Hindawi Publishing Corporation
Journal of Toxicology
Volume 2012, Article ID 760142, 9 pages
Retargeting Clostridiumdifficile ToxinB to
NeuronalCellsas a Potential VehicleforCytosolic Deliveryof
TherapeuticBiomoleculesto Treat Botulism
1Division of Infectious Diseases, Department of Biomedical Sciences, Tufts Cummings School of Veterinary Medicine,
200 Westboro Road, North Grafton, MA 01536, USA
2Synaptic Research LLC, 1448 South Rolling Road, Baltimore, MD 21227, USA
Correspondence should be addressed to Greice Krautz-Peterson, greice.krautz email@example.com
Received 16 May 2011; Accepted 13 July 2011
Academic Editor: S. Ashraf Ahmed
Copyright © 2012 Greice Krautz-Peterson et al.ThisisanopenaccessarticledistributedundertheCreativeCommonsAttribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
Botulinum neurotoxins (BoNTs) deliver a protease to neurons which can cause a flaccid paralysis called botulism. Development
of botulism antidotes will require neuronal delivery of agents that inhibit or destroy the BoNT protease. Here, we investigated the
potential of engineering Clostridium difficile toxin B (TcdB) as a neuronal delivery vehicle by testing two recombinant TcdB chi-
meras. For AGT-TcdB chimera, an alkyltransferase (AGT) was appended to the N-terminal glucosyltransferase (GT) of TcdB. Re-
combinant AGT-TcdB had alkyltransferase activity, and the chimera was nearly as toxic to Vero cells as wild-type TcdB, suggesting
efficient cytosolic delivery of the AGT/GT fusion. For AGT-TcdB-BoNT/A-Hc, the receptor-binding domain (RBD) of TcdB was
replaced by the equivalent RBD from BoNT/A (BoNT/A-Hc). AGT-TcdB-BoNT/A-Hc was >25-fold more toxic to neuronal cells
and >25-fold less toxic to Vero cells than AGT-TcdB. Thus, TcdB can be engineered for cytosolic delivery of biomolecules and
improved targeting of neuronal cells.
Clostridial toxins in nature are remarkably efficient cell
cytosol delivery vehicles with highly evolved cell-specific
delivery features that may be ideal for therapeutic applica-
tions. Specifically these toxins (1) gain entry to animals; (2)
survive in blood; (3) bind to target cells expressing a specific
receptor; (4) penetrate the target cells; (5) deliver an enzy-
(TcdA and TcdB) contain a receptor-binding domain (RBD)
that binds to receptors that are broadly expressed on cells
and then enters by endocytosis. Once in the endosome, the
toxins employ a translocation domain (TD) to deliver a glu-
cosyltransferase (GT) to the cytosol which inactivates Rho
GTPases and leads to cell death . The toxins also contain
a cysteine protease (CPD), located between GT and TD, that
cleaves the GT enzymatic “cargo” from the “delivery vehicle”
at the endosomal membrane and liberates it into the cytosol
C. difficile bacteria generally reside in the gut where
the released toxins intoxicate intestinal epithelial cells and
cause the disruption of tight junctions of epithelium and
its barrier function. It is likely that in severe cases of the
infection, the toxins penetrate into the submucosa and
toxins in the blood of the experimentally infected animals
[6, 7], suggesting that the toxins may be reasonably stable in
Recent developments have enabled the application of
TcdA and TcdB as therapeutic delivery vehicles. The Bacillus
megaterium (B. megaterium) expression system has been
shown to permit high-level expression of functional recom-
binant TcdB (5–10mg/L culture) . Secondly, the toxicity
of these agents is virtually eliminated by introducing two
2Journal of Toxicology
point mutations within the GT domain that should have
no effect on endosomal uptake and translocation to the
cytosol (Haiying Wang and Hanping Feng, unpublished
data). Finally, the limits of the GT, TD, and RBD domains
have recently been carefully defined in the literature ,
facilitating efforts to replace one or more domains with
similar domains from other toxins. We recently showed that
it was possible to replace the RBD from TcdB with the
RBD from TcdA and retain most or all of the toxin activity
(Haiying Wang and Hanping Feng, unpublished data).
