See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/319986158

Immuno-detection of cleaved SNAP-25 from differentiated mouse embryonic

stem cells provides a sensitive assay for determination of botulinum A toxin
and antitoxin potency

Article  in  Journal of Immunological Methods · September 2017

DOI: 10.1016/j.jim.2017.09.007

CITATIONS READS

14 311

5 authors, including:

Shalini P Rajagopal Dorothea Sesardic

Medicines and Healthcare products Regulatory Agency (MHRA) National Institute for Biological Standards and Control
9 PUBLICATIONS   169 CITATIONS    179 PUBLICATIONS   5,254 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

EU AntiBOTABE View project

Replacing animals for botulinum testing View project

All content following this page was uploaded by Shalini P Rajagopal on 29 September 2017.

The user has requested enhancement of the downloaded file.

 Journal of Immunological Methods xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Journal of Immunological Methods

journal homepage: www.elsevier.com/locate/jim

Research paper

Immuno-detection of cleaved SNAP-25 from differentiated mouse

embryonic stem cells provides a sensitive assay for determination of
botulinum A toxin and antitoxin potency
G. Yadirgi, P. Stickings, S. Rajagopal, Y. Liu, D. Sesardic⁎
Division of Bacteriology, National Institute for Biological Standards and Control, a center of the Medicines and Healthcare Products Regulatory Agency, South Mimms,
Potters Bar, Hertfordshire EN6 3QG, UK

A R T I C L E I N F O A B S T R A C T

Keywords: Botulinum toxin type A is a causative agent of human botulism. Due to high toxicity and ease of production it is
BoNT/A classified by the Centres for Disease Control and Prevention as a category A bioterrorism agent. The same
Mouse embryonic stem cells serotype, BoNT/A, is also the most widely used in pharmaceutical preparations for treatment of a diverse range
Cell-based assay of neuromuscular disorders. Traditionally, animals are used to confirm the presence and activity of toxin and to
SNAP-25
establish neutralizing capabilities of countermeasures in toxin neutralization tests. Cell based assays for BoNT/A
Capture ELISA
have been reported as the most viable alternative to animal models, since they are capable of reflecting all key
In vitro potency
Toxin neutralization steps (binding, translocation, internalization and cleavage of intracellular substrate) involved in toxin activity. In
this paper we report preliminary development of a simple immunochemical method for specifically detecting
BoNT/A cleaved intracellular substrate, SNAP-25, in cell lysates of neurons derived from mouse embryonic stem
cells. The assay offers sensitivity of better than 0.1 LD50/ml (3 fM) which is not matched by other functional
assays, including the mouse bioassay, and provides serotype specificity for quantitative detection of BoNT/A and
anti-BoNT/A antitoxin. Subject to formal validation, the method described here could potentially be used as a
substitute for the mouse bioassay to measure potency and consistency of therapeutic products.

1. Introduction considerable ethical and safety limitations and alternative non-animal

Botulinum neurotoxins (BoNT/s) are considered the most lethal
toxins (Gill, 1982). They induce prolonged muscle paralysis and block
muscle function through inhibition of neuronal transmitters from per-
ipheral cholinergic nerve endings (Rossetto, et al., 2014). Although
they are highly toxic and the causative agents of botulism, BoNTs have
been used for over twenty years in many medical interventions to treat
muscle hyperactivity, such as dystonia and spastic conditions, exocrine
gland hyperactivity, such as hyperhidrosis, and pain disorders
(Dressler, 2012). A number of licensed products are manufactured
globally and for many years the in vivo mouse bioassay has been ap-
plied to determine toxin activity per vial, monitor lot to lot variability
and to define shelf life. Animals are still used for assessment of toxin
potency by many manufacturers, for clinical diagnosis, and for potency
testing of antitoxins (Sesardic, 2012). The mouse bioassays have
methods have been sought as replacements for a number of years
(Sesardic and Gaines Das, 2008; NIH Publication, 2008; Sesardic, 2012;
Adler et al., 2010).
To date, seven serologically distinct botulinum toxin serotypes
(A–G) have been identified. They all comprise a 150 kDa single-chain
progenitor toxin which is subsequently activated by protease to gen-
erate a disulphide bond-linked structure containing a 50 kDa light chain
(LC) and a 100 kDa heavy chain (HC). The HC contains two functional
domains that are both required for toxin binding and uptake into the
nerve cells which involves a dual-receptor mechanism involving gang-
liosides, such as GT1b, and a protein receptor. Once the toxin is in-
ternalized, the inter-chain disulphide bond is broken by reducing con-
ditions within the endocytic vesicles. This activates and releases the
endopeptidase light chain (LC) which specifically cleaves one of three
proteins involved in neurotransmitter release (Pirazzini et al., 2017).

Abbreviations: BoNT/A, botulinum neurotoxin A; SNAP-25, synaptosomal protein of molecular mass 25 kDa; EB, embryoid bodies; ESDN, embryonic stem cell-derived neurons; mESC,
mouse embryonic stem cells; HC, heavy chain; LC, light chain; LHn/A, recombinant endopeptidase fragment of BoNT/A LC and the N-terminus of the HC; ELISA, enzyme linked
immunosorbent assay; WB, Western blotting; PBS, phosphate buffered saline; CM, culture medium; RA, retinoic acid; LIF, leukemia inhibitory factor; GMEM, Glasgow's Minimal Essential
Medium; Shh, Sonic Hedgehog; FD, final differentiation; NGF-β, nerve growth factor beta; BDNF, brain-derived neurotrophic factor; RT, room temperature
⁎
Corresponding author.
E-mail addresses: gokhanyadirgi@yahoo.co.uk (G. Yadirgi), paul.stickings@nibsc.org (P. Stickings), shalini.rajagopal@nibsc.org (S. Rajagopal), Yvonne.Liu@nibsc.org (Y. Liu),
thea.sesardic@nibsc.org (D. Sesardic).

http://dx.doi.org/10.1016/j.jim.2017.09.007
Received 12 July 2017; Received in revised form 19 September 2017; Accepted 20 September 2017
0022-1759/ © 2017 Published by Elsevier B.V.

