Liposome polymerase chain reaction assay for the sub-attomolar detection of cholera toxin and botulinum neurotoxin type A

Abstract

We describe an ultrasensitive immunoassay for detecting biotoxins that uses a liposome with encapsulated DNA reporters, and ganglioside receptors embedded in the bilayer, as the detection reagent. After immobilization of the target biotoxin by a capture antibody and co-binding of the detection reagent, the liposomes are ruptured to release the reporters, which are quantified by real-time polymerase chain reaction. The new assays for cholera and botulinum toxins are several orders of magnitude more sensitive than current detection methods. A single 96-well microtiter plate can analyze approximately 20 specimens, including calibration standards and controls, with all measurements conducted in triplicate. Using pre-coated and blocked microtiter plates, and pre-prepared liposome reagents, a liposome polymerase chain reaction assay can be carried out in about 6 h.

Full text

Liposome polymerase chain reaction assay for the
sub-attomolar detection of cholera toxin and
botulinum neurotoxin type A
Jeffrey T Mason
1
, Lixin Xu
1,4
, Zong-mei Sheng
2
, Junkun He
1
& Timothy J O’Leary
3
1
Departments of Biophysics and
2
Molecular Pathology, Armed Forces Institute of Pathology, 1413 Research Boulevard, Rockville, Maryland 20850, USA.
3
Biomedical
Laboratory Research and Development Service, Veterans Health Administration, 810 Vermont Avenue NW, Washington, DC 20420, USA.
4
Present address: Division of
Monoclonal Antibodies, Center for Drug Evaluation and Research, US Food and Drug Administration, 29 Lincoln Drive, Bethesda, Maryland 20892, USA.
Correspondence should be addressed to J.T.M. (mason@afip.osd.mil).
Published online 30 November 2006; doi:10.1038/nprot.2006.331
We describe an ultrasensitive immunoassay for detecting biotoxins that uses a liposome with encapsulated DNA reporters, and
ganglioside receptors embedded in the bilayer, as the detection reagent. After immobilization of the target biotoxin by a capture
antibody and co-binding of the detection reagent, the liposomes are ruptured to release the reporters, which are quantified by
real-time polymerase chain reaction. The new assays for cholera and botulinum toxins are several orders of magnitude more sensitive
than current detection methods. A single 96-well microtiter plate can analyze B20 specimens, including calibration standards and
controls, with all measurements conducted in triplicate. Using pre-coated and blocked microtiter plates, and pre-prepared liposome
reagents, a liposome polymerase chain reaction assay can be carried out in about 6 h.
INTRODUCTION
The potential use of biological toxins as weapons of mass destruc-
tion has created an imperative to develop rapid, field-deployable
and highly sensitive assays for the detection of these agents. In
addition, assays for biological toxins have applications in such
diverse fields as microbiology, clinical diagnostics, the evaluation of
therapeutic agents, agriculture and environmental and food testing.
Biological toxins typically exhibit extreme potency. For example,
botulinum neurotoxin type A (BoNT/A), which is produced by the
anaerobic bacterium Clostridium botulinum, is about 100 billion
times more toxic than cyanide; it is the most lethal human toxin
known, with an LD
50
of approximately ng per kg
1
(ref. 1). Thus,
assays for biological toxins must be not only highly specific, but also
highly sensitive with the ability, in some applications, to detect
toxins down to the level of a few hundred molecules. The only
current assay technology capable of this level of sensitivity couples
the protein detection specificity of antibody–protein binding with
the powerful amplification capability of the polymerase chain
reaction (PCR). Immuno-PCR, first introduced by Sano et al.
2
,
uses a reporter oligonucleotide that is either covalently
2
or non-
covalently
3
attached to an antibody specific for the protein of
interest. Although these methods allow for the highly specific and
sensitive detection of protein targets, they have limitations that
have prevented their widespread use. These shortcomings include
poor sensitivity and reproducibility with complex environmental
or biological specimens, the expense and short shelf life of the
reagents, the complex synthesis necessary to fabricate the detection
reagents and the high susceptibility of the assay to contamination
with reporter oligonucleotide. Here we describe an ultrasensitive
assay for the detection of biotoxins, which we call liposome
polymerase chain reaction (LPCR)
4
, that overcomes many of the
limitations of conventional immuno-PCR. This assay uses a lipo-
some with encapsulated DNA reporters, and ganglioside receptors
embedded in the bilayer, as a detection reagent (Fig. 1). After
immobilization of the target biotoxin in a microtiter plate well by a
capture antibody and co-binding of the DNA-liposome detection
reagent, the vesicles are ruptured to release the DNA reporters,
p
uor
G
gn
i
h
s
i
lb
uP eru
t
a
N
600
2
©
natureprotocol
s
/moc.erut
a
n.w
ww//:ptth
Figure 1
|
Representation of a liposome detection reagent shown in cross
section. The dsDNA reporters (green with red bars) are encapsulated inside the
lipid bilayer of the liposome (yellow) into which monosialoganglioside G
M1
receptors (blue) have been incorporated. The liposome is shown bound to a
cholera toxin beta subunit (CTBS) pentamer, which is co-bound to a capture
antibody.
