Research paper

Bacteriophage 1:2, 86-93; May/June 2011; © 2011 Landes Bioscience

Enumeration of bacteriophage particles

Comparative analysis of the traditional plaque assay
and real-time QPCR- and nanosight-based assays
Bradley Anderson,1,* Mohammed H. Rashid,2 Chandi Carter,1 Gary Pasternack,1 Chythanya Rajanna,2 Tamara Revazishvili,2
Timothy Dean,3 Andre Senecal4 and Alexander Sulakvelidze1,2
1
Intralytix Inc.; Baltimore, MD USA; 2Emerging Pathogens Institute; University of Florida; Gainesville, FL USA; 3United States Environmental Protection Agency;
Durham, NC USA; 4United States Army Natick Soldier Research Development and Engineering Center; Combat Feeding Directorate; Natick, MA USA

Key words: bacteriophage, phage, plaque assays, phage titer

Bacteriophages are increasingly being utilized and considered for various practical applications, ranging from
decontaminating foods and inanimate surfaces to human therapy; therefore, it is important to determine their
concentrations quickly and reliably. Traditional plaque assay (PA) is the current “gold standard” for quantitating phage
titers. However, it requires at least 18 h before results are obtained, and they may be significantly influenced by various
factors. Therefore, two alternative assays based on the quantitative real-time polymerase chain reaction (QPCR) and
NanoSight Limited (NS) technologies were recently proposed for enumerating phage particles. The present study
compared the three approaches’ abilities to quantitate Listeria monocytogenes-, Escherichia coli O157:H7- and Yersinia

©2011L
andesBiosc
ienc
e.
pestis-specific lytic phages quickly and reproducibly. The average coefficient of variation (CVS) of the PA method including
all three phages was 0.15. The reproducibility of the PA method decreased dramatically when multiple investigators
performed the assays, and mean differences of as much as 0.33 log were observed. The QPCR method required costly
equipment and the synthesis of phage-specific oligonucleotide primers, but it determined phage concentrations faster
(within about 4 h) and more precisely than did PA (CVS = 0.13). NS technology required costly equipment, was less precise

Donotdi
str
ibut
e.
(CVS = 0.28) than the PA and QPCR methods, and only worked when the phages were suspended in clear medium.
However, it provided results within 5 min. After the overall correlation is established with the PA method, either of the
two assays may be useful for quickly and reproducibly determining phage concentrations.

Introduction States and United Kingdom,7,8 and more trials are being planned
(http://clinicaltrials.gov/ct2/results?term=bacteriophage). One
Viruses that kill bacteria by intracellular lysis were first iden- of the key requirements for developing “phage technology” for
tified during the early part of the 20th century by Frederick practical applications is the ability to quantitate viable bacterio-
Twort and Felix d’Herelle who called them bacteriophages (i.e., phages accurately and reproducibly, so that rigorous studies of
Greek for “bacteria-eaters”) or phages.1 Since their discovery, their potencies, minimal effective doses, side effects, etc., can be
phages were used for various practical applications, including in performed.
human and veterinary medicine (reviewed in refs. 2–4). They Traditionally, three methods have been used to quantitate
also served as model microorganisms for some of the most sig- bacteriophages: (1) plaque counts on agar plates seeded with
nificant discoveries in the field of molecular biology, including the bacteria in which the phages can propagate, (2) a dilution
deciphering the genetic code and the discovery of the trans- method, where bacterial lysis is used as an indicator of phage
duction phenomenon.5 Although therapeutic applications of presence and (3) measuring the length of time required to lyse
bacteriophages were largely forgotten in the west after the wide- a standardized bacterial suspension.9 However, only the first
spread acceptance of antibiotics during the 1940s and 1950s, method (originally developed by d’Herelle in 1917) has been gen-
the emergence of antibiotic-resistant bacteria and the increasing erally useful for determining actual phage titers, and it has been
popularity of environmentally-friendly, “green” technologies has used—with minor modifications—since then in numerous labo-
rekindled interest in practical uses for phages. Indeed, several ratories throughout the world. The basis of that traditional phage
bacteriophage-based products recently were approved for food assay (PA) involves the interaction of a single lytic phage particle
safety applications and are being commercialized in the United and a permissive bacterium, which results in the host bacteri-
States and western Europe.6 Also, at least two phage prepara- um’s lysis and the release of newly formed phage progeny. When
tions were examined during recent clinical trials in the United mixtures of the phages and their specific bacterial host cells in

*Correspondence to: Bradley Anderson; Email: [email protected]

Submitted: 01/19/11; Revised: 01/19/11; Accepted: 03/14/11
DOI: 10.4161/bact.1.2.15456