Please cite this article as: Yadirgi, G., Journal of Immunological Methods (2017), http://dx.doi.org/10.1016/j.jim.2017.09.007
G. Yadirgi et al.
The endopeptidase activity of BoNT LC is highly specific so that each
toxin has a unique substrate/target bond combination. Whereas the LCs
of botulinum toxin serotypes A, E and C1 all cleave SNAP-25 (sy-
naptosomal associated protein of molecular mass 25 kDa), they target
this protein at distinct sites (Q197-R198, R180-I181, and R198-A199,
respectively) (Schiavo et al., 1993; Vadakkanchery et al., 1999;
Rossetto et al., 2014; Pirazzini et al., 2017).
Once the molecular mechanism of BoNT/s was unravelled in the
early 1990′s, it became possible to develop immunoassays designed to
detect the endopeptidase activity of the BoNT LCs. This approach re-
quired production of a recombinant or synthetic substrate (SNAP-25 for
BoNT/A and BoNT/E or VAMP2 for BoNT/B) and a specific neoepitope
antibody, recognising only a peptide sequence that becomes exposed
upon toxin-mediated cleavage of the substrate. Taking this approach,
serotype specific assays have been developed for several BoNT/s (Ekong
et al., 1997; Jones et al., 2008; Jones et al., 2009; Sesardic et al., 2004;
Simon et al., 2015). However, endopeptidase assays can suffer from
unwanted matrix interference from albumin and human serum (Jones
et al., 2011; Simon et al., 2015), cannot detect changes to the binding
and translocation domains of the toxin, and can detect free LC activity
in the absence of a functional binding domain, thus potentially giving
false positive results (Sesardic, 2012). Furthermore, these methods are
not considered to be indicators of stability. While these endopeptidase
assays are suitable for monitoring lot-to-lot consistency following va-
lidation for a given toxin product, they cannot be considered as a full
replacement for the mouse LD50 potency test (NIH Publication, 2008,
Adler et al., 2010).
To circumvent the limitations of the first generation of in vitro en-
dopeptidase assays, second generation methods were developed that
require both functional binding to toxin receptor or antibody and cat-
alytic endopeptidase activity for detection of BoNT/A and BoNT/B
(Evans et al., 2009; Liu et al., 2012; Rosen et al., 2016). However, the
concept of biochemical assays which mimic only two steps in the bo-
tulinum intoxication process is also considered limited as these assays
do not reflect all the events required for toxin activity in vivo, such as
the internalization step and, particularly, the persistence of toxin within
the cell, which are known to contribute to toxin potency (Keller and
Neale, 2001).
Cell based bioassays for BoNT/s are superior to other in vitro models
because they are capable of reflecting all major steps of botulinum toxin
action (Pellet, 2013) and they are therefore considered to offer the most
realistic strategy towards complete replacement of the mouse LD50 test
(NIH Publication, 2008; Adler et al., 2010). Primary cells have been
used for BoNT research for many years, and the initial studies for cell
based methods focused on neurons derived from rat (Keller et al.,
2004), mouse (Dong et al., 2007) and chicken embryos (Nuss et al.,
2010). However, these studies did not offer sensitivity comparable to
the mouse bioassay since only nM rather than low pM concentrations of
BoNT/A could be detected. The first highly sensitive cell based assay for
detection of BoNT/A was developed using primary rat embryonic spinal
cord cells. Comparable sensitivity to the mouse bioassays was reported
using detection of cleaved SNAP-25 quantified by Western blotting
(Pellet et al., 2010). The approach to use primary rat spinal cord cells
for detection and quantification of botulinum neurotoxin antibodies
was also reported as an opportunity to replace the mouse toxin neu-
tralization test (Hall et al., 2004). Still, it was recognised that primary
cells do not provide an easily standardized and robust model that would
be preferred for routine laboratory use, and they also require animals
for the supply of cells.
Immortalized cell lines can offer the desired standardization prop-
erties, but they are typically at least 1000-fold less sensitive for BoNT/A
compared to the mouse bioassay (Purkiss et al., 2001; Yowler et al.,
2002, Pellet, 2013), even after lengthy differentiation protocols and
addition of GT1b to improve sensitivity (Rasetti-Escargueil, et al.,
2011). The notable exception is the cancer derived SiMa cell line
(human neuroblastoma cell line) which, with an EC50 of ~1 pM
2
Journal of Immunological Methods xxx (xxxx) xxx–xxx
(~ 35 LD50/ml), offers a sufficiently sensitive and robust method for
potency testing of BoNT/A in Botox® (Fernandez-Salas et al., 2012).
The availability of neurons derived from pluripotent stem cells has
provided a new platform for highly sensitive detection of botulinum
neurotoxins (Kiris et al., 2011; McNutt et al., 2011, Pellet et al., 2011;
Whitemarsh et al., 2012; Beske et al., 2015; Jenkinson et al., 2017). In
addition to the source of cells, development of cell based assays is
highly dependent on the use of a relevant and quantitative endpoint to
measure the botulinum toxin activity. Measurement of neuro-
transmitter release has been applied in assay development studies
(Keller and Neale, 2001; Rasetti-Escargueil et al., 2011; Hall et al.,
2004) and, while it is an important endpoint and research tool, it is
considered less useful for routine quality control purposes because it is
relatively nonspecific and highly dependent on cell type (Pellet, 2013).
Most recently, blockade of synaptic transmission in a network culture of
neurons derived from human and rodent stem cells has been described
as a functionally relevant readout for BoNT intoxication (Beske et al.,
2015, 2016; Jenkinson et al., 2017). However, this approach relies on
the use of specialized whole cell electrophysiology equipment or multi-
electrode arrays which is not applicable in a routine control laboratory
setting and does not offer high throughput capability. Western blotting
(Pellett et al., 2007, 2015, 2017, Whitemarsh et al., 2012; McNutt et al.,
2011), quantitative immunofluorescence (Dong et al., 2004; Gilmore
et al., 2011; Kiris et al., 2011) or ELISA (Nuss et al. 2010; Pellett et al.,
2017) are methods that are all based on detection of intracellular
substrates after cleavage by BoNT/s which is considered to be the most
specific endpoint and can be applied to any neuronal cell type.
In this study, we report preliminary development of a simple cap-
ture ELISA method for detection of BoNT/A activity in neurons derived
from mouse embryonic stem cells. Our results suggest that the assay is
likely to be suitable for testing the activity of therapeutic and com-
mercial grade BoNT/A toxins and could also be used to measure the
potency of antitoxin to BoNT/A. The method described here is more
sensitive than the mouse bioassay for detection of BoNT/A and other
cell based assays that have been developed for this toxin serotype.
2. Materials and methods
2.1. Toxins
Purified type BoNT/A haemagglutinin (HA) free toxin purchased
from Metabiologics Inc., (Madison, USA, @ 2.3 × 10 (Dong et al.,
2007) LD50/mg, 1 mg/ml from Hall strain) was used. A working stock
was prepared at 20,000 LD50/ml (87 ng/ml) in Gelatine (0.2% w/v)
Phosphate (50 mM di-sodium hydrogen orthophosphate) Buffer (GPB,
pH 6.5) as previously described (Jones et al., 2008). All stocks of ali-
quoted toxins were stored at −80 °C prior to use.
Two formulations of therapeutic BoNT/A toxin for injection, from
the same manufacturer, containing different concentrations of BoNT/A
as active component were also used. Recombinant LHn/A fragment,
with an LD50 of 0.3 mg/mouse, was a generous gift from Syntaxin Ltd,
and was used as reported previously (Liu et al., 2012). All BoNT/A used
in this study were sub-type A1. Purified BoNT/B1 and BoNT/E3 were
purchased from Metabiologics (USA) and working stocks of 20,000 and
12,560 mouse LD50/ml, respectively were prepared in GPB buffer and
stored at − 80 °C before use.
2.2. Antibodies
BoNT/A cleavage site-specific polyclonal anti-peptide antibody
against SNAP-25190–197 was made in rabbits as previously described
(Ekong et al., 1997; Jones et al., 2008) and affinity purified against the
immunizing peptide. The same antibody was also custom made by Bio
Trend (Chemikalin GmbH, Köln, Germany) and purchased affinity
purified to capture BoNT/A cleaved SNAP-25 from cell lysates.
Two separate polyclonal detection antibodies (sheep anti-SNAP-
G. Yadirgi et al.
251–57 and SNAP25111–157) were raised as previously reported (Ekong
et al., 1997) and were used to detect captured SNAP-25 in order to
increase the signal and improve the sensitivity of the ELISA. Optimal
results were obtained by mixing two antibodies together at equivalent
concentrations.
Botulinum antitoxin equine type A (NIBSC product code 59/021),
with assigned potency of 2000 IU/ampoule, and botulinum antitoxin
equine type E (NIBSC product code 02/318), with assigned potency of
197 IU/ampoule, were used in the toxin neutralization assays (Jones
et al., 2006).
2.3. Maintenance and neuronal differentiation of mouse embryonic stem
(mES) cells
The mES cell line used in this study was E14tg2a (provided by Dr.
Jennifer Nichols of Cambridge University, Welcome Trust Centre for
Stem Cell Research). These feeder-independent pluripotent cells were
maintained at 37 °C in a 5% (v/v) CO2 incubator in 0.2% gelatin-coated
flasks (Nunc) in culture medium (CM) made up of Glasgow's Minimal
Essential Medium (GMEM, Invitrogen) supplemented with 10% ESC-
qualified fetal bovine serum (Invitrogen), 10 ng/ml Leukemia in-
hibitory factor (LIF, Invitrogen) and 0.1 mM 2-mercaptoethanol.
Passage 18 of mES cells (1 × 10) (Bigalke et al., 1985) were cultured in
a T25 flask (Nunc) for 3 days until 90–95% confluence was reached. To
form embryoid bodies (EBs), cells were washed with pre-warmed PBS
prior to trypsinization [2 ml trypsin-EDTA/PBS (1:1)] for 60 s. Re-
sulting clumps of 100 to 200 cells (primed to form small and large sized
EBs) were transferred into bacterial grade Petri dishes (Nunc) in 12 ml
of CM without LIF, for 24 h (Day 1).
EBs were treated with 1 μM retinoic acid (RA, Sigma) for two con-
secutive days in fresh CM without LIF (Day 2 and 3). On Day 4, Sonic
Hedgehog (Shh) 100 ng/ml, R & D Systems) and purmorphamine (1 μM,
Calbiochem) was added to cultures for 24 h. On Day 5 EBs were
transferred to new dishes containing CM without LIF, supplemented
with glial cell-derived neurotrophic factor (GDNF, 100 ng/ml, R & D
Systems) and brain-derived neurotrophic factor (BDNF, 100 ng/ml,
R & D Systems), for 60 h.
After 9–12 days, EBs were dissociated into single cell cultures by
washing with PBS and then incubating for 5 min with trypsin-EDTA/
PBS (1:1) at 37 °C in a 5% (v/v) CO2 incubator. Dissociation into single
cells was encouraged using mechanical force by pipetting. For the final
differentiation (FD) into embryonic stem cell derived neurons (ESDNs),
single cells were seeded at a density of 4–5 × 104 cells/well on laminin
coated surfaces (1 μg/well, Invitrogen) in 96-well plates (Nunc). For
this FD stage, cells were maintained in differentiation medium (DM)
consisting of: Neurobasal medium (Invitrogen) supplemented with 1×
B27 (Invitrogen), 0.5 mM L-glutamine and nerve growth factor beta
(NGF-β, 20 ng/ml, Sigma). Media was changed every 2–3 days for up to
21 days (DIV21). DIV indicates number of days that cells are cultured as
a monolayer in the final differentiation medium.
2.4. Immunocytochemistry and cell imaging
Cells at the FD stage were plated on laminin coated (20 μg/ml)
16 mm glass coverslips (Fisher Scientific), and differentiated for 3, 10
or 21 days in vitro (DIV3, 10 or 21 respectively). ESDNs were fixed for
10 min in 4% paraformaldehyde, permeabilized with either Tween-20
(0.2% v/v) or TritonX-100 (0.2% v/v) in PBS for 10 min and blocked
with 5% BSA for 1 h at room temperature (RT). Primary antibodies
were from Abcam unless otherwise stated, and were diluted in 1% (v/v)
BSA in PBS and applied overnight at 4 °C: rabbit anti-Map2 (1:250;
ab32454); mouse anti-Tau (1:250; ab80579); mouse anti-VAMP-1/2/3
(1:100; Santa Cruz Biotechnology sc-133,129); rabbit anti-SV2A R-300
(1:1000; Santa Cruz Biotechnology sc-28,955); mouse anti-synapto-
tagmin I/II (1:500; ab13259); mouse anti-choline acetyltransferase
(1:250; ab68779) and, mouse anti-noradrenaline (1:250; ab8887).
3
Journal of Immunological Methods xxx (xxxx) xxx–xxx
Cells were washed three times with PBS followed by application of
the appropriate Alexa-conjugated secondary antibody (Invitrogen) - all
used at 1:500 dilution in 1% (v/v) BSA in PBS, and incubated in the
dark for 2 h at RT. Application of secondary antibody in the absence of
primary antibody was used as a negative control. Cells were counter-
stained with DAPI (1 μg/ml, Sigma) and coverslips were mounted with
Fluoroshield (Sigma). Cells were visualised using a fluorescent micro-
scope (Olympus IX73).
2.5. Treatment of cells
For the ELISA based detection assay, all treatments were performed
on 96-well plates (Nunc), with ESDNs seeded at a density of
4–5 × 104 cells per well and differentiated between 9 and 12 days
(DIV9 to DIV12). Initial studies included treatment of cells from DIV3.
Toxin dose finding and time course studies involved addition of a range
of concentrations of BoNT/A (0.08 to 40 LD50/ml) to DIV9 differ-
entiated cells in 200 μl volumes of cell culture medium and incubating
tissue culture plates for different lengths of time (12, 24, 48 or 72 h) at
37 °C.
In all other studies, ESDNs were used at DIV9 were used unless
otherwise indicated. Purified BoNT/A, BoNT/B or BoNT/E, or com-
plexed BoNT/A, from therapeutic preparations diluted in DM, were
added in 200 μl volumes (range 0.002 to 800 LD50/ml) and cells were
incubated for 48 h at 37 °C. For specificity studies, inactivated BoNT/A
or recombinant LHn/A (at molar excess) were used. Controls included
wells with cells but no toxin or treatment.
A BoNT/A toxin neutralization study was performed to determine
the concentration of botulinum type A reference antitoxin needed to
prevent intracellular cleavage of SNAP-25 induced by a fixed dose
(20 LD50/ml, 4 LD50/dose) of BoNT/A. Toxin was pre-incubated for
1 h at RT with an equal volume of botulinum type A or E reference
antitoxin (ranging between 0.1 IU/ml and 0.03 mIU/ml) in a separate
96-well tissue culture plate. All dilutions were performed in DM and
200 μl of toxin/antitoxin mixture was added to ESDNs at DIV12 and
tissue culture plates incubated for 48 h at 37 °C. Controls in this study
included cells exposed to antitoxin alone with no toxin treatment.
For Western blotting studies, cells were plated into laminin-coated 6
well plates at 1 × 106 cells per well and treated with BoNT/A toxin
(diluted in DM, ranging from 0.2–125 LD50/ml) for 48 h.
2.6. Preparation of cell lysate
At the end of the treatment, cells were carefully washed once for
5 min with PBS. Total protein from cells was harvested by lysing with
110 μl per well (for 96 well plates) or 400 μl per well (for 6 well plates)
of freshly M-PER lysis buffer (Mammalian Protein Extraction Reagent,
Pierce 87786) containing EDTA-free protease inhibitor (Pierce) and
150 mM NaCl. Cell lysates were stored at − 20 °C until use in ELISA or
Western blotting.
2.7. Immuno-detection of cleaved SNAP-25 by ELISA
A schematic diagram showing the general principle of the capture
ELISA for immuno-detection of BoNT/A cleaved SNAP-25 is shown in
Fig. 1. Polystyrene Maxisorp 96-well plates (Nunc) were coated with
50 μl/well of affinity purified capture antibody against SNAP-25190–197
diluted to 5 μg/ml in 50 mM carbonate buffer (pH 9.6). Sealed plates
were incubated overnight at 4 °C. The coating antibody was decanted
and plates were blocked for 1 h at RT with 250 μl per well of 2.5% (w/
v) of skimmed milk powder (Marvel) in PBS, containing 0.05% v/v
Tween-20 (M-PBST). Plates were then washed 3 times with PBST,
which was the same for all subsequent wash steps. Cell lysates (50 μl
per well) were added together with an equal amount of M-PBST and
plates were shaken at 400 rpm for 10 min, and then incubated without
shaking for 90 min at RT. M-PBST (100 μl per well) was added to buffer
G. Yadirgi et al. Journal of Immunological Methods xxx (xxxx) xxx–xxx

