by one log10 (D10 value) was
0?31 and 0?35 kGy for S. epidermidis and E. coli, respectively. D10 values for amplifiable DNA
extracted from bacteria were 2?58 and 3?09 kGy for S. epidermidis and E. coli, respectively,
whereas D10 values for amplifiable DNA were significantly higher for DNA extracted from irradiated
viable bacterial cells (22?9 and 52?6 kGy for S. epidermidis and E. coli, respectively; P<0?001).
This study showed that gamma irradiation of DNA in viable bacterial cells has little effect on
amplifiable DNA, was not able to eliminate amplifiable 16S rRNA genes at a dose of up to 12 kGy
and cannot therefore be used for elimination of DNA contamination of PCR reaction components or
Received 25 December 2005 laboratory equipment when this DNA is present in microbial cells. This finding has practical
Accepted 8 June 2006 implications for those using molecular diagnostic techniques in microbiology.

INTRODUCTION PCR, is unknown. DNA may fail to amplify due to DNA

degradation, such as alteration in primer binding sites or
Gamma irradiation is electromagnetic radiation of short reduction of the DNA into fragments smaller than the target.
wavelength emitted by radioactive isotopes as the unstable If gamma irradiation effectively eliminates amplifiable DNA,
nucleus breaks up and decays to reach a stable form. It is it could be used widely in laboratory and clinical practice for
widely used for sterilization of medical devices, food prevention of DNA contamination of PCR reaction reagents,
preservation and processing of tissue allografts and blood laboratory equipment, surgical instruments and containers
components, obviating the need for high temperatures that for specimen collection and transportation.
can be damaging to such products (Block, 2001; Hansen &
Shaffer, 2001; Kainer et al., 2004; Mendonca et al., 2004; We therefore studied the effect of gamma irradiation on the
Osterholm & Norgan, 2004). DNA is the principal cellular viability of Staphylococcus epidermidis and Escherichia coli
target governing loss of viability after exposure to gamma (using quantitative cultures) and on their DNA (using
irradiation. DNA damage occurs predominantly by the quantitative PCR amplification of the 16S rRNA gene). The
indirect action of gamma rays, which interact with other 16S rRNA gene was selected because this highly conserved
atoms or molecules, particularly water, to produce reactive region of bacterial DNA is often used when the infecting
free radicals. Cell death (defined for proliferating cells as loss agent is not known and the goal is to detect and identify the
of reproductive capability) is predominantly induced by presence of any bacterium (Kolbert et al., 2004). The 16S
double-strand breaks in DNA, separated by not more than a rRNA gene is present as multiple copies in the genomes
few base pairs, which can generally not be repaired by the cell of most bacterial species that belong to the eubacterial
(Hall & Giaccia, 2006). kingdom, but is not present in human, viral or fungal
genomes. The presence of multiple copies of this target in
Although several studies have investigated the effect of bacteria increases assay sensitivity when applied to infected
gamma irradiation on the viability of micro-organisms, little human specimens. However, this target has been associated
information is available regarding its effect on microbial with false-positive results as a result of 16S rRNA gene
DNA. In particular, whether gamma irradiation eliminates contamination of reagents or equipment used for molecular
amplifiable DNA, detectable using quantitative broad-range approaches. We also evaluated differences in radiation