86 Bacteriophage Volume 1 Issue 2

 Research paper research paper

Table 1. CVS for the PA, NS-based and QPCR-based methods

Expected titer (PFU/mL) Phage-specific CVS Overall CVS
Method Phage
1 x 10 8
1 x 10 9
1 x 10 10
Range Average Range Average
ECML-117 0.11–0.23 0.12–0.14 0.12–0.24 0.11–0.24 0.15
PA List-36 0.14–0.26 0.13–0.23 0.06–0.08 0.06–0.26 0.14 0.06–0.26 0.15
YpP-G 0.22 0.12 0.06 0.06–0.22 0.13
ECML-117 0.22 0.28 0.30 0.22–0.30 0.27
NS List-36 0.27 0.86 a
0.27 0.27 0.27 0.22–0.35 0.28
YpP-G 0.30 0.28 0.35 0.28–0.35 0.31
ECML-117 0.09 0.09 0.08 0.08–0.09 0.09
QPCR List-36 0.20 0.11 0.19 0.11–0.20 0.17 0.08–0.20 0.13
YpP-G 0.15 0.09 0.15 0.09–0.15 0.13
a
The outlier excluded from further analysis. n = 390. For the PA, ECML-117 and List-36 phages: 30 specimens per expected titer per phage = 180. For
the YpP-G phage: 10 specimens per expected titer = 30. For all other assays and phages: 10 specimens per phage per expected titer = 180. Total data
points for all specimens: 390.

molten soft agar are poured onto the surface of a base layer of in aqueous solutions. The NTA-based approach utilizes laser-
nutrient-containing agar supporting bacterial growth, the host illuminated optical microscopy for direct, real-time visualization
cells resume their growth. In areas where phages are not present, of nanoparticles in a clear liquid. The nanoparticles are detected
the bacteria grow to the stationary phase and form a confluent, as light-scattering centers moving under Brownian motion, and

©2011L
andesBiosc
ienc
e.
opaque layer or “lawn” in the soft agar overlay. However, in areas they are counted in a few seconds or a few minutes. NTA using
where phages are present, the phage progeny released from each NS technology has been used to analyze various nanoparticles in
infected bacterium will lyse neighboring bacteria and produce suspension and, more recently, at least one bacteriophage prepa-
a growing zone of liberated phages, which eventually becomes ration has been examined by that procedure (www.nanosight.
visible to the naked eye as a clear circular area or “plaque” in com/applications/biological-nanoparticles/bacteriophage). The

Donotdi
str
ibut
e.
the otherwise confluent lawn. The plaques are counted, and the
phage concentration/titer is commonly expressed as the number
of plaque-forming units (PFU)/mL of the assayed preparation.
Although the PA is currently considered the “gold standard” for
goal of the current study was to compare the speed and precision
of the classical PA, the QPCR-based approach and NTA using
NS-based technology for determining phage concentrations.

determining phage concentrations, it is not without drawbacks, Results and Discussions

including (1) poor reproducibility; e.g., a change in the salt con-
centration of the nutrient agar can change the viable titer of T2 Plaque assay (PA). PA studies were performed within a seven-day
phage preparations by more than ca. 1,000-fold, and assaying the time period. The CVS ranged from 0.11 to 0.24, 0.06 to 0.26
same phage against different bacterial host strains will often yield and 0.06 to 0.22 for ECML-117, List-36 and YpP-G, respectively.
a different titer against each host and (2) a relatively long time The mean CVS were tightly grouped: 0.15 for ECML-117, 0.14
(usually 18 to 24 h) is required to complete the assay.9,10 for List-36 and 0.13 for YpP-G. The overall range of the CVS was
Phage particles also can be enumerated by transmission elec- 0.06 to 0.26, with a mean value of 0.15 (Table 1). The CVS may
tron microscopy (TEM) after negative staining or by epifluo- be used to estimate the highest and lowest values in a 90% CI
rescence microscopy after staining with DNA fluorochromes. for any given sample (using the Z value for 90% confidence), by
However, those techniques also have limitations; e.g., epifluores- the following formula (derived from ref. 14): V H = V x (1 + CVS
cence microscopy suffers from significant background problems x Z90) and V L = V x (1 - CVS x Z90), where V = sample value, V H
and the equipment for TEM is too expensive to make that method = highest value for 90% confidence, V L = lowest value for 90%
commonly available.10 Recently, two new rapid assays were pro- confidence and Z90 = Z value for 90% confidence (Z = 1.6445).15
posed for determining phage concentrations: (1) a quantitative For example, a phage preparation with a PA-determined titer of
real-time PCR (QPCR)-based approach,11 and (2) a nanopar- ca. 1 x 108 PFU/mL and a CVS of ca. 0.15, would have a low
ticle tracking analysis (NTA)-based approach using NanoSight interval of ca. 7.5 x 107 PFU/mL and a high interval of ca. 1.3 x
Limited (NS) technology. The QPCR method is based on mea- 108 PFU/mL, with 90% confidence. To put this into further per-
suring the total fluorescence generated during each PCR cycle spective, phage specimens whose titers (determined by the same
as the phage-containing specimen is amplified; i.e., performing investigator in the same laboratory) are within that range may be
time-dependent analysis of phage DNA molecule accumulation. considered to be identical to a titer of ca. 1 x 108 PFU/mL.
The underlying idea behind the assay is that establishing the cor- The CVS was reduced as the phage concentration increased:
relation between DNA counts (as determined by the QPCR) and they averaged 0.18, 0.14 and 0.12 for phage preparations contain-
the number of viable phage particles (as determined by PA) will ing 108, 109 and 1010 PFU/mL, respectively. The data presented
enable one to use the former to determine phage concentrations in Table 1 provide insight into the expected variations and the

www.landesbioscience.com Bacteriophage 87
 ©2011L
andesBiosc
ienc
e.
Donotdi
str
ibut
e.
Figure 1. Range of possible variance in phage titers determined by the PA. The diagonal line represents perfect agreement. The numeric values before
and after the semi-colons were obtained by two different investigators.