Fig. 1. Schematic overview of the capture ELISA for immuno-detection of BoNT/A cleaved SNAP-25. BoNT/A cleaves SNAP-25 (pink) between amino acids 197 and 198 in embryonic
stem cell derived neurons, and the cleavage product is captured using a specific neoepitope antibody raised against a peptide corresponding to amino acids 190–197 of SNAP-25 (SNAP-
25190–197). This antibody only recognises the BoNT/A-cleaved SNAP-25 and does not recognise intact SNAP-25. The captured cleavage product is then detected using two polyclonal
detection antibodies that bind to two distinct sites, SNAP-251–57 and SNAP-25111–157. (For interpretation of the references to colour in this figure legend, the reader is referred to the web
version of this article.)

control wells. After washing, an equal concentration of polyclonal
sheep antibodies raised against SNAP-25 epitopes (1–57, 111–157),
diluted 1/1000 from 10 mg/ml stock in M-PBST was added to plates
(100 μl per well), which were then incubated for 90 min at RT. Plates
were washed and 100 μl per well of rabbit anti-sheep-HRP conjugate
(Thermo 31480) diluted 1:16,000 in M-PBST was added. Plates were
incubated for 90 min at RT followed by a final wash and addition of
100 μl/well of substrate solution [50 mM citric acid pH 4.0, 0.05% w/v
ABTS (2,2′-Azino-bis(3-thylbenzothiazoline-6-sulfonic acid) diammo-
nium salt (A9941, Sigma), and 0.05% v/v of 30% Hydrogen Peroxide
Solution (Merck)]. Colour was allowed to develop for 30 min at RT
prior to reading absorbance at 405 nm using a Multiskan plate reader.
2.8. Immuno-detection of cleaved SNAP-25 by Western blotting
165 V for 60 min and then transferred onto 0.2 μm nitrocellulose
membranes at 10 V for 60 min using Bolt transfer buffer, both using a
Novex Bolt system (Invitrogen).
Presence of total or truncated SNAP-25 in cells before and after
treatment with BoNT/A was performed by incubating membranes with
commercially, or in house, sourced antibodies to SNAP-25 (monoclonal
mouse antibody from Synaptic systems #111 111, rabbit anti-SNAP-25
from abcam #ab109105 or rabbit antibody to SNAP-25190–197) diluted
in PBS with 1% skimmed milk powder (Marvel). Specific detection was
visualised by addition of either goat anti-mouse IgG (Sigma, A9044) or
goat anti-rabbit IgG (Sigma, A0545) conjugated to HRP, followed by
detection using ECL2 chemiluminescence reagent (Pierce 80,196) ac-
cording to the manufacturer's instructions. Gels were developed and
capture images recorded using Gel Logic 2200 Pro Imaging Systems.