46488 G 2006 SGM Printed in Great Britain 1271

A. Trampuz and others

sensitivity of extracted DNA in comparison with DNA re- during irradiation to minimize variations in the absorbed dose.
siding within viable bacterial cells at the time of irradiation. Normal saline, processed in the same way as bacterial cells, served as
a negative control. A self-contained 137Cs gamma irradiation cell
irradiator (Mark I) was used. The source strength was ~6000 Ci
METHODS (2?261014 Bq) with a dose rate of 9?35 Gy min21, as established by
the National Institutes of Standards and Technology. Actual
Bacterial cultures. Stock cultures of S. epidermidis ATCC 12228 absorbed doses were within 3 % of target doses as assessed by dosi-
and E. coli ATCC 10798 were frozen in Microbank cryovials (Pro- metric measurement using 5 mm diameter alanine dosimeters
lab Diagnostics) and stored at 270 uC until studied. One cryovial (Bruker Biospin).
bead from each stock culture was streaked on trypticase soy agar
containing 5 % sheep blood (BD Diagnostic Systems) and incubated Assessment of irradiation effect. Bacterial suspensions were
for 24 h. An isolated colony was removed aseptically from the agar exposed to radiation doses of 0–4 kGy, in increments of 0?2 kGy
plate and inoculated into 150 ml sterile trypticase soy broth. After (Fig. 1a). After irradiation, serial dilutions were prepared in normal
incubation for 18 h on a rotary shaker at 150 r.p.m. at 37 uC, the saline, plated on trypticase soy agar containing 5 % sheep blood and
broth was centrifuged (5000 g for 10 min) and the pellet was resus- incubated at 37 uC for 48 h. Viable cells were expressed as mean
pended in 150 ml normal saline to keep the bacteria viable, but log10[c.f.u. (ml suspension)21] ± SD of triplicates. The gamma irra-
minimize their replication. One millilitre aliquots of bacterial sus- diation effect on DNA was studied at doses of 0–12 kGy, in incre-
pensions were assayed after exposure to gamma irradiation in three ments of 1 kGy, using either DNA extracted from bacterial
ways (Fig. 1). First, viability of bacteria was evaluated after irradia- suspensions before irradiation (Fig. 1b) or DNA extracted from irra-
tion of bacterial suspensions (Fig. 1a). Second, the effect of gamma diated bacterial suspensions (Fig. 1c). Tubes with extracted DNA
irradiation was studied on amplifiable free DNA extracted from bac- were stored at 21±2 uC for a maximum of 12 h until irradiation.
terial cells before irradiation (Fig. 1b). Third, the effect of gamma DNA quantity was determined by quantitative PCR.
irradiation was evaluated on amplifiable DNA where viable cells
were irradiated first and then the DNA was extracted (Fig. 1c). Quantitative 16S rRNA gene PCR. Real-time PCR (LightCycler)
DNA extraction. DNA was extracted using a QIAamp DNA mini was used to quantify the 16S rRNA gene. Universal primers (forward
kit (Qiagen). One millilitre aliquots of bacterial suspension were primer: 59-TGGAGAGTTTGATCCTGGCTCAG-39; reverse primer:
centrifuged (5000 g for 10 min) and the pellet was resuspended in 59-TACCGCGGCTGCTGGCAC-39) spanning positions 5–532 (inclu-
180 ml buffer ATL. Twenty microlitres of proteinase K was added sive) of E. coli K-12 (GenBank accession no. NC_000913) were used
and the mixture was vortexed and incubated at 56 uC for 60 min. (Kolbert et al., 2004; Tang et al., 1998, 2000). Each PCR mix con-
After incubation, 200 ml buffer AL was added and the mixture was sisted of 2 ml target DNA added to 18 ml mastermix (LightCycler
incubated at 70 uC for 10 min before 200 ml absolute ethanol FastStart DNA Master SYBR Green I; Roche Applied Science), con-
(Aaper) was added and the mixture transferred to a QIAamp spin taining final concentrations of 2?5 mM MgCl2, 0?04 mM each primer
column that was centrifuged at 6000 g for 1 min. Five hundred micro- and 0?05 U thermolabile uracil-N-glycosylase (Roche Applied
litres buffer AW1 was then added to the column and the sample was Science). Cycling parameters consisted of one cycle at 95 uC for
centrifuged at 6000 g for 1 min. Five hundred microlitres buffer AW2 10 min (pre-incubation), followed by 45 cycles of denaturation at
was then added to the column and the sample was centrifuged at 95 uC for 15 s, annealing at 62 uC for 5 s and elongation at 72 uC for
20 000 g for 3 min. After centrifugation, 1 ml distilled water was 20 s. These PCR conditions were optimized to produce the least
added. The sample was incubated at room temperature for 5 min and non-specific signal by primer dimers, as evaluated by post-amplifica-
then centrifuged at 6000 g for 1 min to elute the DNA. tion melting curve analysis. Mastermix on its own was used as a
negative control for PCR. For quantification, the second derivative
Gamma irradiation. One millilitre aliquots of bacterial suspension maximum method with Savitzky–Golay polynomial estimation was
and extracted DNA were irradiated in triplicate in closed 1?5 ml used. The standard curve was determined by depicting the amplifica-
polyethylene microcentrifuge tubes at 21±2 uC, with rotation tion threshold cycle number (crossing point) against the logarithm

Fig. 1. Study design for evaluation of the

effect of gamma irradiation on bacterial via-
bility (a), free (extracted) DNA (b) and DNA
in viable bacterial cells subsequently sub-
jected to DNA extraction (c). Samples for
quantification of bacterial cells (a) were irra-
diated with 0–4 kGy in increments of
0?2 kGy and those for quantification of DNA
(b, c) with 0–12 kGy in increments of
1 kGy.