possible range of titration errors when evaluating PA-determined the means, standard errors of the means and 95% CI (Fig.
phage titers obtained by different investigators in the same 1). We found that the (1) mean difference between titrations
laboratory. was -0.1238 log, (2) standard error of the mean difference was
The reproducibility/precision of the PA-based method among 0.1008 log, (3) upper 95% CI of the mean difference between
various laboratories was poor. For example, only one dataset titrations was 0.3303 log and (4) lower 95% CI of the mean
(for ECML-117 at an expected concentration of 1010 PFU/mL) difference between titrations was 0.0827 log. Our data suggest
was not significantly different (p > 0.05) when the phages were that titrations whose log10 transformation values are within 0.33
assayed in two different participating laboratories (Intralytix in log of one another may be considered equivalent since the dif-
Maryland and EPI UF in Florida). However, all other matched ferences are within the 95% CI of the mean differences between
means were significantly different from one another (p < 0.05). the titrations. Furthermore, if the absolute values of the differ-
Thus, in order to determine the range of possible variance in ences between the titrations were used to calculate the mean dif-
phage titers determined by PA, 10 Listeria phages (including ferences, the range was even larger at ca. 0.5 log. This analysis
List-36) were analyzed by the PA assay. The purpose of the anal- took both inter-operator and day-to-day variations into account.
yses was to determine the variance between paired mean PA val- To put the above numbers into practical perspective: if investiga-
ues obtained on separate days by different investigators during a tors in one laboratory determine the mean titer is 1 x 108 PFU/
time period when the phage preparations should be completely mL, investigators in other laboratories may expect the same
stable; i.e., when changes in their titers should not occur. We phage preparation’s titer will range from 4.7 x 107 to 2.1 x 108
chose to compare the mean values because they represent the PFU/mL (based on mean difference of 0.33 log), assuming that
working values upon which decisions are made, rather than the all participants use the same host strain and standardized PA for
individual triplicate values obtained by each titration. Twenty- their analyses.
nine paired titrations were performed, within seven days of one The possible effect of the agar overlay’s volume on phage titer
another, by at least two different investigators quantitating the enumeration was examined during additional titration stud-
10 different Listeria phages. Each titration value was subjected ies performed with each phage preparation (ca. 1010 PFU/mL),
to a log10 transformation and the calculated difference values using agar overlay volumes of 2, 3 and 4 mL. The values were
between the titrations yielded 29 different values, one for each not significantly different from one another and from the results
pair of titrations. Paired two tailed t test was used to calculate obtained with the initial 1 mL agar overlay (data not shown).

88 Bacteriophage Volume 1 Issue 2

 Table 2. Key differences and pros and cons of the PA, QPCR-based and NS-based methods
Method
Characteristics
PA QPCR-based NS-based
Needs many dilution steps requir- Requires multiple software and
Specimen pretreatment is required.
ing numerous pipetting steps. ­hardware adjustments.
Handling requirements Requires only minimal pipetting
Requires manual (visual) plaque Specimen chamber must be cleaned
steps.
recognition and counting. after each reading.
Time requirements 18 to 24 h 3 to 4 h ≤5 min
Specialized equipment not There is an upfront equipment cost There is an upfront equipment cost
Cost considerations required and there is only a mod- of $40,000 to $50,000, but a low of $40,000 to $60,000, but no addi-
erate reagent cost per result. reagent cost per result. tional reagent costs.
Coefficient of variation (CVS) 0.15 0.13 0.28
Optimal concentration range 30 to 300 PFU/mL 106 to 108 PFU/mL 107 to 109 PFU/mL
Involves direct enumeration of Reproducibility among different Reproducibility among different
Additional pros
viable phage particles. laboratories is expected to be good. laboratories is expected to be good.
Abnormal storage conditions likely
Abnormal storage conditions likely to to affect phage viability may delete-
Results are affected by numerous affect phage viability may deleteri- riously impact correlation with PA
Additional cons factors and are poorly reproduc- ously impact correlation with PA Results only are accurate when the
ible among different laboratories. Requires the synthesis of phage- phages are in clear solutions rather
specific oligonucleotide primers. than in cloudy ones; e.g., in the host
bacterium’s broth culture.

©2011L
andesBiosc
ienc
e.
Thus, within those parameters, the volume of the agar overlay
does not significantly influence the results.
QPCR-based analyses. The CVS for the QPCR-based analy-
ses ranged from 0.08 to 0.09, 0.11 to 0.20 and 0.09 to 0.15 for
Some shortcomings of the QPCR-based assay include its: (1)
requirement for expensive equipment (although real-time PCR
machines are becoming increasingly common in many laborato-
ries), (2) requirement for optimizing PCR amplification condi-