For Western blotting studies, 20 μg of total protein from cell lysates,
diluted in Tris-Tricine sample buffer containing 2% β-mercaptoethanol,
was separated on a 10–12% Tris-Tricine gel (Bio-Rad 456–3113) with
Tris-Tricine SDS buffer under denaturing conditions. Gels were trans-
ferred onto 0.2 μm nitrocellulose membrane (Bio-Rad 162–0212) using
a wet transfer system, blocked with PBS containing 5% skimmed milk
powder and exposed to primary antibody. Alternatively, cell lysates
were diluted with 4 × LDS sample buffer, 10 × reducing agent (both
from Invitrogen) and ultrapure water, to 7.5 μg total protein, denatured
at 70 °C for 10 min, then loaded onto 12% Bis-Tris Plus gels
(Invitrogen). Molecular weight markers were a 1:1 mixture of Novex
Sharp pre-stained protein standard and MagicMark XP western protein
standard (Invitrogen). Gels were run with Bolt MES running buffer at
3. Results
3.1. Differentiation and characterization of mouse ES cells
Differentiated cells were fully characterized prior to assay devel-
opment by immunocytochemistry as shown in Fig. 2. By day 10 of
differentiation (DIV10), presence of complex axodendritic arbours and
networks of cells are observed appearing together with positive im-
munostaining for specific markers of neuronal differentiation, axonal
marker Tau and dendritic marker MAP-2 (Fig. 2a). At DIV21, im-
munostaining confirmed presence of synaptotagamin I/III and SV2A,
indicating presence of botulinum toxin specific receptors (Fig. 2b). At
DIV21, immunostaining confirmed the presence of choline

4
G. Yadirgi et al. Journal of Immunological Methods xxx (xxxx) xxx–xxx

Fig. 2. Characterization of neurons derived from mouse ESDNs by immunostaining and Western Blotting. (a) Cells at DIV10 show positive immunostaining for specific markers of
neuronal differentiation: Tau (red) and MAP-2 (green). (b) Cells at DIV21 cell show positive immunostaining for synaptotagmin I/III (green), SV2A (red) and Tau (grey). Cells at DIV21
have positive immunostaining for neurotransmitters: c) choline acetyltransferase (red) and Tau (green) and d) noradrenaline (red) and Tau (green). Cells at DIV21 also had positive
immunostaining for VAMP 1/2/3 (green), SV2A (red) and Tau (grey). Scale bar is 50 μm, and nuclear counterstaining is shown in blue. (f) Western blotting for presence of SNAP-25 in
differentiated cells at DIV12 (lane 2) DIV13 (lane 3), DIV14 (lane 4), DIV19 (lane 5) and DIV21 (lane 6). Lanes 1 and 7 are molecular weight markers. (For interpretation of the references
to colour in this figure legend, the reader is referred to the web version of this article.)

acetyltransferase and noradrenaline neurotransmitters (Fig. 2c and d SNAP-25 cleavage product in samples exposed to BoNT/A at con-
respectively), and the presence of VAMP-1/2/3 (Fig. 2e). SNAP-25 ex- centrations of 125, 25 and 5 LD50/ml but not in the control sample,
pression in differentiated cells was confirmed by Western blotting using with a very faint band also visible in the lysate from samples exposed to
antibody detecting only intact SNAP-25 (Fig. 2f). 1 LD50/ml (Fig. 3b).

3.2. Western blotting
Western blotting was also used to confirm cleavage of SNAP-25 in
lysates from ESDNs that had been treated with a range of BoNT/A
concentrations (125, 25, 5.0, 1.0 and 0.2 LD50/ml). Immunodetection
was performed using two different antibodies: a mouse monoclonal
antibody recognising both cleaved and un-cleaved SNAP-25 (Fig. 3a)
and a rabbit polyclonal antibody specific for BoNT/A cleavage site only
(Fig. 3b). The mouse monoclonal antibody was able to detect intact
SNAP-25 in all samples and there was a visibly reduced intensity of the
band corresponding to intact SNAP-25 at the highest concentration of
BoNT/A (125 LD50/ml). The reduction of intensity of signal for intact
SNAP-25 was accompanied by the appearance of a band corresponding
to the smaller SNAP-25 cleavage product, and this band was clearly
visible at the highest BoNT/A concentrations of 125 and 25 LD50/ml
(Fig. 3a). The rabbit polyclonal (neoepitope) antibody detected the
3.3. Capture ELISA for detection of BoNT/A cleaved SNAP25 from
differentiated mouse ES cells
In a preliminary study, ESDNs at DIV3 and DIV9 were treated with
pure BoNT/A (80–0.039 LD50/ml = 16.0–0.0078 LD50/dose) for 48
and 72 h. One set of cells from DIV3 were also given the same treatment
but BoNT/A was incubated with cells for 24 h after which time the
toxin was removed and cells incubated for further 48 h in culture
medium without toxin. Results from this preliminary study (Fig. 4a)
show a positive dose response for all 3 treatments, with a dose de-
pendent increase in the detection of cleaved SNAP-25. There was a clear
increase in sensitivity when the differentiation time increased from
DIV3 to DIV9 despite the longer incubation time with toxin at DIV3
(EC50 of 3.25 LD50/ml for DIV3 and 1.08 LD50/ml for DIV9). A shorter
exposure to BoNT/A followed by 2 days of culture in medium without
toxin only did not improve sensitivity (EC50 of 7.17 LD50/ml).

5
G. Yadirgi et al. Journal of Immunological Methods xxx (xxxx) xxx–xxx

Fig. 3. Detection of cleaved SNAP-25 from ESDNs treated with BoNT/A, by Western blotting. (a) Presence of SNAP-25 and BoNT/A cleaved SNAP-25 in lysed ESDNs at DIV12 treated for
48 h with 125 (lane 3), 25 (lane 4), 5.0 (lane 5), 1.0 (lane 6) and 0.2 (lane 7) LD50/ml of BoNT/A. Lane 2 = control (no toxin). Lane 1 = molecular weight markers. Blots were visualised
with mouse monoclonal antibody, recognising both cleaved and un-cleaved SNAP25. (b) Presence of BoNT/A cleaved SNAP-25 in lysed ESDNs at DIV12 treated for 48 h with 125 (lane 3),
25 (lane 4), 5.0 (lane 5), 1.0 (lane 6) and 0.2 (lane 7) LD50 of BoNT/A. Lane 2 = control (no toxin). Lane 1 = molecular weight markers. Blots were visualised with rabbit BoNT/A
cleavage site specific antibody, detecting only cleaved SNAP-25.

A preliminary time course study using ESDNs at DIV9 was per- Table 1
formed by incubating cells with BoNT/A, 80–0.039 LD50/ml Variation in EC50 values for different preparations of ESDNs and different ELISA plates.
(16–0.0078 LD50/dose) for between 12 and 72 h (Fig. 4b), followed by
Plate No. Cell batch 1 Cell batch 2 Cell batch 3
detection of cleaved SNAP-25 by ELISA. Assay sensitivity increased on
extended exposure to toxin, with EC50 values (LD50/ml) ranging from DIV9 DIV9 DIV12
9.41 at 12 h to 0.74 LD50/ml at 72 h, although there was little differ-
1 1.08 1.70 1.94
ence beyond the 48 h time point.
1.82
The dose response and EC50 for BoNT/A was confirmed for two 2 1.06 1.15 1.10
different preparations of ESDNs at DIV9 and for a third cell preparation 1.26
at DIV12, and in each case cleaved SNAP-25 was measured on at least 3 1.46 1.52 ND
two ELISA plates. The results (shown as EC50 in LD50/ml) are sum- 1.43
1.41
marised in Table 1. Taking all the data together, the minimum level of
1.41
detection was determined as at least 0.1 LD50/ml (3 fM) with an EC50 4 ND 1.42 ND
of 1.39 (1.24–1.55) LD50/ml (6 pg/ml; 45 fM). The EC50 data from all
3 cell batches for all plates were log10 transformed and a random effects Data are the EC50 for the dose response to pure BoNT/A (48 h treatment). Results are
model was used to determine between-plate variance, which was con- shown for 3 different cell batches (DIV9–DIV12), titrated on up to 4 plates with single or
replicate titrations per plate as indicated. EC50 values were calculated from log trans-
verted to geometric coefficient of variation (GCV). The between-plate
formed data normalised as a percentage of the largest value in each data set. ND = not
GCV was estimated as 22.4%. done.