NATURE PROTOCOLS
|
VOL.1 NO.4
|
2006
|
2003
PROTOCOL

which are quantified by real-time PCR. Encapsulation of reporters
inside liposomes offers two advantages over current immuno-PCR
methods. First, B60 reporters can be encapsulated inside each
liposome increasing the sensitivity of the assay. Second, any
contaminating reporter DNA in the plate wells can be degraded
by DNase digestion immediately prior to lysis of the liposomes, as
the enzyme cannot pass through the liposomal bilayer. Limitations
of LPCR relative to the simpler, but less sensitive, enzyme-linked
immunosorbent assay (ELISA) include the need to prepare
the DNA-liposome detection reagent and the requirement of
performing real-time PCR. LPCR assays for cholera toxin beta
subunit (CTBS) and BoNT/A yield detection thresholds below 1 fg
ml
–1
, which is 2–3 orders of magnitude more sensitive than current
detection methods
5–7
. Other biotoxins that could potentially be
detected using ganglioside receptors include tetanus, pertussis,
shiga, ricin and heat-labile enterotoxin
8,9
. We are currently devel-
oping LPCR assays that employ antibodies as antigen receptors in
place of gangliosides. These assays are being used to detect addi-
tional biotoxins as well as biomarkers for cancer, prion disease,
dengue virus and human immunodeficiency virus.
MATERIALS
REAGENTS
.
Monosialoganglioside G
M1
from bovine brain (Sigma)
.
Cholera toxin beta subunit (non-toxic subunit) from Vibro cholerae (CTBS;
Sigma)
.
Ficoll, 70 kDa (Sigma)
.
DNase I from bovine pancreas (Type IV; Sigma)
.
Exonuclease III from Escherichia coli (Sigma)
.
Triton X-100, ultra grade (Sigma)
.
Bovine serum albumin, RIA grade (BSA; Sigma)
.
Sepharose CL-4B (Sigma)
.
Octyl-b-D-glucopyranoside (Sigma)
.
Trisialoganglioside G
T
1b
from bovine brain (Calbiochem)
.
1,2-Dioleoyl-sn-glycero-3-phosphocholine in chloroform (DOPC; Avanti
Polar Lipids)
!
CAUTION Chloroform is a carcinogen and can cause liver
and kidney damage. It should be handled and disposed of appropriately
(see www.osha.gov for further information).
.
Lissamine rhodamine B 1,2-dihexadecyl-sn-glycero-3-phosphoethanolamine
triethylammonium salt (rhodamine-DHPE; Invitrogen/Molecular Probes)
.
Anti-cholera toxin beta subunit mouse monoclonal antibody (anti-CTBS;
Biodesign Laboratories)
.
Botulinum neurotoxin type A from Clostridium botulinum (BoNT/A;
Metabiologics)
!
CAUTION BoNT/A is extremely toxic with a human LD
50
of
B1 ng/kg. Only trained personnel should work with this toxin, and
registration with the Center for Disease Control and Prevention (CDC) may
be required (http://www.cdc.gov). Appropriate laboratory safety procedures
should be employed. See the following website and references contained therein
for additional details (http://pathema.tigr.org/pathema/BoNT_protocols.shtml).
.
Affinity-purified polyclonal rabbit IgG antibody against BoNT/A (anti-
BoNT/A; Metabiologics)
.
Costar flat-bottom EIA and Easy Wash high-binding 96-well microtiter
plates (Fisher Scientific)
.
PCR primers (DNA Technologies)
.
AmpliTaq Gold DNA polymerase (Applied Biosystems)
.
Taqman universal PCR mastermix (Applied Biosystems)
.
PCR Taqman probe (Applied Biosystems)
.
10 PCR buffer (Invitrogen)
.
TOPO TA cloning kit with pCR2.1-TOPO T/A plasmid vector and One-Shot
E. coli (Invitrogen)
.
TRizol Plus RNA purification kit (Invitrogen)
.
SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen)
.
Hela cells (ATCC)
.
Plasmid DNA Mini-Prep kit (QIAGEN)
.
TO-PRO-1 DNA intercalating fluorescent dye (Invitrogen/Molecular Probes)
.
QIAquick PCR purification kit (QIAGEN)
.
3M sodium acetate (Sigma)
.
Glycogen, ultrapure (Sigma)
.
Ethanol, absolute (Aldrich)
REAGENTS SETUP
Coating buffer 50 mM carbonate/bicarbonate buffer (Kirkegaard & Perry),
pH 9.6
Blocking/dilution buffer 1% (w/v) BSA in PBS, pH 7.8
Wash b u ffer A 2 mM imidazol/0.02% (w/v) Tween-20 in PBS (Kirkegaard &
Perry), pH 7.4
Wash buff er B PBS, pH 7.4
Digestion buffer 10 mM CaCl
2
/10 mM MgCl
2
/20 mM HEPES (Sigma),
pH 7.8
Lysis buffer 10 mM Triton X-100 in 10 mM borate (Sigma), pH 9.0
EQUIPMENT
.
Probe-tip sonicator, Sonic Dismembrator/model 500, with 1/8’’ probe
(Fisher Scientific)
.
Dynamic light-scattering spectrometer, Nicomp/model 370 (Particle Sizing
Systems)
.
Microtiter plate washer, model ELx405 (Bio-Tek)
.
ABI PRISM genetic sequencer, model 7700 (Applied Biosystems)
.
GeneAmp 9600 for reporter amplification (Perkin-Elmer Corporation)
.
Eppendorf model 5417R bench-top centrifuge with model FA 45-24-11 rotor.
.
Optima TLX ultracentrifuge and model TLS 55 swinging bucket rotor
(Beckman Instruments)
PROCEDURE
Preparation of reporter
1| The double-stranded DNA (reporter) that is encapsulated inside the liposomes serves only as a PCR amplification surrogate
for detection and quantification of the corresponding biotoxin target of the assay. Thus, any convenient sequence can be used;
however, the sequence should be o100 bp in length to ensure the best amplification efficiency, and to maximize the number of
reporters encapsulated into the liposomes. The reporter should also be a sequence not likely to be found in the specimens to be
analyzed.