Donotdi
str
ibut
e.
ECML-117, List-36 and YpP-G, respectively (Table 1). The mean
CVS were 0.09, 0.17 and 0.13 for ECML-117, List-36 and YpP-
G, respectively (the differences were not statistically significant
[p > 0.05]). The overall range of the CVS for the phage titers
tions and ongoing calibration with known standards analyzed
concurrently with the test samples and (3) need for phage-specific
oligonucleotide primers. The latter necessitates obtaining at least
partial phage sequences and poorly designed primers may form
determined by the QPCR-based method was 0.08 to 0.20, with a self-dimers and decrease PCR efficiency.11 The advantages of the
mean value of 0.13 (Table 1). The CVS and the above-described QPCR-based assay include its (1) good precision/reproducibility
formula may be used to calculate the range of titration values which correlates well with that of the PA, (2) its requirement for
obtained by the QPCR-based method. For example, a phage only a few reagents, (3) high throughput (many phage prepara-
preparation with a QPCR method-determined titer of ca. 1 x 108 tions may be analyzed concurrently when the assay is performed
PFU/mL and a CVS of ca. 0.13 (Table 1), would have a low in 96-well microtiter plates), (4) fairly fast turn-around time of 3
interval of ca. 7.9 x 107 PFU/mL and a high interval of ca. 1.2 x to 4 h and (5) potential for greater precision than PA performed
108 PFU/mL, with 90% confidence. by investigators in different laboratories (Table 2).
The QPCR-based method, which had the lowest CVS, was NS-based analyses. The CVS of the NS-based assays ranged
the most precise of the three assays we examined. However, since from 0.22 to 0.30, 0.27 to 0.86 and 0.28 to 0.35 for ECML-117,
that assay involves amplifying phage DNA, we were concerned List-36 and YpP-G, respectively (Table 1). The mean CVS values
that it may overestimate the concentration of viable phage par- were 0.27 for ECML-117 and List-36, and 0.31 for YpP-G. One
ticles by detecting free phage DNA and/or DNA inside damaged specimen of List-36 had an abnormally high CVS of 0.86 at the
phage capsids rather than the DNA in viable phage particles. expected concentration of 109 PFU/mL. After that one sample
Although, we did observe that the phage concentrations esti- was rejected using Dixon’s Q test with 99% confidence (Q =
mated by the QPCR method were higher than those obtained 0.80), the overall range of the CVS values were calculated to be
by PA, we found that was not the case with all phages. Thus, 0.22 to 0.35, with a mean CVS of 0.28 for the NS method (Table
the correlation between the results obtained by QPCR (which 1). Thus, the CVS values and the above-described formula may
measures the DNA of both viable and nonviable phages) and PA be used to calculate the range of titration values obtained by
(which only measure viable phage particles) must be established the NS method. For example, a NS-obtained value of ca. 1 x
for each phage before using the QPCR-based assay to quanti- 108 PFU/mL and an overall CVS of 0.28 (Table 1) would have
tate the number of its viable particles in a specific preparation. a low interval of ca. 5.4 x 107 PFU/mL and a high interval of
Moreover, the correlation may need to be updated every time ca. 1.5 x 108 PFU/mL, with 90% confidence. In other words,
the production, purification and general handling practices that NS-obtained values in the range of 5.4 x 107 to 1.5 x 108 PFU/mL
might affect phage viability are changed. may be considered equivalent to 1 x 108 PFU/mL.

www.landesbioscience.com Bacteriophage 89
 Figure 2. Correlation between the results obtained by the PA and by the NS- and QPCR-based assays for determining phage titers. The symbols repre-
sent matched-pair, ten-point data set means plotted on a log scale. The diagonal lines represent perfect agreement. (A) Results obtained with the NS
method and the PA. (B) Results obtained with the QPCR method and the PA. (C) Results obtained with the NS- and QPCR-based assays.

The NS method was found to be affected by background par- titers and it is the only method among the three approaches we
ticle distributions, and many commercially available media for examined that directly enumerates viable phage particles (QPCR

©2011L
andesBiosc
ienc
e.
propagating bacteria (including BHI broth and LB broth) elicited and NS methods quantitate both viable and nonviable phages).
high levels of “background noise,” which made it very difficult (if Therefore, as previously mentioned, correlations must be estab-
not impossible) to interpret the results. Diluting phage specimens lished between the data obtained by the PA and the other two
to a concentration of ca. 1 x 107 PFU/mL, using 0.9% saline or assays before using only the QPCR and/or NS methods to deter-
deionized water, did not entirely alleviate the problem. Diluting mine viable phage concentrations. Furthermore, even after the