activity even when this component was used at a 4-order of magnitude

3.4. Specificity of the capture ELISA for active BoNT/A
The specificity of the cell based assay with SNAP-25 capture ELISA
was determined using botulinum toxin from serotype B (which cleaves
a different SNARE protein) and E (which cleaves SNAP-25 but at a
different position to toxin A). A dose dependent cleavage of SNAP-25
was confirmed for BoNT/A but there was no measureable response for
either the B or E toxin and OD values for these samples were compar-
able to those seen for cells incubated without any toxin (Fig. 5a). The
positive response for toxin A was also absent in lysates from cells that
had been incubated with a heat inactivated BoNT/A preparation
(Fig. 5a). Specificity of the assay for active A toxin was further con-
firmed using the non-toxic fragment of the BoNT/A (BoNT/A LHn)
which lacks the receptor binding domain. There was no detectable
higher concentration (Fig. 5b).
3.5. Detection of BoNT/A in pharmaceutical products
The cell based assay was used to measure the activity of two
pharmaceutical products containing BoNT/A in the form of complexed
toxin with bulking/stabilising excipients. Both products had a very si-
milar dose response profile to the commercial BoNT/A (Fig. 6) sug-
gesting that this assay might be suitable to assess products of this type.
Using the commercial BoNT/A as a reference toxin, the precision of the
relative potency estimate for both product 1 and product 2 was calcu-
lated as 85–118% which complies with recommended validity criteria
for a valid potency estimate according to the European Pharmacopeia

a 3.0 DIV9 (48h) b

3.0 DIV9 (12h)
DIV3 (72h)
DIV9 (24h)
2.5 DIV3 (24h + 48h)
2.5 DIV9 (48h)
DIV9 (72h)
2.0 2.0
OD 405 nm

OD 405 nm

1.5 1.5

1.0 1.0

0.5 0.5

0.0 0.0
0.01 0.1 1 10 100 0.01 0.1 1 10 100
BoNT/A LD50/ml BoNT/A LD50/ml

Fig. 4. Detection of cleaved SNAP-25 from ESDNs treated with BoNT/A by capture ELISA. (a) cells were differentiated for 3 (DIV3) or 9 (DIV9) days on 96-well tissue culture plates and
treated with 200 μl purified BoNT/A (80 to 0.04 LD50/ml;16 to 0.008 LD50/well) for 24, 48 and 72 h. (b) cells at DIV9 were exposed to same dose range of pure BoNT/A (80 to
0.04 LD50/ml) for 12, 24, 48 and 72 h. Results presented are from one assay and each data set is from a single titration. In both figures the dotted line shows the absorbance reading for
control cells control without toxin.

6
G. Yadirgi et al. Journal of Immunological Methods xxx (xxxx) xxx–xxx

a 1.5 b
BoNT/A
BoNT/B 1.5

BoNT/E
1.0 BoNT/A (Heat Inactivated)
OD 405 nm

BoNT/A LHn
1.0

OD 405 nm
BoNT/A

0.5
0.5

0.0 0.0
0.001 0.01 0.1 1 10 100 1000

-8

-6

-4

-2

4
-1

10

10

10
10

10

10

10
10
BoNT/A LD50/ml BoNT/A [µg/ml]

Fig. 5. Specificity of capture ELISA for detection of biologically active BoNT/A. ESDNs at DIV9 were incubated for 48 h with (a) purified BoNT/A, BoNT/B, BoNT/E (650 to 0.0037 LD50/
ml) and equivalent concentration of heat inactivated BoNT/A. The heat inactivation of Metabiologics pure holotoxin of BoNT/A1 was achieved by boiling at 95 °C for 5 min. b) purified
BoNT/A (10− 2 and 10− 8 μg/ml) and non-toxic fragment of BoNT/A, LHn/A (300 to 0.0017 μg/ml) lacking receptor binding domain. Results are from a single assay and each data set is
from a single titration. Dotted lines in each experiment indicate controls where cells were not exposed to any treatments.

1.5 1.5
BoNT/A Anti-A + BoNT/A
Product 1 Anti-E + BoNT/A
Product 2 Anti-A
1.0 1.0
OD 405 nm

OD 405 nm
0.5 0.5

0.0 0.0
0.0001 0.001 0.01 0.1 1 10 100 1000 0.00001 0.0001 0.001 0.01 0.1 1
BoNT/A LD50/ml Antitoxin IU/ml

Fig. 6. Detection of cleaved SNAP-25 from ESDNs treated with BoNT/A in therapeutic
formulations. After 48 h incubation cells were lysed and the lysates from each individual
well were subjected to BoNT/A specific capture ELISA to detect cleaved SNAP-25. Dotted
lines are cell controls with no toxin. Results are from a typical experiment performed on
two independent occasions and each data set is the mean from triplicate titration ± SD.
monograph for Botulinum toxin A for injection (01/2012:2113).
Fig. 7. Inhibition of BoNT/A cleavage of SNAP-25 in ESDNs by antibodies against bo-
tulinum A toxin. Cells at DIV12 were treated for 48 h with a mixture of purified BoNT/A
toxin (20 LD50/ml; 4 LD50/well) and reference standard for type A antitoxin (NIBSC
product code 59/021), or reference antitoxin for BoNT/E (NIBSC product code 02/318),
used in the range between 0.1 IU and 0.05 mIU. Results are from a representative ex-
periment where each data set is a mean from a duplicate titration ± SD. Negative controls
included cells treated with reference antisera against BoNT/A in the absence of BoNT/A
toxin (n = 1).

3.6. Neutralization of BoNT/A effect on neurons from mouse EC cells

determined by capture ELISA
To evaluate the suitability of the cell based assay for measuring
antitoxin potency, neutralization assays were performed with a fixed
dose (20 LD50/ml) of pure BoNT/A which corresponds to 4 LD50/dose
(~ 15 pg of pure BoNT/A per dose). This dose of toxin had been shown
to be towards the top of the linear part of the dose response curve. This
concentration of toxin was neutralized by anti-A antitoxin in a dose
dependent manner in the range between 1 and 100 mIU/ml (Fig.7).
There was no neutralization of BoNT/A activity in the presence of an
equine polyclonal antibody to BoNT/E further demonstrating the spe-
cificity of the assay and confirming that neutralization is serotype
specific. Furthermore, the anti-A antitoxin alone had no effect on SNAP-
25 cleavage (Fig. 7).
4. Discussion
Cell based assays for detection of BoNT/s are considered to be the
most suitable approach to replace the mouse bioassay. This is due to the
fact that they are capable of reflecting all key steps required for toxin
activity in vivo and can be standardized and applied for high
throughput routine testing of therapeutic toxin products (NIH
Publication number, 2008; Adler et al., 2010). In this study we report
preliminary development of a cell based assay that has a sensitivity
exceeding that of the mouse bioassay and is the most sensitive cell
based assay for detection of BoNT/A reported to date.
During the past decade, much effort has focused on development of
sensitive assays for BoNT/s which could provide sensitivity equivalent
to the mouse bioassay. Primary rodent neuronal cells have been used in
BoNT research for many years (Keller et al., 2004; Bigalke et al.,1985;
Dong et al., 2007), but it was not until rat spinal cord cells were first
applied for quantitative detection of purified BoNT/A that a cell based
assay for potency determination was realistically possible (Pellet et al.,
2010). This assay is dependent on immunodetection of cleaved and
uncleaved SNAP-25, quantified by Western blotting (WB) using a
commercial antibody that detects both molecules. The level of sensi-
tivity (EC50) was reported to be 133 LD50/ml (5.7 pg or 3.8 LD50/
well) which was claimed to be comparable to the mouse bioassay
(Pellet et al., 2010). However assays using primary neuronal cells do
not provide an easily standardized and robust model suitable for rou-
tine laboratory use, require relatively lengthy manipulation of cells, and
use of animals as a source of the cells. Assays using these cells were
therefore not used widely as an alternative to the animal potency test,
despite their high sensitivity in comparison to all other established cell
lines (Pellet, 2013; Hubbard et al., 2012).
Availability of neurons derived from pluripotent stem cells (Bibel
et al., 2004, 2007) has significantly contributed to the field of cell based
assay development and provided a new platform for highly sensitive
detection of BoNT/s. Improvements in procedures for the culture and
differentiation of stem cells, and reduction in cost of critical reagents,