2| We use an 84-base segment derived from the human b
2
-microglobin transcript. Since the final assay does not involve a
reverse-transcriptase step, this sequence, which spans an intron, will not be found in contaminating human DNA. The following
is the complete sequence of the b
2
-microglobin reporter used in this assay
10
:5¢–TGA CTT TGT CAC AGC CCA AGA TAG TAA GTG GGA
TCG AGA CAT GTA AGC AGC ATC ATG GAG GTT TGA AGA TGC CGC ATT TGG ATT–3¢
3| Pellet HeLa cells suspended in PBS buffer, pH 7.4, by centrifugation at 300g for 10 min to produce a pellet of
B1  10
7
cells. Extract the total RNA from the cell pellet using the TRizol Plus RNA Purification kit by following the
manufacturer’s instructions.
p
uor
G
gn
i
h
s
i
lb
uP eru
t
a
N
600
2
©
natureprotocol
s
/moc.erut
a
n.w
ww//:ptth
2004
|
VOL.1 NO.4
|
2006
|
NATURE PROTOCOLS
PROTOCOL

4| Convert the RNA isolated from the cell pellet to cDNA using the SuperScript First-Strand Synthesis System for RT-PCR
(with random primers) by following the manufacturer’s instructions.
5| Prepare reporter DNA by first amplifying cDNA from b
2
-microglobin transcripts derived from the HeLa cells using b
2
M-246F
and b
2
M-330R primers (vide infra). All primer and probe designs were performed using Taqman Probe & Primer Design software
(ABI). The primer sequences are:
b
2
M246FðforwardÞ : 5
0
-TGA CTT TGT CAC AGC CCA AGA TA-3
0
b
2
M330RðreverseÞ : 5
0
-AAT CCA AAT GCG GCA TCT TC-3
0
Check for the presence of reporter DNA by agarose gel electrophoresis.
6| Clone the amplified reporter into a pCR2.1-TOPO T/A plasmid vector and use this vector to transform One-Shot E. coli using
the TOPO TA cloning kit by following the manufacturer’s instructions.
7| Extract the plasmid DNA using the Plasmid DNA Mini-Prep kit. Amplify a 328 bp DNA fragment from the above recombinant
plasmid using M13 forward and reverse primers. This is done to ensure that only the b
2
-microglobin reporter is amplified in the
final PCR step (vide infra). The M13 primer sequences are:
Forward : 5
0
-GTA AAA CGA CGG CCA G-3
0
Reverse : 5
0
-CAG GAA ACA GCT ATG AC-3
0
8| Amplify the DNA fragment using a protocol of 29 cycles as follows:
Use a PCR reaction mixture consisting of
Following PCR, confirm the presence of the 328 bp DNA fragment by agarose gel electrophoresis.
9| Generate the 85-bp reporter by amplifying the 328-bp fragment using the b
2
M-246F and b
2
M-330R primer set (15 mM
each) using the same PCR conditions as above. Following PCR, confirm the presence of the 85-bp DNA fragment by agarose
gel electrophoresis.
10| Purify the 85-bp reporter using a QIAquick PCR purification kit and then precipitate the reporter at –20 1C overnight by
adding 1/10 (v/v) of 3M sodium acetate, pH 5.2, and three volumes of absolute ethanol containing glycogen (1 ng ml
–1
)
as a carrier.
11| Centrifuge the DNA solution at 16,000g for 25 min at 23 1C. Wash the DNA pellet with 70% ethanol and dry it under
a stream of nitrogen. Confirm the purity of the reporter by agarose gel electrophoresis.
12| Dissolve the dry reporter in 500 ml of 10 mM Tris, pH 7.4. Determine the reporter concentration by measuring the DNA
solution absorbance at a wavelength of 260 nm. An absorbance of 1.0 at 260 nm corresponds to a reporter concentration of
50 mgml
–1
. The weight concentration of the reporter can be converted into molar concentration using the reporter molecular
weight, which is 54.9 kDa.
13| Finally, dilute the reporter to a concentration of 667 mgml
–1
and store at –80 1C.
’ PAUSE POINT The reporter is stable for at least 2 years at this temperature.
p
uor
G
gn
i
h
s
i
lb
uP eru
t
a
N
600
2
©
natureprotocol
s
/moc.erut
a
n.w
ww//:ptth
Cycle number Denature Anneal Extend
1–25 95 1C for 60 s 55 1C for 1 min 72 1Cfor3min
26 72 1C for 10 min
Plasmid DNA 10 ng
10 PCR Buffer 5 ml
10 mM dNTP mix 1 ml
25 mM MgCl
2
5 ml
M13 forward primer (0.1 mgml
–1
)1ml
M13 reverse primer (0.1 mgml
–1
)1ml
Nuclease-free water 35.6 ml
Taq polymerase (1 unit ml
1
)0.4ml
NATURE PROTOCOLS
|
VOL.1 NO.4
|
2006
|
2005
PROTOCOL

An alternative to preparing your own reporter is to purchase it commercially. There are numerous vendors (e.g., Integrated
DNA Technologies) that offer synthetic PAGE-purified oligonucleotides of any desired sequence and length. A second
option is to purchase an optimized set consisting of a reporter, forward and reverse primers and Taqman probe. Applied
Biosystems is one commercial source of such optimized reagents. This approach greatly simplifies the LPCR assay, but is
more expensive.
Preparation of liposome detection reagents
14| Dissolve 58 mg of DOPC, 5.8 mg of rhodamine-DHPE and either 3 mg of monosialoganglioside G
M1
or 4.2 mg of
trisialoganglioside G
T1b
in chloroform to a final volume of 2–4 ml. The molar ratio of the three components in this solution is
92.2:5.4:2.4. The rhodamine-DHPE is added to determine the concentration of the liposome solution and as a visualization aid
during purification. It does not interfere with the real-time PCR measurement.