1272 Journal of Medical Microbiology 55

 Effect of gamma irradiation on bacterial cells and DNA

of the initial target concentration (Shepley & Wolk, 2004; Wittwer & E. coli D10 values in our study are comparable to those
Kusukawa, 2004). Standard curves for S. epidermidis and E. coli were reported by others, ranging from 0?20 to 0?65 kGy (Block,
generated from five serial dilutions of known quantities of S. epider-
2001; Osterholm & Norgan, 2004). The radiation suscept-
midis [limit of detection, 150 c.f.u. (ml bacterial suspension)21; coeffi-
cient of determination (r2), 0?97; amplification efficiency (E), 1?78] ibility of cells is known to be affected by a number of
and E. coli [limit of detection, 25 c.f.u. (ml bacterial suspension)21; factors, including replication rate, intracellular water con-
r2, 0?95; E, 1?82). Amplification efficiency was calculated according tent, amount of DNA, medium composition, temperature,
to the formula E=1021/k, where k represents the slope of the quan- pH, oxygenation status and the ability to repair radiation-
tification standard curve. DNA quantity was expressed as c.f.u. induced DNA damage (Thayer & Boyd, 1993, 2001; Thayer
equivalent (ml bacterial suspension)21. Random amplification pro-
ducts after irradiation were sequenced in both 59–39 and 39–59 direc-
et al., 2003). To our knowledge, D10 values have not been
tions with BigDye terminator version 1.1 Taq kit and an ABI reported previously for S. epidermidis, as most studies have
3730XL DNA sequencer (Applied Biosystems), using the above uni- focused on micro-organisms important in the food industry
versal PCR primers as sequencing primers. Sequence data were ana-
lysed using MicroSeq software and GenBank. The strain of S.
epidermidis used has five copies of 16S rRNA genes and the E. coli
strain has seven copies.

Radiation dose–response curves and D10 values. Responses to

gamma irradiation were expressed as the logarithm of the ratio of
survivors (N/N0), where N represents the mean c.f.u. ml21 or c.f.u.
equivalent ml21 of irradiated bacterial suspension or DNA, as
appropriate, and N0 the mean number of c.f.u. ml21 or c.f.u.
equivalent ml21 of non-irradiated control. The log10N/N0 (outcome
variable, y) was plotted against the corresponding radiation dose
(explanatory variable, x) to obtain the semi-logarithmic dose–
response curve. D10 values, defined as the radiation dose (in kGy)
required to reduce the number of c.f.u. ml21 or c.f.u. equivalent
ml21 by one log10, were determined by calculating the negative reci-
procal of the slope of the linear regression curve (Aziz et al., 1997;
Bari et al., 2003; Lamb et al., 2002; Rajkowski et al., 2003; Sommers
& Fan, 2003; Thayer & Boyd, 1993, 2001; Thayer et al., 2003).

Statistical analysis. Variables in the dose–response curve were

fitted using a simple linear regression model, as determined by least-
squares analysis (Woodward, 1999). The zero radiation value was
excluded from the linear regression analysis to avoid a possible
shoulder effect. The analysis was limited to the linear portion of the
curve and r2 values were calculated. The 95 % confidence intervals
(CIs) for the regression curve were weighted by standard deviations
of triplicate samples. Regressions were tested for differences by ana-
lysis of covariance (Woodward, 1999). SD and 95 % CI were calcu-
lated for D10 values. A P value of <0?05 (for a 2-sided test) was
considered statistically significant. All calculations were performed
using the statistical software package JMP (version 6.0; SAS Ins-
titute). Origin software (version 7.5; OriginLab) was used for gra-
phic analysis.