Donotdi
str
ibut
e.
further to reduce the background noise was not feasible because
the optimal reading of the NS system is in the range of 107 to 109
PFU/mL.
The NS- and QPCR-based assays measure both viable and
correlations are established, it may be prudent to continue to use
PA to verify the outcomes of the QPCR- and NS-based analy-
ses, at least for a period of time until the “correction factor” and
reproducibility of testing is established with absolute certainty for
nonviable phage particles simultaneously. Thus, NS and PA data the bacteriophage under investigation. The data presented below
must be obtained for each phage and its production and storage and in Figure 2 provide preliminary insight into the correlation
protocols before using the NS-based method to determine viable among the phage titers determined by the three assays.
concentrations of that phage. Some drawbacks of the NS method In all instances, R 2 values were calculated with n = 3. In gen-
include: (1) it requires fairly expensive equipment (although eral, the QPCR-based assay under-reported the number of viable
prices are likely to come down in the future), (2) it only reliably phage particles determined by the PA for List-36, over-reported
detects phages with a size larger than 40 to 50 nm (which is a the number for ECML-117 and yielded mixed results for YpP-G
minor concern since most phages used for food safety, environ- (Fig. 2A). Among the three phages examined, the strongest cor-
mental decontamination and clinical applications are larger than relation observed between the QPCR- and PA-determined means
that), (3) the phages need to be suspended in a clear medium of the titers was R 2 = 1 for YpP-G with an average difference of
(which is a significant limiting factor because it negates one of the 50%, followed by R 2 = 0.99 for List-36 with an average difference
most attractive potential applications of the technology: using it of 76% and R 2 = 0.98 for ECML-117 with an average differ-
to optimize phage production) and (4) in order for the results ence of 38%. Thus, the values determined by the QPCR method
to be reliable, the phage concentrations must be within a range were ca. 0.5-fold different (higher or lower), 0.76-fold lower and
of 107 to 109 PFU/mL. However, NS provides results within 3.9-fold higher for YpP-G, List-36 and ECML-117, respectively,
an impressive ≤5 min timeframe, which is significantly faster than those determined by the PA. Although the percent- and/or
than does PA and the QPCR method (18 to 24 h and 3 to 4 h, fold-differences (i.e., the “correction factor”) seem high, the R 2
respectively), and its performance does not require any additional values determined for all three phage preparations were excel-
reagents. Once optimized, it is likely that the NS-based method lent (ranging from a very high value of 0.98 for ECML-117, to
will be reproducible among various laboratories, with accuracy a perfect match of 1 for YpP-G), which suggests that the results
comparable to PA performed by various investigators (but signifi- obtained by the QPCR and PA methods correlate very well with
cantly faster). one another. In other words the viable phage titers obtained by
Comparison of phage titers determined by the PA and by the QPCR-based assay are essentially identical to those obtained
the QPCR- and NS-based assays. At the present time, the PA by the PA, for the same sample, when an appropriate adjust-
is considered to be the gold standard for determining phage ment is made with the above-determined “correction factor.” We

90 Bacteriophage Volume 1 Issue 2

expected (as mentioned above) that the QPCR-based assay would Myoviridae (ECML-117) and Podoviridae (YpP-G) families of
over-report the data obtained by the PA, which turned out to be double-stranded DNA-containing bacteriophages, respectively.
the case for many, but not for all, of the specimens examined. Studies of ECML-117 and List-36 were done at Intralytix and
Also, half of the readings obtained for YpP-G were over-reported the Emerging Pathogens Institute of the University of Florida in
and half-were under-reported. At the present time, it is unclear Gainesville (EPI-UF), under BSL-2 biosafety conditions required
why some readings obtained by the QPCR method were under- for their bacterial host species; i.e., E. coli O157:H7 and L. mono-
reported, but possible contamination of the phage specimens or cytogenes, respectively. Studies of YpP-G were done at the EPI-UF,
PCR reaction mixture, which interfered with PCR-mediated under BSL-3 biosafety conditions required for its host bacterium
amplification, may be one explanation. (Y. pestis). ECML-117 and List-36 phages were filter purified and
The NS-based assay over-reported the number of viable phage buffer exchanged via diafiltration into sterile PBS and 100 mM
particles determined by the PA, except when List-36 was enu- saline, respectively, while YpP-G phages were filter purified in the
merated at a concentration of 1010 PFU/mL, which was slightly original growth media.
under-reported (Fig. 2B). Among the three phages examined, the Equipment calibration, culture media and reagents. Prior to
strongest correlation between the NS- and PA-determined means use, all equipment and pipettes were inspected and calibrated.
of the titers was R 2 = 1 for YpP-G with an average difference of Reconstituted culture media and most of the reagents were pre-
544%, followed by R 2 = 0.98 for List-36 with an average dif- pared at the sites taking part in the studies. Luria-Bertani (LB)
ference of 52% and R 2 = 0.98 for ECML-117 with an average Broth and Luria-Bertani Agar were purchased as powders from
difference of 461% (Fig. 2B). In other words, the phage titers Neogen Corporation (LB broth, 7279B, LB agar, 7213B). The
determined by the NS method were ca. 0.5-fold higher, 4.6-fold soft agar overlay with which the phages and their bacterial host
higher and 5.4-fold higher for List-36, ECML-117 and YpP-G, species were mixed contained sterile isotonic saline from Fisher
respectively, than they were when the phage preparations were BioReagents (0.9% saline, 50-700-066), Tris-HCl buffer, pH
quantitated by the PA. A possible explanation for these differ- 7.5 (50 ml/L) from Teknova (T1075) and agar-agar (7 g/L) from

©2011L
andesBiosc
ienc
e.
ences is that NS enumerated viable phage particles, non-viable EMD Chemicals (1.01614). Brain-heart infusion (BHI) broth
phage particles and background noise where as PA only enumer- and agar were purchased from Becton Dickinson (bhi broth,
ated viable phage particles. The correlation between the phage 237200, bhi agar, 241810).
titers determined by the NS method and PA was very good, with Preparation of phage specimens for assaying. PA was used
an R 2 ranging from a very high value of 0.98 for List-36 and to determine the initial titers of all three bacteriophage prepara-