15| Add the solution to a test tube, and remove the chloroform by incubation in a water bath (heated to 45 1C) under a stream
of nitrogen gas.
!
CAUTION Chloroform is a carcinogen and can cause liver and kidney damage. It should be handled and disposed of appropri-
ately (see http://www.osha.gov for further information). Evaporation should be carried out under a chemical fume hood.
16| Remove residual chloroform by incubation in a vacuum dryer for at least 4 h.
17| Disperse the dry lipid mixture in 1 ml of 10 mM Tris buffer, pH 7.4, to yield a total lipid concentration of 80 mM. Use a
vortex mixer set on high speed and continue until there is no lipid film remaining on the sides or bottom of the tube.
18| Prepare small unilamellar vesicles (SUVs) by sonication with a probe-tip sonicator. Use a sonication program of 10 cycles
of 4 min on/1 min off. Immerse the tube in an ice bath throughout the process to minimize sample heating.
19| Centrifuge the resulting SUVs at 1,500g in a microcentrifuge for 5 min to remove undispersed lipid and titanium from the
probe tip.
20| Combine SUVs (250 ml, 20 mmol total lipid) and reporter (150 ml, 100 mg).
21| To this mixture, add 600 ml of ethanol/calcium chloride solution (8.3 mM CaCl
2
in 16.6 mM Tris, pH 7.4, containing 79%
(v/v) ethanol). Add the solution dropwise over approximately 30 s with maximum vortex mixing.
m CRITICAL STEP The ethanolic/calcium chloride solution must be added slowly to the rapidly vortexed liposome-DNA solution
to prevent high local concentrations of calcium, which would lead to undispersed DNA-lipid aggregates.
22| Dialyze the resulting DNA-containing large unilamellar liposomes against 500 volumes of PBS, pH 7.4, for 24 h at 4 1Cwith
two changes of buffer
11
.
’ PAUSE POINT Theliposomemixturecanbestoredat41C for up to 1 week.
Purification of the liposome detection reagents
23| Mix the liposome suspension (0.2 ml) with 0.4 ml of 30% (w/v) Ficoll dissolved in PBS, pH 7.4, to give a final
concentration of 20% (w/v) Ficoll in PBS. Transfer the liposome suspension to an ultracentrifuge tube in a swinging
bucket rotor.
24| Gently layer a 1.2-ml volume of 10% (w/v) Ficoll in PBS, pH 7.4, on top of the liposome suspension. Cover the Ficoll layers
with a 0.4-ml layer of PBS, pH 7.4.
25| Centrifuge the discontinuous gradient for 30 min at 100,000g at 23 1C.
26| Collect the liposomes at the interface between the saline and 10% (w/v) Ficoll layers. Unencapsulated reporter remains in
the lowest Ficoll layer.
27| Dialyze the purified liposomes at 4 1C against 500 volumes of PBS, pH 7.4 (12 h), followed by 500 volumes of 10 mM Tris,
pH 7.8 (12 h).
28| Store the purified liposome detection reagent under nitrogen in a sealed dark vial at 4 1C.
’ PAUSE POINT The liposome detection reagent can be stored for up to 6 months with little loss of encapsulated reporter.
For an alternate liposome purification procedure see Box 1.
Preparation of blocking liposomes
29| SUVs are used as a blocking reagent in the microtiter plate assay.
p
uor
G
gn
i
h
s
i
lb
uP eru
t
a
N
600
2
©
natureprotocol
s
/moc.erut
a
n.w
ww//:ptth
2006
|
VOL.1 NO.4
|
2006
|
NATURE PROTOCOLS
PROTOCOL

30| The blocking SUVs are prepared using a lipid mixture of DOPC and rhodamine-DHPE (94.6:5.4), but no ganglioside or
reporter is added during the preparation.
31| Prepare and purify the blocking SUVs as described above (Steps 15–19).
32| Store the purified liposome detection reagent under nitrogen in a sealed dark vial at 4 1C.
’ PAUSE POINT The blocking liposomes can be stored for up to 3 months.
Characterization of the liposome detection reagents
Determination of total lipid concentration
33| Mix 25–100 ml of liposome solution in a test tube along with 1.5 ml of methanol and 20 ml of 0.1 M NaOH. Vortex the solution
and allow it to stand for 5 min.
34| Prepare a blank by substituting PBS for the liposome solution.
35| Read the absorbance of the liposome solution at 560 nm after zeroing the spectrophotometer against the blank.
36| Determine the rhodamine-DHPE concentration using an extinction coefficient of 95,000 M
–1
cm
–1
after compensating for
the dilution factor.
37| Calculate the total lipid concentration based upon the mole percent of rhodamine-DHPE in the original lipid mixture
8
.
m CRITICAL STEP Normally, the liposome detection reagents require no further characterization beyond the calculation of the total
lipid concentration. The effect on the assay of variations in the ganglioside concentration or the number of encapsulated reporters
per liposome is compensated for by determining a standard curve using known concentrations of biotoxin. However, a brief
discussion of a more complete characterization of the liposome detection reagents is provided for those desiring to develop their
own assays using different gangliosides or reporters.
Determination of liposome concentration
38| Determine the hydrodynamic diameter of the liposomes using any dynamic light scattering spectrometer that can measure
particle diameters from 10 to 1,000 nm. Follow the manufacturer’s instructions for the use of the spectrometer. If a suitable
spectrometer is not available, a diameter of 150 nm can be assumed
4
.