One drawback to the use of TcdA or TcdB as cytosolic
delivery vehicles is the lack of cell specificity. This is in
contrast to botulinum neurotoxins (BoNTs) which display
a marked specificity for neuronal cells. BoNTs are CDC
Category A biodefense threat agents that cause paralysis
by entering the presynaptic terminal of motor neurons
and inhibiting neurotransmitter release. All seven BoNT
serotypes bind to a neuronal receptor through a receptor-
binding domain. The toxins then undergo endocytosis,
delivery of the BoNT protease cargo to the cytosol, and
subsequent cleavage of SNARE proteins [9–11]. Reversal
of neuronal intoxication must involve either the inhibition
and/or elimination of the protease. We and others have
reported development of biomolecules that potently inhibit
BoNT protease [12–14] or promote its degradation [15, 16].
In this study, we demonstrate that biomolecules fused to the
amino terminus of TcdB can be successfully delivered to the
cytosol of cells and that replacement of the TcdB RBD with
the equivalent RBD from BoNT serotype A (BoNT/A) leads
to a chimeric toxin with enhanced specificity for neurons.
These results indicate that it may be possible to develop
therapeutic agents based on TcdB in which biomolecules are
delivered to BoNT-intoxicated neurons that inhibit and/or
destroy the toxin protease. Such a treatment would promote
accelerated neuronal recovery from intoxication and thus
could serve as the first antidotes for treatment of botulism.
2.1. Bacterial and Mammalian Cell Cultures. Bacterial cul-
tures of Escherichia coli (TOP10 cells; Invitrogen, Carlsbad,
CA) and B. megaterium (MS941 strain; kindly provided by
Dr. Rebekka Biedendieck, Germany) were grown at 37◦C in
Luria-Bertani (LB) medium, containing ampicillin (Amp)
and tetracycline (Tet), respectively.
Mammalian cell lines were obtained from ATCC (Man-
assas, VA) and cultured as monolayers in 100mm cell culture
dishes at 37◦C and 5% CO2. Cells were reseeded twice a
week after harvest using 0.05% trypsin to suspend cells. The
murine neuroblastoma cell line, Neuro2A, and the human
neuroblastoma line, M17, were cultured in DMEM/F12
medium (Invitrogen) supplemented with 10% fetal calf
serum, 2mML-glutamine, 100units/mL penicillin, and
50μg/mL streptomycin sulfate. Vero cells (kidney epithelial
cells from African green monkey) were cultured in DMEM
(Invitrogen) supplemented with 10% fetal calf serum,
2mML-glutamine, 100units/mL penicillin, and 50ug/mL
2.2. Cloning of TcdB Constructs. Full-length TcdB was ex-
pressed in B. megaterium as described previously . To
generate AGT-TcdB, the DNA encoding an alkylguanine-
DNA alkyltransferase (AGT) flanked by 5?-BsiWI and 3?-
BsrGI sites was synthesized (Geneart, Germany). AGT is
a commercially available tag for producing AGT-fusion
proteins. AGT catalyzes its own covalent binding to the
can be labeled with probes such as biotin or fluorescein
to permit detection and/or cell localization of AGT-fusion
proteins . The AGT was appended in frame to TcdB
by digestion of AGT with BsiWI and BsrGI and ligation
into pHis1525/TcdB digested with BsrGI as represented
in Figure 1. To generate AGT-aTcdB-ΔGT construction,
a unique BamHI site (position 542) was created between
coding sequences of GT and CPD by overlapping PCR. Then
the AGT-tag coding DNA precisely replaced the GT in frame
with the CPD, by ligation into pHis1525/TcdB digested with
5?-BsrGI and 3?-BamHI.