Donotdi
str
ibut
e.
ECML-117, to a perfect R 2 of 1 for YpP-G. Thus, as with the
QPCR-determined method, we expect that the titers obtained
by the NS-based assay adjusted by the above-determined average
“correction factor” will very accurately predict PA-determined
tions, after which they were diluted (with sterile isotonic saline)
to yield preparations with phage concentrations of 1010, 109 and
108 PFU/mL. Mixing was done with a Mini-Vortexer VM-3000
(VWR Scientific, 945300) after each dilution step. Aliquots
titers. of the diluted List-36 and ECML-117 phage preparations were
The NS method over-reported the number of viable phage par- stored at Intralytix for subsequent analysis and they were also
ticles determined by the QPCR-based assay, except when ECML- express-mailed (on wet ice) to the EPI-UF. Studies of YpP-G were
117 was enumerated at a concentration of ca. 108 PFU/mL. Among done only at the EPI-UF.
the three phages examined, the strongest correlation between the Plaque-based assays (PA). PA was performed essentially as
results obtained with the NS and QPCR methods was R 2 = 1 for described by Adams.9 Briefly, LB agar in Petri dishes were streaked
ECML-117 with an average difference of 27%, followed by R 2 = (for purity and well-isolated colonies) with thawed specimens of
0.99 for List-36 (with an average difference of 452%) and YpP-G the bacterial host strains (stored in 70% LB broth/30% glyc-
(with an average difference of 568%) (Fig. 2C). erol in a -80°C freezer) and the inoculated media were incubated
overnight at 30°C, 37°C and 28°C for the L. monocytogenes,
Materials and Methods E. coli and Y. pestis hosts, respectively. Aliquots (10 mL) of sterile
LB broth were inoculated with one colony of the appropriate host
Bacteriophages. Three different bacteriophages were used dur- strains and incubated (2 to 6 h, with shaking) until the opti-
ing our studies: (1) Listeria monocytogenes-specific bacteriophage cal density reached an OD600 of 0.2 to 0.3. The phage-host cell
List-36 (ATCC #PTA-5376), (2) Escherichia coli O157:H7- interaction mixtures consisted of 10 aliquots (1 mL each) of each
specific bacteriophage ECML-117 (ATCC #PTA-7950) and (3) diluted phage preparation (103, 102 and 101 PFU/mL) and 100
Yersinia pestis-specific bacteriophage YpP-G. List-36 and ECML- μL of the appropriate host culture in sterile, 10-mL graduated
117 are part of the commercial phage preparations ListShieldTM Falcon culture tubes. The phage-specific bacterial host-soft agar
and EcoShieldTM, respectively (www.intralytix.com). YpP-G overlays were prepared by incubating (10 min, room tempera-
is a component of a diagnostic phage preparation for detect- ture) the phage-host mixtures and gently mixing them with ali-
ing Y. pestis, which was produced in the former Soviet Union quots (3 mL) of molten (45° to 50°C) soft agar (0.7% w/v). After
and was kindly supplied by Dr. Nikoloz Tsertsvadze (Georgian the individual mixtures were quickly but very gently poured onto
National Center for Disease Control; Tbilisi, Georgia). Based on appropriate solid culture media in Petri dishes (10 cm diameter),
the classification scheme of Ackermann and Berthiaume,12 List- the dishes were rotated gently upright on a flat surface to distrib-
36, ECML-117 and YpP-G belong to the Siphoviridae (List-36), ute the mixtures evenly, after which they were incubated (room

www.landesbioscience.com Bacteriophage 91
temperature, 15 min) until the top layer solidified. Subsequently, standard range was used during the QPCR-based assays of the
the Petri dishes were inverted and incubated (18 to 20 h) at the test specimens.
appropriate temperature. Petri dishes exhibiting between 30 to NTA/NS technology-based assays. The assays were per-
300 well isolated plaques were used to calculate the phage titers formed exclusively at the EPI-UF, according to the recommended
and for subsequent statistical analyses. protocols of the required machine’s manufacturer (NanoSight,
QPCR-based assays. The assays were done at the EPI-UF. LM20). Briefly, aliquots (300 to 400 μL) of 10 replicates (1-mL)
Primer sets specific to each phage were required for the assays, of diluted phage preparations (108 PFU/mL) in sterile 10-mL cul-
and they were designed after determining full-genome sequences ture tubes were injected aseptically into the NanoSight LM20
of each phage, using the Primer3 software.13 After identifying machine’s specimen chamber until the liquid reached the nozzle’s
the required primer sequences, the primers were synthesized by tip. The specimens were tracked and measured at room tempera-
Invitrogen Corporation on a fee-for-service basis. The primer ture for 60 s with a manual shutter and the brightness, gain and
pairs for each bacteriophage were as follows: Ec3F (5'-GCG ATA blur ranged between 3 to -6, 0.5 to 2.0 and 3 x 3 to 5 x 5, respec-
GTT GCA TCC GTC TT-3') and Ec3R (5'-GCA GGA AGT tively. The data were captured and analyzed with NTA Build 127
TGT GAC CGA CT-3') for ECML-117; Lp3F (5'-GCA GAA software (version 2.0).
GCT TCG GTA CCT TTA-3') and Lp3R (5'-CAA GCC GTG Data collection and statistical analyses. The PA-based data
GTA CGG TTT AT-3') for List-36; and Yp1F (5'-CGC GGT were collected by visual examination of the plates and manual
ACT CTA GGA TGA GC-3') and Yp1R (5'-CTA TTG GGG plaque counting. Data collected during the QPCR- and NS-based
AAG GGG GTT TA-3') for YpP-G. assays were processed with Bio-Rad’s iQ5 2.0 Standard Edition
The assays were performed, with minor modifications, as Optical System software (version 2.0.148.060623) and NTA
described by Edelman and Barletta.11 Briefly, aliquots (500 μL) Analytical software (version 2.0), respectively. Duplicate QPCR
of each undiluted phage stock were incubated (37°C, 90 min, results were averaged and all data were saved to Open Office
on a heat block) with proteinase K (1 mg/50 μL), after which Calc spreadsheets. Calculations of confidence intervals (CI),