39| Estimate the number of lipid molecules per liposome (N
tot
) using equation (1), where d is the hydrodynamic diameter of
the liposomes as determined by dynamic light scattering. This equation assumes a bilayer thickness of 4 nm and a lipid head-
group area of 0.71 nm
2
for phosphatidylcholine. The contribution of ganglioside and rhodamine-DHPE to the average headgroup
area are ignored in this approximation.
N
tot
¼ð4:43nm
2
Þ½d
2
+ðd  8nmÞ
2
 ð1Þ
40| Estimate the concentration of liposomes in the solution (L
tot
) using equation (2):
L
tot
ðmmol ml
1
Þ¼½total lipidðmmol ml
1
Þ=N
tot
ð2Þ
Determination of reporter concentration
41| Mix 100 ml of liposome solution with 900 mlof100mMOctyl-b-
D-glucopyranoside. Vortex the mixture and incubate at
37 1C for 15 min. Prepare a corresponding blank using 100 mlofPBS,pH7.4.
42| Read the optical absorbance of the sample at 260 nm and subtract the corresponding reading for the blank.
43| The total reporter concentration (R
tot
) is calculated using equation (3):
R
tot
ðmgml
1
Þ¼A
260
0:020ðmgml
1
Þ10 ð3Þ
p
uor
G
gn
i
h
s
i
lb
uP eru
t
a
N
600
2
©
natureprotocol
s
/moc.erut
a
n.w
ww//:ptth
BOX 1
|
ALTERNATE LIPOSOME PURIFICATION PROC EDURE
An alternate purification procedure is to degrade the unencapsulated reporter with Dnase I and exonuclease III. The DNA-liposomes are then
resolved from the free nucleotides by gel permeation chromatography
18
.
To the dialyzed liposomes from Step 22, add 2,000 units of pancreatic DNase I, 300 units of exonuclease III and 5 mM MgCl
2
to the external
aqueous phase.
Incubate the reaction mixture for 3 h at 37 1C, and then stop the reaction by adding 7 mM EDTA.
Remove the nucleotides and enzymes from the DNA-liposomes by elution from a 5 ml Sepharose CL-4B column equilibrated in 10 mM Tris,
pH 7.8. Proceed from Step 28.
NATURE PROTOCOLS
|
VOL.1 NO.4
|
2006
|
2007
PROTOCOL

where A
260
is the optical absorbance of the sample, 0.020 mgml
–1
is the absorbance of a 1 mgml
–1
solution of b
2
-microglobin
reporter, and 10 is the dilution factor.
Determination of the number of reporters per liposome
44| Add 1 ml of 1 mM TO-PRO-1 in dimethylsulfoxide to a 1 ml solution of DNA-liposomes diluted 100-fold in PBS, pH 7.4.
Prepare a scattering blank by substituting 1 ml of dimethylsulfoxide for the TO PRO-1 solution
11
.
45| Measure the fluorescence emission of the liposome solution at 531 nm using an excitation wavelength of 514 nm and
5 nm slit widths. Subtract the fluorescence of the blank solution from that of the sample.
46| Add 20 ml of 100 mM Triton X-100 to both liposome solutions, vortex, and allow the solutions to incubate in capped tubes
at 37 1C for 15 min. This serves to rupture the liposomes.
47| Re-measure the fluorescence emission of the liposome solution at 531 nm using an excitation wavelength of 514 nm and
5 nm slit widths. Subtract the fluorescence of the blank solution from that of the sample and correct for the dilution of the
detergent solution.
48| Determine the percent encapsulation as the ratio of fluorescent intensity before to that after the addition of 20 mlof
100 mM Triton X-100 to rupture the liposome detection reagents.
49| Determine the concentration of encapsulated reporter by using the percent encapsulation and the total reporter concentra-
tion determined in Step 43.
50| Determine the number of reporters per liposome by dividing the concentration of encapsulated reporter by the liposome
concentration (L
tot
) determined in Step 40.
LPCR microtiter plate assay for CTBS
51| Coat each well of a 96-well EIA high-binding flat plate with 150 ml of anti-CTBS mouse monoclonal antibody (1.0 mgml
–1
)
in coating buffer.
52| Cover the microtiter plate with a plate sealer and incubate the plate at 4 1C on a plate shaker at 600 r.p.m. for 18 h.
53| Aspirate the coating buffer and wash the plate wells five times with wash buffer A using a microtiter plate washer.
54| Add 300 ml of blocking buffer to each well and incubate the plate for 2 h at 23 1C.
55| Aspirate the blocking buffer and wash the wells twice with wash buffer A.
56| Add 150 ml of serially diluted CTBS (concentration range: 10
–14
to 10
–19
M in dilution buffer) or 150 mlofdilutionbuffer
(blank) to the plate wells. Also, include a ‘no template’ control. Prepare 3–5 replicates for each antigen concentration, including
the blank and control. Incubate the plate at 23 1C for 1 h.
57| Aspirate the sample solutions and wash the wells five times with wash buffer A and twice with wash buffer B.
58| Add 150 ml of blocking liposomes (2.0 mmol ml
–1
total lipid, diluted 1:1,000 in dilution buffer) to the plate wells, and
incubate the plate at 23 1C for 1 h.
59| Aspirate the blocking liposome solution and wash the plate three times with wash buffer B.
60| Add 150 ml of the monosialoganglioside G
M1
-containing liposome detection reagent (0.8 mmol ml
–1
total lipid, diluted
1:1,000 in dilution buffer) to the wells, and incubate the microtiter plate at 23 1C for 1 h.
61| Aspirate the detection liposomes and wash the wells ten times with wash buffer B.
62| Degrade any unencapsulated DNA by adding 150 IU of pancreatic DNase I in 100 ml of digestion buffer to each plate well.