RESULTS AND DISCUSSION

Irradiation effect on viability of bacterial cells

The effect of gamma irradiation on the viability of stationary-
phase cells of S. epidermidis and E. coli is shown in Fig. 2(a)
and (b), respectively. Non-irradiated bacterial suspensions
contained a mean±SD of 8?78±0?12 log10(c.f.u. ml21) for Fig. 2. Effect of gamma irradiation on S. epidermidis (a) and E.
S. epidermidis or 9?47±0?07 log10(c.f.u. ml21) for E. coli. coli (b). This is shown as the effect on bacterial viability after irra-
Bacterial viability was abrogated at 2?8 kGy for S. epidermidis diation of bacterial suspensions (determined by subsequent quan-
and 3?6 kGy for E. coli. D10 values for bacterial cells were titative cultures; &), on amplifiable free DNA extracted from
0?31 kGy for S. epidermidis and 0?35 kGy for E. coli (P>0?1) bacterial cells before irradiation (determined by subsequent quan-
(Table 1). titative PCR; m) and on amplifiable DNA where viable cells were
first irradiated and then the DNA was extracted and subjected to
Gamma irradiation at 4 kGy sterilized stationary-phase quantitative PCR (*). Dotted lines represent the 95 % CIs for the
populations of both S. epidermidis and E. coli. The calculated estimates of each regression.

http://jmm.sgmjournals.org 1273
A. Trampuz and others

Table 1. Radiation D10 values for bacterial cells and DNA

Irradiation procedure D10 [kGy (95 % CI)]* r2

S. epidermidis
Irradiated bacteria 0?31 (0?29–0?34)a 0?985
Irradiated extracted DNA 2?58 (2?37–2?83)b 0?984
Irradiated bacteria, followed by DNA extraction 22?9 (18?9–28?9)b 0?920
E. coli
Irradiated bacteria 0?35 (0?32–0?38)a 0?981
Irradiated extracted DNA 3?09 (2?78–3?46)b 0?976
Irradiated bacteria, followed by DNA extraction 52?6 (34?5–111?1)b 0?638

*Determined from the slope of the simple linear regression analysis: a, from 0?2 to 4 kGy; b, from 1 to
12 kGy.

(Block, 2001). However, S. epidermidis is the predominant highest radiation dose tested (12 kGy). Potential reasons for
pathogen causing device-associated infections and is clini- the radiation resistance of DNA in viable cells are manifold.
cally important in the implant industry (Zimmerli et al., DNA in viable cells may be more resistant to irradiation than
2004). free (extracted) DNA because of low molecular mass
scavengers that mop up free radicals in cells, physical
protection of DNA by packaging in cells and/or cellular
Irradiation effect on amplifiable DNA
repair of damaged DNA (Hall & Giaccia, 2006). In dying
Table 1 shows radiation D10 values for DNA extracted from cells, DNA fragmentation may also occur because of the
bacteria before irradiation and for DNA extracted from action of nucleases. Less likely, irradiated bacteria may be
irradiated bacterial cells. Fig. 2 shows that gamma irradia- more easily lysed than non-irradiated bacteria; conse-
tion at 4, 8 and 12 kGy reduced the free amplifiable DNA quently, larger amounts of extracted DNA would be
quantity (extracted before irradiation) by 1?20±0?06, available for PCR. However, it is unlikely that relatively
2?65±0?02 and 4?44±0?03 log10(c.f.u. equivalent ml21) small differences in DNA extraction efficiency in irradiated
for S. epidermidis, respectively, and by 0?55±0?05, and non-irradiated cells could explain the significant
1?80±0?08 and 3?60±0?04 log10(c.f.u. equivalent ml21) differences in D10 values of cell-associated and free DNA.
for E. coli, respectively. D10 values for extracted DNA were DNA extraction is less efficient for Gram-positive bacteria
lower for S. epidermidis than for E. coli (2?58 versus than for Gram-negative bacteria. Failure to extract the DNA
3?09 kGy, P=0?02). from S. epidermidis may make it appear easier to eliminate.
In contrast, irradiation of DNA in viable bacterial cells, Importantly, the amplification assay used in this study
which were subsequently subjected to extraction, had less quantified amplifiable DNA using universal primers
effect on amplifiable DNA than did irradiation of extracted annealing to 16S rRNA genes present as multiple copies
DNA (P<0?001). Even at the highest radiation dose tested in the genomes. Whether or not the use of a specific rather
(12 kGy), a reduction in the quantity of amplifiable DNA in than broad-range PCR assay, targeting a single copy gene,
irradiated viable bacterial cells corresponding to just would yield different results is unknown. However, broad-
0?43±0?05 log10(c.f.u. S. epidermidis ml21) or 0?10±0?06 range PCR is commonly used in diagnostic microbiology
log10(c.f.u. E. coli ml21) was achieved (Fig. 2). D10 values for and was therefore chosen for study. Different results may
DNA extracted from irradiated viable bacterial cells were arise with different sizes of target; for example, a shorter
22?9 and 52?6 kGy for S. epidermidis and E. coli, respectively. partial 16S rRNA gene target may have yielded greater
The DNA quantity after amplification of normal saline residual amplifiable DNA.
without bacteria (negative control) was below the detection
limit. Sequence data of 15 randomly chosen amplification The results of our study indicate that gamma irradiation
products with positive signals confirmed the specific target cannot be used for elimination of DNA contamination of
with >99 % identity. PCR reaction components, surgical instruments or labora-
tory equipment, when this DNA is present in microbial cells.
This subject is important in clinical practice as molecular
Comparison of effects on viability and
amplification techniques are increasingly deployed in
amplifiable DNA
microbiological diagnostics due to their high sensitivity,
We have demonstrated that gamma irradiation of viable rapidity and ability to detect organisms that are not growing
bacterial cells has a smaller effect on amplifiable 16S rRNA because of prior antimicrobial therapy or are not culturable
genes than does irradiation of extracted DNA. Importantly, on conventional growth media. Possible strategies to
gamma irradiation did not eliminate amplifiable DNA at the enhance elimination of DNA residing in viable cells by