7
G. Yadirgi et al.
has further advanced the field of assay development for detection of
BoNT/A (Hubbard et al., 2012; Kiris et al., 2014).
Several studies have used mESC combined with immunoblotting to
monitor dose dependent proteolysis of SNAP-25 by BoNT/A (McNutt
et al., 2011; Pellet et al., 2011; Kiris et al., 2011; Hubbard et al., 2012,)
and other BoNT/serotypes (Hubbard et al., 2012). McNutt et al. (2011)
determined an EC50 of 0.4 pM for BoNT/A in mESC derived neurons at
DIV12 (based on detection of cleaved SNAP-25 by WB) following 48 h
treatment with toxin. Pellet et al. (2011) reported a minimum detection
of 104 fM, with EC50 of 1.7 pM, using neurons from mES cells after
9 days in culture and 24 h incubation with pure BoNT/A. A study by
Hubbard et al. (2012) confirmed only 50% cleavage of SNAP-25 with
67 pM of BoNT/A in enriched glutamatergic neurons from mESCs. Kiris
et al. (2011) used a high-throughput ELISA method combined with
immunofluorescence to monitor disappearance of full length SNAP-25
using BoNT/A cleavage sensitive antibodies. In this study they reported
a dynamic dose response range between 25 and 1000 pM for BoNT/A,
which is considerably inferior in sensitivity compared to results pre-
sented in this study. Furthermore, in the previous study of Nuss et al.
(2010) using these same antibodies, sensitive to the BoNT/A cleavage
site, in combination with primary neuronal cells (embryonic chicken
spinal motor neurons) detected BoNT/A only in the range between 0.1
and 10 nM.
Our method includes a capture ELISA read-out which is comparable
to studies of Kiris et al. (2011) and Nuss et al. (2010) and offers sig-
nificantly higher throughput capacity compared to WB, with higher
sensitivity. The assay reported in this study has a minimum detection
level of 0.1 LD50/ml which is equivalent to 0.02 LD50 per dose
(0.44 pg/ml or 3 fM) and an EC50 of 1.5 LD50/ml (6.5 pg/ml; 48 fM).
Since neurons derived from mESC were also used in previous studies, it
is likely that the differentiation protocol and format of the capture
ELISA, including the capture and detection antibodies used contributed
to the enhances detection of BoNT/A in comparison to previous studies.
One key difference to the method previously reported (Nuss et al.,
2010; Kiris et al., 2011) is that in our approach, the capture antibody
used in the ELISA specifically and selectively detects only BoNT/A-
cleaved SNAP-25 product so that appearance of product, rather than
loss of SNAP-25 is detected. The same antibody has been used pre-
viously in development of highly sensitive biochemical assays for
BoNT/A, that have even higher sensitivity compared to the cell based
assay (Jones et al., 2008; Liu et al., 2012), but suffer from the limitation
previously noted that they are incapable of reflecting all key steps in-
volved in toxin activity in vivo. It is worth noting that antibody de-
tecting only the BoNT/A cleaved SNAP25, and used as a capture anti-
body, was at least 25-fold more effective in detecting this fragment
compared to the commercially sourced antibody detecting both cleaved
and un-cleaved SNAP-25 as determined by WB studies reported here.
This suggests that the cleavage site specific antibody plays an important
role in determining sensitivity of the assay. In addition to the capture
antibody, the cell based assay described here includes two separate
detection antibodies directed against SNAP-25111–157 and SNAP251–57
(regions of SNAP25 protein which are unaffected by toxin-mediated
proteolysis) that are used to amplify the detection signal. Although the
study presented here does not represent a formal validation, the results
obtained suggest that the assay is likely to be suitable for development
into a robust, high throughput routine assay. Although the dose re-
sponse curve and absolute EC50 differs between plates (from the same
cell batch) and between cell batches, it should be noted that in routine
use test toxin or antitoxin samples would be titrated alongside a re-
ference (as is the case for the mouse bioassay) which should sig-
nificantly reduce the between plate and between cell batch variability
observed in this preliminary study.
Recent studies on neurons derived from human induced pluripotent
stem cells also confirm highly sensitive detection of BoNTs (Pellett
et al., 2015; Whitemarsh et al., 2012). Two commercially sourced
human stem cell derived neurons (iCell and HIP™) were exposed to pure
8
Journal of Immunological Methods xxx (xxxx) xxx–xxx
BoNT/A for 48 h and effect of toxin on SNAP-25 determined by WB and
capture ELISA (Pellett et al., 2017). In the capture ELISA, the EC50 for
both HIP and iCell neurons was 0.34 LD50 per dose which is very
comparable to our findings presented in this study using ESDNs. The
capture ELISA was more sensitive compared to WB where EC50s of 0.49
and 1.5 LD/50 per dose were reported for iCell and HIPS neurons, re-
spectively. Different detection antibodies were used in the capture
ELISA and WB which may contribute to the differences in sensitivity.
Whereas human stem cells may be the preferred option in some studies,
more robust differentiation and a higher protein yield in a 96 well
format makes mES cells more attractive for replacement of the mouse
bioassay in the majority of applications currently relying on animal
models. It is worth noting that the low protein yield in cell lysates from
human neurons necessitated modification of capture ELISA to improve
the signal detection (Pellett et al., 2017).
Detection of cleavage products is specific and quantitative for
BoNTs and therefore is a desirable endpoint for potency determination
of pharmaceutical BoNTs. An ELISA based detection method for cleaved
SNAP-25 has been developed for detection of BoNT/A in Botox®
pharmaceutical products as a replacement of the mouse LD50 potency
test (Fernandez-Salas et al., 2012). In this study, differentiated SiMa
cells (human neuroblastoma cell line) were used in an ELISA format
incorporating a monoclonal antibody detecting cleaved SNAP-25. The
detection of BoNT/A in Botox® formulations was reported with an EC50
of 4–6 LD50/ml (1.0–0.4 LD50/well), for BoNT/A complex an EC50 of
3.3 pM, and for pure BoNT/A an EC50 of 0.12 nM. In our study, EC50s
for formulated HA complexed toxin products were ~1.0 LD50/ml and
for pure BoNT/A ~1.5 LD50/ml (48 fM) confirming that ESDNs are
more sensitive to BoNT/A compared to SiMa cells.
There are limited reports describing cell based assays for toxin
neutralization from which an appropriate level of sensitivity can be
determined. Pellett et al. (2007) used rat primary neuronal cells and
confirmed detection of antibodies in patients resistant to BoNT/A
therapy. The authors did not include an International Standard to de-
termine the level of sensitivity, but applied a theoretical value that one
International Unit of reference type A antitoxin would neutralize 104
mouse LD50 units. They used this extrapolation to estimate potencies of
6–7.5 mIU/ml for samples derived from patients resistant to toxin
therapy. However, it should be noted that this extrapolation is not
necessarily transferrable to systems other than the mouse bioassay.
Furthermore, the mouse toxin neutralization assay has been confirmed
not to be sufficiently sensitive to detect low levels of serum antibodies
in patients resistant to toxin therapy (Göschel et al., 1997). Hall et al.
(2004) used rat embryonic spinal cord cells combined with a [3H]
glycine release assay and reported a linear dose response in the range of
0.001 to 0.1 IU/ml. This is comparable to our study where we confirm a
similar sensitivity and dose response for the same International Re-