Cover with a plate sealer, and incubate the plate at 37 1C on a plate shaker with gentle shaking for 30 min (ref. 12).
63| Heat the plate at 80 1C for 10 min to inactivate the DNase I. Aspirate the enzyme solution and wash the wells five times
with wash buffer B.
p
uor
G
gn
i
h
s
i
lb
uP eru
t
a
N
600
2
©
natureprotocol
s
/moc.erut
a
n.w
ww//:ptth
2008
|
VOL.1 NO.4
|
2006
|
NATURE PROTOCOLS
PROTOCOL

64| Add 100 ml of lysis buffer to the wells, cover with a plate sealer and incubate the plate on a shaker at 600 r.p.m. for
15 min at 23 1C. The lysis buffer serves to rupture the membranes of the liposomes, which releases the encapsulated reporters.
Blocking with a nonspecific DNA to prevent loss of reporter is not required.
m CRITICAL STEP All of the above steps are critical to the success of the LPCR assay. A high concentration of DNase I is used since
DNA absorbed to the plate walls or the outer surface of the liposomes can be difficult to digest. It is acceptable to use a partially
purified grade of DNase I (non RNase free) to minimize cost.
’ PAUSE POINT The plate can remain sealed overnight prior to analysis by real-time PCR.
? TROUBLESHOOTING
LPCR microtiter plate assay for BoNT/A
65| The LPCR microtiter plate assay for BoNT/A is carried out as described above with the following modifications.
 Trisialoganglioside G
T1b
-containing liposomes are used as the detection reagent due to the high affinity of this ganglioside for BoNT/A
13
.
 Corning flat-bottom EasyWash high-binding 96-well plates are used.
 The capture antibody (anti-BoNT/A) concentration in coating buffer is 2.5 mgml
–1
.
Real-time PCR
66| Add a 2-ml aliquot from each microtiter plate well to a 50-ml PCR reaction mixture prepared from Taqman universal PCR
Mastermix, which contains Taqman buffer A; 3.5 mM MgCl
2
;200mM each of dATP, dCTP and dGTP; 400 mM dUTP; 1.25 units of
AmpliTaq Gold; and 0.5 units of AmpErase UNG.
67| Add forward and reverse primers (300 nM each) along with the probe (200 nM).
68| Set up and initiate a PCR protocol consisting of a 2-min UNG incubation step at 50 1Canda10-minAmpliTaqGold
activation step at 95 1C. Then perform 40 cycles of PCR, where each cycle consists of a 15-s denaturation step at 95 1Canda
1-min annealing/extension step at 60 1C.
69| The forward (b
2
M-246F) and reverse (b
2
M-330R) primer sequences were given in Step 5. The fluorescent probe is:
5
0
-½VIC TGA TGC TGC TTA CAT GTC TCG ATC CCA½TAMRA-3
0
Data analysis
70| For each antigen concentration in the dilution series, including the blank, calculate the mean C
t
value and the
standard deviation of the 3–5 replicate measurements from the real-time PCR analysis. The no template control should
have a C
t
value of 37 to 40. A lower value could indicate the presence of reporter contamination in the PCR reaction
mixture
14
.
71| Construct a standard curve by plotting the average C
t
values and their standard deviations for the serially diluted antigen
versus the log
10
of the antigen concentration. Perform a linear regression analysis of the data to obtain the equation for the
standard curve and the 95% confidence limits. The linear correlation coefficient should be Z 0.98. Data analysis is performed
with Origin version 7.0 or equivalent software.
72| Determine the detection threshold of the LPCR assay, which is defined as the average C
t
value of the blank minus three
times the standard deviation of the blank, as is used for immuno-PCR assays
5,15
.
73| Convert the C
t
value of an unknown sample into toxin concentration by interpolation using the standard curve and linear
regression equation determined with the serially diluted standards. The unknown sample must be within the dynamic range of
the standard curve, which is typically 5–6 orders of magnitude.
m CRITICAL STEP Always prepare the standard curve using the same sample matrix as the unknowns. For example, if the samples
to be analyzed are urine specimens, the standards should also be prepared in urine. Always include serially diluted standards
with every plate to ensure the highest possible precision in the LPCR measurements. The PCR amplification efficiency
16
(E)can
normally be determined from the slope of the standard curve as E ¼ 10
–1/slope
. If the efficiency of a PCR reaction is 100%, a log
10
increase in reporter concentration will require about 3.3 cycles, which yields a value of 2 for E. For the LPCR assays reported here,
the slopes are B1, thus E 42 indicating efficiencies greater than 100%. This arises from the fact that the slopes of the LPCR
assays are a function of the change in biotoxin concentration superimposed upon a relatively constant level of nonspecific
binding of the DNA-liposomes. For this reason, the apparent amplification efficiencies are greater than 100%, a fact that does
not in itself compromise the accuracy of the assay
16
.
p
uor
G
gn
i
h
s
i
lb
uP eru
t
a
N
600
2
©
natureprotocol
s
/moc.erut
a
n.w
ww//:ptth
NATURE PROTOCOLS
|
VOL.1 NO.4
|
2006
|
2009
PROTOCOL


TIMING
Steps 1–13: 2 d (if amplifying existing b
2
-microglobin transcript), or 2–4 weeks (if cloning is required)
Steps 14–22: 2 d
Steps 23–28: 1 d
Steps 29–32: 6 h
Steps 33–50: 2 h
Steps 51–65: 4 h
Steps 66–73: 2 h
? TROUBLESHOOTING
Troubleshooting advice can be found in Table 1.