1274 Journal of Medical Microbiology 55

 Effect of gamma irradiation on bacterial cells and DNA

gamma irradiation include inactivation of cellular repair Kolbert, C. P., Rys, P. N., Hopkins, M., Lynch, D. T., Germer, J. J.,
mechanisms using low temperatures for irradiation, expo- O’Sullivan, C. E., Trampuz, A. & Patel, R. (2004). 16S ribosomal
sure to high temperatures before irradiation or DNA DNA sequence analysis for identification of bacteria in a clinical
microbiology laboratory. In Molecular Microbiology: Diagnostic
extraction before irradiation. Alternatively, other methods
Principles and Practice, pp. 361–378. Edited by D. H. Persing, F. C.
for DNA elimination, such as chemical (e.g. bleach) or Tenover, J. Versalovic, Y.-W. Tang, E. R. Unger, D. A. Relman & T. J.
enzymic (e.g. nuclease) treatment, might be considered. White. Washington, DC: American Society for Microbiology.
Finally, radiation resistance of DNA in microbial cells may Lamb, J. L., Gogley, J. M., Thompson, M. J., Solis, D. R. & Sen, S.
be beneficial for diagnostic purposes if the goal is to reduce (2002). Effect of low-dose gamma irradiation on Staphylococcus
the infectivity of the specimen while preserving microbial aureus and product packaging in ready-to-eat ham and cheese
DNA as a target for molecular diagnostics. This strategy has sandwiches. J Food Prot 65, 1800–1805.
been validated for herpes viruses and Bacillus anthracis using Mendonca, A. F., Romero, M. G., Lihono, M. A., Nannapaneni, R. &
autoclaving (Espy et al., 2002), but has not yet been Johnson, M. G. (2004). Radiation resistance and virulence of Listeria
described with gamma irradiation. monocytogenes Scott A following starvation in physiological saline.
J Food Prot 67, 470–474.
In summary, our observations have important implications Osterholm, M. T. & Norgan, A. P. (2004). The role of irradiation in
for those using molecular techniques in diagnostic micro- food safety. N Engl J Med 350, 1898–1901.
biology. The inability of gamma irradiation to eliminate Rajkowski, K. T., Boyd, G. & Thayer, D. W. (2003). Irradiation D-
microbial DNA in viable cells needs to be taken into account values for Escherichia coli O157 : H7 and Salmonella sp. on inoculated
when using irradiated specimens. broccoli seeds and effects of irradiation on broccoli sprout keeping
quality and seed viability. J Food Prot 66, 760–766.
Shepley, D. P. & Wolk, D. M. (2004). Quantitative molecular
ACKNOWLEDGEMENTS methods: result standardization, interpretation and laboratory
quality control. In Molecular Microbiology: Diagnostic Principles
The authors would like to thank Jann N. Sarkaria for useful suggestions and Practice, pp. 95–129. Edited by D. H. Persing, F. C. Tenover,
and review of the manuscript. This work was supported by the Mayo J. Versalovic Y.-W. Tang, E. R. Unger, D. A. Relman & T. J. White.
Foundation and Roche Research Foundation. Presented in part at
Washington, DC: American Society for Microbiology.
the 44th Interscience Conference on Antimicrobial Agents and
Chemotherapy, Washington, DC, USA, October 30–November 2, 2004. Sommers, C. & Fan, X. (2003). Gamma irradiation of fine-emulsion