©2011L
andesBiosc
ienc
e.
the mixtures were immediately subjected to another heat treat- coefficient(s) of variation (CVS); i.e., the population’s standard
ment (70°C, 30 min). After the heat treatments, two sets of ten deviation divided by its mean,14 and power relation coefficients
replicates were prepared for each of the samples that contained of determination (R 2) between the three assaying methods were
ca. 107 PFU/mL prior to the heat treatments. The QPCR assay completed in Open Office Calc. software. Paired two-tailed t test
reaction mixtures (25 μL) consisted of 12.5 μL of iQ SYBR and one-way analysis of variance (ANOVA) calculations were

Donotdi
str
ibut
e.
Green Supermix (Bio-Rad 170-8880), forward primer (10
mM in 0.5 μL), reverse primer (10 mM in 0.5 μL), template
phage DNA preparation (2 μL) and sterile MilliQ water (9.5
μL). Amplification was performed with a iQ5 detection system
performed with the GraphPad InStat (version 3.05) program
(GraphPad Software, www.graphpad.com) and Open Office
Calc and the supplemental OooStat statistics package (version
0.5) (written by David Hitchcock and released under the GNU
Real-Time PCR Detection System (Bio-Rad, 170-9780). QPCR general public license), respectively.
cycling was done at 95°C (5 min), followed by 35 3-step cycles
at 95°C (30 s), 50°C (20 s) and 72°C (45 s), and a final cycle at Summary
72°C (5 min). Real-time data capture was performed with Bio-
Rad’s iQ5 Standard Edition Optical System software, version Bacteriophages are increasingly being used and/or considered
2.0.148.060623. for use in many practical applications. For example, at least two
Crude phage DNA (used to create standard curves) and the bacteriophage-based products are currently being marketed for
heat-treated phage preparations were analyzed simultaneously food safety applications, to reduce or eliminate contamination of
by agarose gel electrophoresis, and the latter were amplified via various food products with the foodborne bacterial pathogen L.
standard PCR. Aliquots (5 μL) of the crude phage DNA prepa- monocytogenes.6 In addition, bacteriophage preparations are being
rations and each of the heat-treated phage specimens (containing considered for use as natural, “green” decontaminating agents
ca. 1010 PFU/mL prior to heating) were analyzed by electropho- effective in reducing or eliminating contamination of various
resis in a 1% agarose gel to confirm band sizes. PCR reaction inanimate surfaces naturally or intentionally (for bioterrorist
mixtures (50 μL) consisted of 10x-concentrated XPCR buffer attacks) contaminated with pathogenic bacteria, including class
(5 μL), MgCl2 (50 mM in 2 μL), dNTPs (10 mM in 2 μL), A agents (e.g., Y. pestis); i.e., those capable of causing outbreaks
forward primer (10 mM in 1 μL), reverse primer (10 mM in 1 and epidemics with significant human morbidity and mortality.16
μL), crude phage DNA preparation (2 μL) and a Taq-Pro Red Also, lytic bacteriophages are being increasingly considered in
DNA polymerase preparation (0.5 unit in 0.5 μL; TaqPro Red the West as therapeutic modalities, for preventing and treating
Core Kit; Denville Scientific, CB4060). PCR cycling was done human diseases caused by bacteria resistant to currently avail-
at 94°C (5 min), followed by 35 3-step cycles at 94°C (45 s), able antibiotics.17 Finally, phages continue to be used as impor-
48°C (30 s) and 72°C (45 s), and a final cycle at 72°C (5 min). tant diagnostic modalities and/or research tools in many research
The PCR products were purified with a Geneclean Spin Kit laboratories around the world.18 All of those applications require
(MP Biomedical, 1101-200) and DNA was eluted with sterile phage preparations whose titers have been rigorously character-
MilliQ water (30 μL). The DNA standards were serially diluted ized by an appropriate assay. The present study was designed to
down to 101 copies/2 μL. The most relevant and encompassing compare the speed and precision of the classical PA assay with

92 Bacteriophage Volume 1 Issue 2

those of two recently developed methods: the QPCR- and NTA/ between the classical PA and either of the two new assays. After
NS-based assays. the overall correlation (including the CVS, correction factor
The pros and cons of each assay are summarized in Table 2. and R 2) is established, either of the new assays may be useful
Each approach has advantages and disadvantages. In general, for quickly and reproducibly determining phage concentrations.
however, both of the new methods appear to be viable alterna- The NS method may be considered to be superior to the QPCR-
tives to the PA for quantifying high-concentration (>106 PFU/ based assay because of the former’s very high speed and its lack
mL) phage preparations. Both are significantly faster to perform of a requirement for phage-specific primers and other expensive
than the classical PA as the QPCR-based assay and the NS pro- reagents. In addition, it enables simultaneous assessment of a
tocol yield results within 3 to 4 h and 5 min, respectively. The phage preparation’s purity and degree of aggregation, which will
faster turnaround time for the NS-based assay is counterbal- be valuable when the preparation is being quality-tested for, for
anced by the reduced precision when compared to the QPCR- example, clinical applications. Given the poor reproducibility
based assay. However, we believe the NS method’s precision may and potentially very large error range of 0.33 log of the tradi-
be increased by, for example, performing it with monodisperse tional PA assay, the availability of the two new complementary
phage specimens that are not contaminated with phage aggre- methods may be very useful for future basic and applied research
gates. The manufacturer of the equipment used to perform the with bacteriophages.
assay suggested (www.nanosight.com/faq/validity-of-method)
that a reproducibility within 2 to 3% may be achieved under Acknowledgements
ideal conditions. We are grateful to Tamar Abuladze and Leroy Voelker for their
The current study was limited because, with the exception help with the plaque assays. We thank Dr. Nikoloz Tsertsvadze
of the PA, we quantitated only three bacteriophages with the for providing us with the YpP-G phage preparation and Drs.
assays. However, the results obtained when we analyzed the three David M. Anderson and Arnold Kreger for their editorial assis-
phages and compared their titers determined by (1) the PA and tance. B.A., C.C. and A.S. hold an equity stake in Intralytix,