ANTICIPATED RESULTS
For biotoxin assays performed in
deionized water, detection thresholds
down to 10–50 molecules in 150 mlof
water are typical (zeptomolar (10
–21
M)
to attomolar (10
–18
M) concentration
range). The lower concentration limit is
determined predominately by the
binding affinities of the capture
antibody and ganglioside for the
biotoxin. The affinity of gangliosides for
biotoxins can vary widely
1,9
.For
example, the detection threshold for an
assay for tetanus toxin in deionized
water using G
T1b
-containing liposomes
is B325 molecules (unpublished data)
due to the lower binding affinity of G
T1b
for tetanus toxin relative to BoNT/A.
p
uor
G
gn
i
h
s
i
lb
uP eru
t
a
N
600
2
©
natureprotocol
s
/moc.erut
a
n.w
ww//:ptth
Figure 2
|
Results of an LPCR assay of human
urine spiked with CTBS. A urine specimen was
collected from a healthy human male volunteer.
The urine specimen was spiked with cholera toxin
beta subunit (CTBS), filtered through a 0.2-mm
polycarbonate filter to remove any particulates,
and the pH of the specimen was adjusted to 7.8
using 0.1 M NaOH. Four replicate liposome
polymerase chain reaction (LPCR) measurements
were carried out for each of six serial dilutions of
CTBS in the urine specimen (concentration range:
10
–14
to 10
–19
M) plus an unspiked urine blank. A
plot of the average serial dilution cycle threshold
(C
t
) values versus the log of the number of CTBS molecules per plate well for the four replicate
measurements is shown. The solid black circles are the average serial dilution C
t
values. The solid red line
is a linear regression fit to the C
t
values, and the dashed blue lines are the 95% confidence limits. The
solid horizontal orange line denotes the average blank C
t
value. The standard deviation of the blank is
drawn at each end of this line. The solid horizontal green line that intersects the linear regression line
indicates the detection threshold of the assay. The detection threshold for this LPCR assay is 43 ± 10
molecules of CTBS (0.5 attomolar or 0.09 fg ml
–1
). The assay dynamic range is almost five orders of
magnitude. The slope of the linear regression fit of the data is –1.02 (r
2
¼ 0.998).
TABLE 1
|
Troubleshooting table.
Problem Possible reasons Solutions
Assay background high
Steps 54, 56, 58, 60 Poor quality BSA Use RIA-grade BSA
Steps 29–32, 58 Poor SUV blocking Use fresh blocking liposomes or a higher SUV concentration
Step 60 Detection reagent concentration too high Use a higher dilution of DNA-liposomes
Steps 62–63 Incomplete DNA digestion Use fresh DNase I, a higher enzyme concentration, or a longer digestion
time
Steps 51–64 Improper reagents or assay conditions Check buffer compositions, check proper operation of plate washer, increase
number of wash cycles
Steps 66–69 Contaminated PCR reaction mixture Check no-template control, if C
t
below 37 replace PCR reagents
Loss of assay sensitivity
Steps 14–28 Detection reagent too old Prepare fresh DNA-liposomes
Step 60 Detection reagent concentration too low Use a lower dilution of DNA-liposomes
Step 60 Non-optimal pH Use pH 7.8 for detection reagent binding
Step 51 Capture antibody too old or too dilute Use fresh capture antibody or decrease capture antibody dilution
Steps 14–28 Too little ganglioside in detection reagent Prepare fresh DNA-liposomes using proper lipid composition
Poor dynamic range
Step 60 Formation of confluent liposome monolayer
8
Use a lower concentration of detection reagent
Steps 66–69 Poor PCR reaction conditions Check real-time PCR instrument and protocol. Replace PCR reagents
Poor reproducibility
Step 54 Too few replicate measurements Increase the number of replicates per sample. Particularly important for low
toxin concentrations
4
Steps 49–67 Poor technique Check for proper operation of pipetters, plate washer, etc.
0
25
Cycle threshold (C
t
)
26
27
28
29
123456
Log
10
(CTBS molecules)
2010
|
VOL.1 NO.4
|
2006
|
NATURE PROTOCOLS
PROTOCOL

The selection of the capture antibody is
critical, as it cannot compete with the
ganglioside for the same epitope. For
example, attempts to create and LPCR
assay for ricin using the monoclonal
antibody 2R1 (clone CP23)
17
were
unsuccessful (unpublished data) as
both apparently compete for the same
epitope on the ricin A-chain. In general,
polyclonal antibodies are more effective
than monoclonal antibodies for use as
the capture antibody in the LPCR assays.
Biotoxin assays using environmental
or biological specimens have higher
detection thresholds due to the higher
background (DNA-liposome non-specific
binding) resulting from the more complex matrix. Detection thresholds of 50–500 molecules in 150 ml of solution [attomolar
(10
–18
M) concentration range] are typical for these more complex specimens.
Representative results obtained for LPCR assays of CTBS and BoNT/A are the following
4
(Figs. 2 and 3):
 CTBS in deionized water. The detection threshold is 10 ± 3 molecules of CTBS [17 yoctomoles (17  10
–24
) derived from a
113-zeptomolar solution (0.02 fg ml
–1
)] based upon the linear regression and 95% confidence limits derived from the sample data.
The dynamic range of the assay is almost five orders of magnitude.
 CTBS in human urine. The detection threshold is 43 ± 10 molecules of CTBS (71 yoctomoles derived from a 0.5-attomolar solution
(0.09 fg ml
–1
)). The dynamic range of the assay is almost six orders of magnitude.
 CTBS in farm runoff water. The detection threshold is 377 ± 168 molecules of CTBS [0.6 zeptomoles derived from a 4-attomolar solution
(0.75 fg ml
–1
)]. The dynamic range of the assay is almost five orders of magnitude.