ference Standard for type A antitoxin. It should also be noted that the
detection limit for anti-A antitoxin in the cell based assay we describe
here is likely to be enhanced when a lower toxin dose is used in the
neutralization assay.
Most recently, the blockade of synaptic transmission in network
cultures of neurons derived from stem cells has been described as a
functionally relevant readout for BoNT intoxication (Beske et al., 2015,
2016, Jenkinson et al., 2017). Using stem cell derived neurons on multi
electrodes, Jenkinson et al. (2017) confirmed a significant reduction of
both spontaneous network bursts and average spike rates with 1.66 pM
of BoNT/A and 33 LD50/ml of Botox®. Beske et al. (2016) measured
spontaneous neurotransmission frequency on neurons derived from
mouse ESC and reported synaptic activity for BoNT/A with an IC50 of
0.05pM for BoNT/A. Exposure of mESC to clinical concentrations of
Botox® demonstrated that exposure to 1 or 2 LD50/ml resulted in re-
duced neurotransmission and production of cleaved SNAP-25 after 20 h
exposure. The sensitivity to BoNT/A reported in these studies is com-
parable to the level of detection presented in our study for both com-
mercial BoNT/A and therapeutic toxin preparations. However, despite
G. Yadirgi et al.
being functionally relevant, the measurement of the blockade of sy-
naptic transmission as an endpoint relies on the use of specialized
equipment which is not readily available in most routine control la-
boratory settings, and less favourable for high throughput capability.
In conclusion, our study shows that neurons from mESCs combined
with a simple capture ELISA for detection of toxin-cleaved substrate
could be used to develop a highly sensitive in vitro assay for measuring
the potency of BoNT/A and anti-BoNT/A antitoxin. Subject to formal
validation, this assay could serve as a complete replacement for mouse
bioassays currently used to measure activity of BoNT/A toxin and an-
titoxin.
Acknowledgements
This study was funded by NC3Rs grant ID 94812 (reference no
G1000086) and National Institute for Biological Standards and Control.
Dr. RGA Jones provided training to GY during initial assay set up. The
authors would also like to thank Rob Tierney for proof reading and
Peter Rigsby for statistical assistance.
Authors contributions
DS and PS conceived and designed the experiments. GY, SR, PS and
YL performed the experiments and analysed the data. PS and SR pre-
pared figures and critically reviewed the paper. DS wrote the paper.
Conflict of interest
The authors declare no conflict of interest.
References
Adler, S., Bicker, G., Bigalke, H., Bishop, C., Blumel, J., Dressler, D., Fitzgerald, J., Gessler,
F., Heuschen, H., Kegel, B., Luch, A., Milne, C., Pickett, A., Ratsch, H., Ruhdel, I.,
Sesardic, D., Stephens, M., Stiens, G., Thornton, P.D., Thurmer, R., Vey, M.,
Spielmann, H., Grune, B., Liebsch, M., 2010. The current scientific and legal status of
alternative methods to the LD50 test for botulinum neurotoxin potency testing.
Report and recommendations of the ZEBET Expert meeting. Altern. Lab. Anim 38 (4),
315–330.
Beske, P.H., Scheeler, S.M., Adler, A., McNutt, P.M., 2015. Accelerated intoxication of
GABAergic synapses by botulinum neurotoxin A disinhibits stem cell-derived neuron
networks prior to network silencing. Front. Cell. Neurosci. 9, 159. http://dx.doi.org/
10.3389/fncel.2015.00159.
Beske, P.H., Bradford, A.B., Grynovicki, J.O., Glotfelty, E.J., Hoffman, K.M., Hubbard,
K.S., Tuznik, K.M., McNutt, P.M., 2016. Botulinum and tetanus neurotoxin-induced
blockade of synaptic transmission in networked cultures of human and rodent neu-
rons. Toxicol. Sci 149 (2), 503–515. http://dx.doi.org/10.1093/toxsci/kfv254.
Bibel, M., Richter, J., Schrenk, K., Schrenk, K., Kerry, Tucker L., Staiger, V., Korte, L.,
Goetz, M., Barde, Y.-A., 2004. Differentiation of mouse embryonic stem cells into a
defined neuronal lineage. Nat. Neurosci. 7, 1003–1009. http://dx.doi.org/10.1038/
nn130.
Bibel, M., Richter, J., Lacroix, E., Barde, Y.-A., 2007. Generation of a defined and uniform
population of CNS progenitors and neurons from mouse embryonic stem cells. Nat.
Protoc. 2, 1034–1043. http://dx.doi.org/10.1038/nprot.2007.147.
Bigalke, H., Dreyer, F., Bergey, G., 1985. Botulinum A neurotoxin inhibits non-cholinergic
synaptic transmission in mouse spinal cord neurons in culture. Brain Res. 360 (1–2),
318–324.
Dong, M., Tepp, W.H., Johnson, E., Chapman, E.R., 2004. Using fluorescent sensors to
detect botulinum neurotoxin activity in vitro and in living cells. Proc. Natl. Acad. Sci.
U. S. A. 101, 14701–14706. http://dx.doi.org/10.1073/pnas.0404107101.
Dong, M., Tepp, W.H., Liu, H., Johnson, E.A., Chapman, E.R., 2007. Mechanism of bo-
tulinum neurotoxin B and G entry into hippocampal neurons. J. Cell Biol. 179,
1511–1522. http://dx.doi.org/10.1083/jcb.200707184.
Dressler, D., 2012. Clinical applications of botulinum toxin. Curr. Opin. Microbiol. 15,
325–336. http://dx.doi.org/10.1016/j.mib.2012.05.012.
Ekong, T.A.N., Feavers, I., Sesardic, D., 1997. Recombinant SNAP25 is an effective sub-
strate for clostridium botulinum type A toxin endopeptidase activity in vitro.
Microbiology 143, 3337–3347. http://dx.doi.org/10.1099/00221287-143-10-3337.
Evans, E.R., Skipper, P.J., Shone, C.C., 2009. An assay for botulinum toxin types A, B and
F that requires both functional binding and catalytic activities within the neurotoxin.
J. Appl. Microbiol. 107, 1384–1391. http://dx.doi.org/10.1111/j.1365-2672.2009.
04325.
Fernandez-Salas, E., Wang, J., Molina, Y., Nelson, J.B., Jacky, B.P.S., Aoki, R.K., 2012.
Botulinum neurotoxin serotype A specific cell based potency assay to replace the
mouse bioassay. PLoS One 7 (11), e49516. http://dx.doi.org/10.1371/journal.pone.
0049516.
9
Journal of Immunological Methods xxx (xxxx) xxx–xxx
Gill, D.M., 1982. Bacterial toxins: A table of lethal amounts. Microbiol. Rev. 46, 86–94.
Gilmore, M.A., Williams, D., Okawa, Y., Holguin, B., James, N.G., Ross, J.A., Aoki, R.K.,
Jameson, D.M., Steward, L.E., 2011. Depolarization after resonance energy transfer
(DARET): a sensitive fluorescence-based assay for botulinum neurotoxin protease
activity. Anal. Biochem. 413, 36–42. http://dx.doi.org/10.1016/j.ab.2011.01.043.
Göschel, H., Wohlfarth, K., Frevert, J., Dengler, R., Bigalke, H., 1997. Botulinum A toxin
therapy: neutralizing and nonneutralizing antibodies-therapeutic consequences. Exp.
Neurol. 147, 96–102. http://dx.doi.org/10.1006/exnr.1997.6580.
Hall, Y.H., Chaddock, J.A., Moulsdale, H.J., Kirby, E.R., Alexander, F.C., Marks, J.D.,
Foster, K.A., 2004. Novel application of an in vitro technique to the detection and
quantification of botulinum neurotoxin antibodies. J. Immunol. Methods 288, 55–60.
http://dx.doi.org/10.1016/j.jim.2004.02.011.
Hubbard, K.S., Gut, I.M., Lyman, M.E., Tuznik, K.M., Mesngon, M.T., McNutt, P.M., 2012.
High yield derivation of enriched glutamatergic neurons from suspension-cultured
mouse ESCs for neurotoxicology research. BMC Neurosci. 13, 127. http://dx.doi.org/
10.1186/1471-2202-13-127.
Jenkinson, S.P., Grandgirard, D., Heidemann, M., Tscherter, A., Avondet, M.-A., Leib,
S.L., 2017. Embryonic stem cell-derived neurons grown on multi-electrode arrays as a
novel in vitro bioassay for the detection of clostridium botulinum neurotoxins. Front.
Pharmacol. 8, 73. http://dx.doi.org/10.3389/fphar.2017.00073.
Jones, R.G.A., Corbel, M.J., Sesardic, D., 2006. A review of WHO international standards
for botulinum antitoxins. Biologicals 34, 223–226. http://dx.doi.org/10.1016/j.
biologicals.2005.11.009.
Jones, R.G.A., Ochiai, M., Liu, Y., Ekong, T., Sesardic, D., 2008. Development of improved