sausage containing sodium diacetate. J Food Prot 66, 819–824.
Tang, Y.-W., Ellis, N. M., Hopkins, M. K., Smith, D. H., Dodge, D. E. &
REFERENCES Persing, D. H. (1998). Comparison of phenotypic and genotypic
techniques for identification of unusual aerobic pathogenic gram-
Aziz, N. H., El-Fouly, M. Z., Abu-Shady, M. R. & Moussa, L. A. A. negative bacilli. J Clin Microbiol 36, 3674–3679.
(1997). Effect of gamma radiation on the survival of fungal and
Tang, Y.-W., Von Graevenitz, A., Waddington, M. G., Hopkins, M. K.,
actinomycetal florae contaminating medicinal plants. Appl Radiat
Smith, D. H., Li, H., Kolbert, C. P., Montgomery, S. O. & Persing,
Isot 48, 71–76.
D. H. (2000). Identification of coryneform bacterial isolates by
Bari, M. L., Nazuka, E., Sabina, Y., Todoriki, S. & Isshiki, K. (2003). ribosomal DNA sequence analysis. J Clin Microbiol 38, 1676–1678.
Chemical and irradiation treatments for killing Escherichia coli
Thayer, D. W. & Boyd, G. (1993). Elimination of Escherichia coli
O157 : H7 on alfalfa, radish, and mung bean seeds. J Food Prot 66,
O157 : H7 in meats by gamma irradiation. Appl Environ Microbiol 59,
767–774.
1030–1034.
Block, S. S. (2001). Disinfection, Sterilization and Preservation, 5th
Thayer, D. W. & Boyd, G. (2001). Effect of irradiation temperature on
edn. Philadelphia, PA: Lippincott Williams & Wilkins.
inactivation of Escherichia coli O157 : H7 and Staphylococcus aureus.
Espy, M. J., Uhl, J. R., Sloan, L. M., Rosenblatt, J. E., Cockerill, F. R., J Food Prot 64, 1624–1626.
III & Smith, T. F. (2002). Detection of vaccinia virus, herpes simplex
virus, varicella-zoster virus, and Bacillus anthracis DNA by Thayer, D. W., Rajkowski, K. T., Boyd, G., Cooke, P. H. & Soroka,
LightCycler polymerase chain reaction after autoclaving: implications D. S. (2003). Inactivation of Escherichia coli O157 : H7 and
for biosafety of bioterrorism agents. Mayo Clin Proc 77, 624–628. Salmonella by gamma irradiation of alfalfa seed intended for
production of food sprouts. J Food Prot 66, 175–181.
Hall, E. J. & Giaccia, A. J. (2006). Radiobiology for the Radiologist, 6th
edn. Philadelphia, PA: Lippincott Williams & Wilkins. Wittwer, C. & Kusukawa, N. (2004). Real-time PCR. In Molecular
Microbiology: Diagnostic Principles and Practice, pp. 71–84. Edited by
Hansen, J. M. & Shaffer, H. L. (2001). Sterilization and preservation
D. Persing, F. Tenover, J. Versalovic, Y.-W. Tang, E. R. Unger, D. A.
by radiation sterilization. In Disinfection, Sterilization and Pre-
Relman & T. J. White. Washington, DC: American Society for
servation, pp. 729–746. Edited by S. S. Block. Philadelphia, PA:
Microbiology.
Lippincott Williams & Wilkins.
Woodward, M. (1999). Epidemiology: Study Design and Data Analysis,
Kainer, M. A., Linden, J. V., Whaley, D. N., Holmes, H. T., Jarvis, W. R.,
1st edn. Boca Raton, FL: CRC Press.
Jernigan, D. B. & Archibald, L. K. (2004). Clostridium infections
associated with musculoskeletal-tissue allografts. N Engl J Med 350, Zimmerli, W., Trampuz, A. & Ochsner, P. E. (2004). Prosthetic-joint
2564–2571. infections. N Engl J Med 351, 1645–1654.

http://jmm.sgmjournals.org 1275