©2011L
andesBiosc
ienc
e.
the QPCR-based method, (2) the PA and the NS-based method Inc., a Maryland corporation developing therapeutic phage prep-
and (3) the QPCR- and NS-based assays correlated well with arations. The US Environmental Protection Agency, through
each other (Fig. 2), which suggests that a similar trend may be its Office of Research and Development, partially funded (con-
applicable to other bacteriophages. However, the CVS and cor- tract #EP-C-08-031, to A.S.) and collaborated in the research
rection factors did vary for the three different phages, which indi- described in this publication. Additional funding was provided

Donotdi
str
ibut
e.
cates that whenever a new phage is characterized, initial testing
must be performed to establish the phage’s specific relationship

References 7.
by Intralytix, Inc., and by SBIR award W911QY-07-C-0125 from
the United States Army (to A.S.).
Wright A, Hawkins CH, Anggard EE, Harper DR. A
controlled clinical trial of a therapeutic bacteriophage
13. Rozen S, Skaletsky HJ. Primer3 on the WWW for gen-
eral users and for biologist programmers. In: Krawetz
1. Duckworth DH. “Who discovered bacteriophage?”. preparation in chronic otitis due to antibiotic-resistant S, Misener S, (Eds.,) Bioinformatics Methods and
Bacteriol Rev 1976; 40:793-802. Pseudomonas aeruginosa; a preliminary report of effi- Protocols: Methods in Molecular Biology. Totowa, NJ:
2. Summers WC. Bacteriophage therapy. Annu Rev cacy. Clin Otolaryngol 2009; 34:349-57. Humana Press 2000; 365-86.
Microbiol 2001; 55:437-51. 8. Wolcott R, Rhoads D, Kuskowski M, Ward L, 14. Quinn GP, Keough MJ. Experimental design and
3. Sulakvelidze A, Barrow P. Phage therapy in animals Sulakvelidze A. Bacteriophage therapy of venous leg data analysis for biologists. Cambridge: Cambridge
and agribusiness. In: Kutter E, Sulakvelidze A, Eds. ulcers in humans: results of a Phase I safety trial. University Press 2002.
Bacteriophages: Biology and Applications. Boca Raton, Journal of Wound Care 2009; 18:237-43. 15. Mahajan BK. Methods in Biostatistics. New Delhi,
FL: CRC Press 2005; 335-80. 9. Adams MH. Enumeration of bacteriophage particles. India: Jaypee Brothers Publishers 2002.
4. Sulakvelidze A, Kutter E. Bacteriophage thera- Bacteriophages. London: Interscience Publishers, Ltd., 16. Abuladze T, Li M, Menetrez MY, Dean T, Senecal A,
py in humans. In: Kutter E, Sulakvelidze A, Eds. 1959; 27-34. Sulakvelidze A. Bacteriophages reduce experimental
Bacteriophages: Biology and Application. Boca Raton, 10. Carlson K. Working with bacteriophages: common contamination of hard surfaces, tomato, spinach, broc-
FL: CRC Press 2005; 381-436. techniques and methodological approaches. In: Kutter coli and ground beef by Escherichia coli O157:H7. Appl
5. Cairns J, Stent GS, Watson JD. Phage and the origins E, Sulakvelidze A, Eds. Bacteriophages: Biology and Environ Microbiol 2008; 74:6230-8.
of molecular biology. Cold Spring Harbor, NY: Cold Application. Boca Raton, FL: CRC Press 2004; 437-94. 17. Thiel K. Old dogma, new tricks—21st Century phage
Spring Harbor Laboratory 1966. 11. Edelman DC, Barletta J. Real-time PCR provides therapy. Nat Biotechnol 2004; 22:31-6.
6. Sulakvelidze A, Pasternack G. Industrial and regulatory improved detection and titer determination of bacte- 18. Griffiths MW. Phage-based methods for detection of
issues in bacteriophage applications in food production riophage. Biotechniques 2003; 35:368-75. bacterial pathogens. In: Sabour PM, Griffiths MW,
and processing. In: Sabour PM, Griffiths MW, Eds. 12. Ackermann HW, Berthiaume L. Atlas of virus dia- Eds. Bacteriophages in the control of food- and water-
Bacteriophages in the control of food- and waterborne grams. Boca Raton, FL: CRC Press 1995. borne pathogens. Washington, DC: ASM Press 2010;
pathogens. Washington, DC: ASM Press 2010; 297-326. 31-59.

www.landesbioscience.com Bacteriophage 93