SNAP25 endopeptidase immune-assays for botulinum type A and E toxins. J.
Immunol. Methods 329, 92–101. http://dx.doi.org/10.1016/j.jim.2007.09.014.
Jones, R.J., Liu, Y., Sesardic, D., 2009. Development of a novel highly specific en-
dopeptidase assay for botulinum toxin type C1 activity. J. Immunol. Methods 343,
21–27. http://dx.doi.org/10.1016/j.jim.2009.01.001.
Jones, R.J., Liu, Y., Hallis, C., Thorpe, S.J., Longstaff, C., Matejtschuk, P., Sesardic, D.,
2011. Release of proteolytic activity following reduction in therapeutic human serum
albumin containing products: detection with a new neoepitope endopeptidase im-
munoassays. J. Pharm. Biomed. Anal. 54, 74–80. http://dx.doi.org/10.1016/j.jpba.
2010.08.013.
Keller, J.E., Neale, E.A., 2001. The role of the synaptic protein snap-25 in the potency of
botulinum neurotoxin type A. J. Biol. Chem. 276, 1348–13476. http://dx.doi.org/10.
1074/jbc.M010992200.
Keller, J.E., Cai, F., Neale, E.A., 2004. Uptake of botulinum neurotoxin into cultured
neurons. Biochemistry 43 (2), 526–532. http://dx.doi.org/10.1021/bi0356698.
Kiris, E., Nuss, J.E., Burnett, J.C., Kota, K.P., Koh, D.C., Wanner, L.M., Torres-Melendez,
E., Gussio, R., Tessarollo, L., Bavari, S., 2011. Embryonic stem cell-derived moto-
neurons provide a highly sensitive cell culture model for botulinum neurotoxin stu-
dies, with implications for high-throughput drug discovery. Stem Cell Res. 6,
195–205. http://dx.doi.org/10.1016/j.scr.2011.01.002.
Kiris, E., Kota, K.P., Burnett, J.C., Soloveva, V., Kane, C.D., Bavari, S., 2014. Recent de-
velopments in cell-based assays and stem cell technologies for Botulinum neurotoxin
research and drug discovery. Expert. Rev. Mol. Diagn. 14, 153–168. http://dx.doi.
org/10.1586/14737159.2014.867808.
Liu, Y., Rigsby, P., Sesardic, D., Marks, J.D., Jones, R.G., 2012. A functional dual-coated
(FDC) microtitre plate method to replace the botulinum toxin LD50 test. Anal.
Biochem. 425, 28–35. http://dx.doi.org/10.1016/j.ab.2012.02.038.
McNutt, P., Celver, J., Hamilton, T., Mesngon, M., 2011. Embryonic stem cell-derived
neurons are a novel, highly sensitive tissue culture platform for botulinum research.
Biochem. Biophys. Res. Commun. 405, 85–90. http://dx.doi.org/10.1016/j.bbrc.
2010.12.132.
National Institute of Environment and Health Services, 2008. Report on the ICCVAM-
NICEATM/ECVAM Scientific Workshop on Alternatives Methods to Refine, Reduce or
Replace the Mouse LD50 Assay for Botulinum Toxin Testing. (NIH Publication
number 08-6416).
Nuss, J.E., Ruthel, G., Tressler, L.E., Wanner, L.M., Torres-Melendez, E., Hale, M.L.,
Bavari, S., 2010. Development of cell-based assays to measure botulinum neurotoxin
serotype A activity using cleavage-sensitive antibodies. J. Biomol. Screen. 15 (1),
42–51. http://dx.doi.org/10.1177/1087057109354779.
Pellet, S., 2013. Progress in cell based assays for botulinum neurotoxin detection. Curr.
Top. Microbiol. Immunol. 364, 257–285. http://dx.doi.org/10.1007/978-3-642-
33570-9_12.
Pellet, S., Tepp, W., Toth, S.I., Johnson, E.A., 2010. Comparison of the primary rat spinal
cord cell (RSC) assay and the mouse bioassay for Botulinum neurotoxin type A po-
tency determination. J. Pharmacol. Toxicol. Methods 61 (10), 304–310. http://dx.
doi.org/10.1016/j.vascn.2010.01.003.
Pellet, S., Du, Z.W., Pier, C.L., Tepp, W.H., Zhang, S.C., Johnson, E.A., 2011. Sensitive and
quantitative detection of botulinum neurotoxin in neurons derived from mouse em-
bryonic stem cells. Biochem. Biophys. Res. Commun. 404, 388–392. http://dx.doi.
org/10.1016/j.bbrc.2010.11.128.
Pellett, S., Tepp, W.H., Clancy, C.M., Borodic, G.E., Johnson, E.A., 2007. A neuronal cell-
based neurotoxin assay for highly sensitive and specific detection of neutralizing
serum antibodies. FEBS Lett. 581 (25), 4803–4808. http://dx.doi.org/10.1016/j.
febslet.2007.08.078.
Pellett, S., Schwartz, M.P., Tepp, W.H., Josephson, R., Scherf, J.M., Pier, C.L., Thomson,
J.A., Murphy, W.L., Johnson, E.A., 2015. Human induced pluripotent stem cell de-
rived neuronal cells cultured on chemically-defined hydrogels for sensitive in vitro
detection of botulinum neurotoxin. Sci Rep 5, 14566. http://dx.doi.org/10.1038/
srep14566.
Pellett, S., Tepp, W.H., Johnson, E.A., Sesardic, D., 2017. Assessment of ELISA as endpoint
in neuronal cell-based assay for BoNT detection using hiPSC derived neurons. J.
Pharmacol. Toxicol. Methods 88, 1–6. http://dx.doi.org/10.1016/j.vascn.2017.04.
G. Yadirgi et al.
013.
Pirazzini, M., Rossetto, O., Eleopra, R., Montecucco, C., 2017. Botulinum neurotoxins:
biology, pharmacology, and toxicology. Pharmacol. Rev. 69, 200–235. http://dx.doi.
org/10.1124/pr.116.012658.
Purkiss, J.R., Friis, L.M., Doward, S., Quinn, C.P., 2001. Clostridium botulinum neuro-
toxins act with a wide range of potencies on SH-SY5Y human neuroblastoma cells.
Neurotoxicology 22 (4), 447–453.
Rasetti-Escargueil, C., Machado, C.M., Preneta-Blanc, R., Fleck, R.A., Sesardic, D., 2011.
Enhanced sensitivity to botulinum type A neurotoxin of human neuroblastoma SH-
SY5Y cells after differentiation into mature neuronal cells. Botulinum J. 22 (1),
30–48. http://dx.doi.org/10.1504/TBJ.2011.041814.
Rosen, O., Ozeri, E., Barnea, A., David, A.B., Zichel, R., 2016. Development of an in-
novative in vitro potency assay for anti-botulinum antitoxins. Toxins 8 (10), 276.
http://dx.doi.org/10.3390/toxins8100276.
Rossetto, O., Pirazzini, M., Montecucco, C., 2014. Botulinum neurotoxins: genetic,
structural and mechanistic insights. Nat. Rev. Microbiol. 12, 535–549. http://dx.doi.
org/10.1038/nrmicro3295.
Schiavo, G., Santucci, A., Dasgupta, B.R., Mehta, P.P., Jontes, J., Benfenati, F., Wilson,
M.C., Montecucco, C., 1993. Botulinum neurotoxins serotypes A and E cleave SNAP-
25 at distinct COOH-terminal peptide bonds. FEBS Lett. 335, 99–103.
Sesardic, D., 2012. Bioassays for evaluation of medical products derived from bacterial
toxins. Curr. Opin. Microbiol. 15 (3), 310–317. http://dx.doi.org/10.1016/j.mib.
2012.05.008.
10
View publication stats
Journal of Immunological Methods xxx (xxxx) xxx–xxx
Sesardic, D., Gaines Das, R., 2008. Alternatives to the LD50 for botulinum toxin potency
testing: strategies and progress towards refinement, reduction and replacement. In:
Proceedings of the 6th World Congress on Alternatives, Tokyo, 2007, pp. 581–585.
Sesardic, D., Jones, R.G., Leung, T., Alsop, T., Tierney, R., 2004. Detection of antibodies
against botulinum toxins. Mov. Disord. 19, 85–91. http://dx.doi.org/10.1002/mds.
20021.
Simon, S., Fiebig, U., Liu, Y., Tierney, R., Dano, J., Worbs, S., Endermann, T., Nevers, M.-
C., Volland, H., Sesardic, D., Dorner, M.B., 2015. Recommended immunological
strategies to screen for botulinum neurotoxin-containing samples. Toxins 7 (12),
5011–5034. http://dx.doi.org/10.3390/toxins7124860.
Vaidyanathan, V.V., Yoshino, K., Jahnz, M., Dörries, C., Bade, S., Nauenburg, S.,
Niemann, H., Binz, T., 1999. Proteolysis of SNAP-25 isoforms by botulinum neuro-
toxin types A, C, and E: domains and amino acid residues controlling the formation of
enzyme-substrate complexes and cleavage. J. Neurochem. 72, 327–337.
Whitemarsh, R.C.M., Strathman, M.J., Chase, L.G., Stankewicz, C., Tepp, W.H., Johnson,
E.A., Pellett, S., 2012. Novel application of human neurons derived from induced
pluripotent stem cells for highly sensitive botulinum neurodetection. Toxicol. Sci.
126 (2), 426–435. http://dx.doi.org/10.1093/toxsci/kfr354.
Yowler, B.C., Kensinger, R.D., Schengrund, C.L., 2002. Botulinum neurotoxin A activity is
dependent upon the presence of specific gangliosides in neuroblastoma cells ex-
pressing synaptotagmin I. J. Biol. Chem. 277 (36), 32815–32819. http://dx.doi.org/
10.1074/jbc.M205258200.