 BoNT/A in deionized water. The detection threshold is 12 ± 4 molecules [20 yoctomoles derived from a 0.1 attomolar solution
(0.02 fg ml
–1
)]. The assay is linear over approximately five orders of magnitude.
ACKNOWLEDGMENTS This work was supported by Army Medical Research and
Material Command grant DAMD17-02-1-0178.
COMPETING INTERESTS STATEMENT The authors declare that they have no
competing financial interests.
Published online at http://www.natureprotocols.com
Reprints and permissions information is available online at http://npg.nature.com/
reprintsandpermissions
1. Gill, D.M. Bacterial toxins: a table of lethal amounts. Microbiol. Rev. 46, 86–94
(1982).
2. Sano, T., Smith, C.L. & Cantor, C.R. Immuno-PCR: very sensitive antigen detection
by means of specific antibody-DNA conjugates. Science 258, 120–122 (1992).
3. Hendrickson, E.R., Truby, T.M., Joerger, R.D., Majarian, W.R. & Ebersole, R.C. High
sensitivity multianalyte immunoassay using covalent DNA-labeled antibodies and
polymerase chain reaction. Nucleic Acids Res. 23, 522–529 (1995).
4. Mason, J.T., Xu, L., Sheng, Z.M. & O’Leary, T.J. A liposome-PCR assay for the
ultrasensitive detection of biological toxins. Nat. Biotechnol. 24, 555–557
(2006).
5. Chao, H.Y., Wang, Y.C., Tang, S.S. & Liu, H.W. A highly sensitive immuno-
polymerase chain reaction assay for Clostridium botulinum neurotoxin type A.
Toxicon 43, 27–34 (2004).
6. Ahn-Yoon, S., DeCory, T.R., Baeumner, A.J. & Durst, R.A. Ganglioside-liposome
immunoassay for the ultrasensitive detection of cholera toxin. Anal. Chem. 75,
2256–2261 (2003).
7. Sharma, S.K. & Whiting, R.C. Methods for detection of Clostridium botulinum toxin
in foods. J. Food Prot. 68, 1256–1263 (2005).
8. Singh, A.K., Harrison, S.H. & Schoeniger, J.S. Gangliosides as receptors for
biological toxins: development of sensitive fluoroimmunoassays using
ganglioside-bearing liposomes. Anal. Chem. 72, 6019–6024 (2000).
9. Puu, G. An approach for analysis of protein toxins based on thin films of lipid
mixtures in an optical biosensor. Anal. Chem. 73, 72–79 (2001).
10. Gu¨ssow, D. et al. The human b
2
-microglobin gene: primary structure and
definition of transcriptional unit. J. Immunol. 139, 3132–3138 (1987).
11. Bailey, A.L. & Sullivan, S.M. Efficient encapsulation of DNA plasmids in small
neutral liposomes induced by ethanol and calcium. Biochim. Biophys. Acta 1468,
239–252 (2000).
12. Berg, E.S. & Skaug, K. Liposome encapsulation of the internal control for whole
process quality assurance of nucleic acid amplification-based assays. J. Microbiol.
Methods 55, 303–309 (2003).
13. Eidels, L., Proia, R.L. & Hart, D.A. Membrane receptors for bacterial toxins.
Microbiol. Rev. 47, 596–620 (1983).
14. Bustin, S.A. & Nolan, T. Data analysis and interpretation. in A-Z of Quantitative
PCR (ed. Bustin, S.A.) 439–492 (International University Line, La Jolla, 2004).
15. McKie, A., Samuel, D., Cohen, B. & Saunders, N.A. A quantitative immuno-PCR
assay for the detection of mumps-specific IgG. J. Immunol. Methods 270,
135–141 (2002).
16. Saunders, N.A. Quantitative real-time PCR. in Real-Time PCR: An Essential Guide
(eds. Edwards, K., Logan, J. & Saunders, N.) 103–123 (Horizon Bioscience,
Norfolk, UK, 2004).
17. Huelseweh, B., Ehricht, R. & Marschall, H.-J. A simple and rapid protein array
based method for the simultaneous detection of biowarfare agents. Proteomics 6,
2972–2981 (2006).
18. Monnard, P.-A., Oberholzer, T. & Luisi, P. Entrapment of nucleic acids in
liposomes. Biochimica. et Biophysica. Acta 1329, 39–50 (1997).
p
uor
G
gn
i
h
s
i
lb
uP eru
t
a
N
600
2
©
natureprotocol
s
/moc.erut
a
n.w
ww//:ptth
Figure 3
|
Results of an LPCR assay of deionized
water spiked with BoNT/A. Deionized water (18
MO) was spiked with botulinum neurotoxin type A
(BoNT/A), filtered through a 0.2-mmpolycarbonate
filter to remove any particulates, and the pH of the
specimen was adjusted to 7.8 using 0.1 M NaOH.
Four replicate LPCR measurements were carried out
for each of six serial dilutions of BoNT/A
(concentration range: 10
–14
to 10
–19
M) plus an
un-spiked water blank. A plot of the average serial
dilution cycle threshold (C
t
) values versus the log
of the number of BoNT/A molecules per plate
well for the four replicate measurements is
shown. The symbols are as defined in Figure 2.
The detection threshold is 12 ± 4 molecules of BoNT/A (0.1 attomolar or 0.02 fg ml
–1
). The assay is
linear over approximately five orders of magnitude. The slope of the linear regression fit of the data
is –0.632 (r
2
¼ 0.998).
25.5
01234567
Log
10
(BoNT/A molecules)
26.0
Cycle threshold (Ct)
26.5
27.0
27.5
28.0
NATURE PROTOCOLS
|
VOL.1 NO.4
|
2006
|
2011
PROTOCOL