To generate AGT-TcdB-BoNT/A-Hc, an AgeI site was
installed between the TD and RBD of TcdB. The entire
RBD of TcdB, consisting of the C-terminally combined
repetitive oligopeptides (CROPs, residues 1852–2366), was
then replaced by the heavy chain C-terminus of BoNT/A
(BoNT/A-Hc, residues 861–1296). BoNT/A-Hc coding DNA
 was amplified by PCR from BoNT/A coding DNA
and flanked by 5?-AgeI and 5?-XmaI restriction sites using
primers; sense: 5?-cgaccggtggtggaggcggttcaggcggaggtggct-
to separate TcdB and BoNT/A-Hc in the chimera. The
reverse primer encoded a His6 sequence at the carboxyl
coding end. All plasmid constructions were confirmed by
DNA sequencing and transformed into B. megaterium for
protein expression as described previously . All DNA
cloning and plasmid construction were performed at Tufts
University and approved by the Institutional Biosafety
Committees in agreement with NIH Recombinant DNA
2.3. Characterization of Recombinant TcdB Chimeric Proteins.
Expression and purification of His-tagged TcdB proteins was
performed essentially as described previously  with a few
For Western blots, AGT-TcdB and AGT-TcdB-BoNT/A-
Hc were separated on a 4–20% gradient polyacrylamide
gel by SDS-PAGE. AGT-tag fused to TcdB was detected
using a rabbit polyclonal serum anti-AGT (New England
Biolabs, Boston) at a dilution of 1:1000. Detection of full-
length TcdB was performed using an alpaca polyclonal
anti-TcdB serum, generated in our laboratory and diluted
1:106. The BoNT/A-Hc domain in AGT-TcdB-BoNT/A-Hc
chimeric protein was detected by a mouse anti-BoNT/A-
Hc monoclonal antibody (A11G12.4B- kindly provided by
Dr. Jean Mukherjee, Tufts University) diluted at 1:25,000.
Detection was performed using Amersham ECL Western
Blotting Detection Reagents for chemiluminescence (GE
Journal of Toxicology3
Figure 1: Engineered recombinant TcdB proteins. Native TcdB contains a glucosyltransferase domain (GT), a cysteine-protease domain
(CPD), a translocation domain (TD), and a receptor-binding domain (RBD) as shown. The AGT-tag coding DNA was appended to the
amino terminus in frame with the full-size TcdB coding DNA to create the AGT-TcdB expression vector. The TcdB RBD was replaced in
frame with the full-size BoNT/A heavy chain carboxyl terminus (BoNT/A-Hc, amino acids 861–1296), containing the receptor-binding
domain for BoNT/A, to generate AGT-TcdB-BoNT/A-Hc. The sizes of the boxes are approximately proportional to the sizes of the protein
domains. Restriction sites used in preparing the constructions are indicated.
2.4. AGT-Tag Labeling with Biotin. AGT fused to TcdB was
labeled in vitro with biotin in the absence of DTT, according
to the manufacturer’s instructions (New England Biolabs).
Briefly, 2μM AGT-TcdB was mixed with 3mM BG-biotin
(AGT substrate labeled with biotin) in a 25μL reaction
and incubated overnight at 4◦C. Biotin-labeled AGT-TcdB
was analyzed by Western blot using streptavidin conjugated
to horseradish peroxidase (HRP). DTT was not added in
the labeling reaction as recommended, because it has been
reported that DTT mimics intracellular activation of the
2.5. Cytotoxicity Assay. Cell lines at semiconfluence were
treated with purified TcdB, AGT-TcdB, or AGT-TcdB-
BoNT/A-Hc in 5x serial dilutions starting from 100ng/mL.
Cells were cultured for a period of 24h, and morphological
toxicity was quantified as the percentage of rounded cells per
3.1. Construction and Expression of AGT-TcdB and AGT-
TcdB-BoNT/A-Hc. To test the potential of TcdB-based vec-
tors for delivery of biomolecules to the cytosol of neuronal
cells, DNAs were created that encode two chimeric forms
of TcdB fused to a C-terminal His6-tag (see diagrammatic
representations in Figure 1). In the AGT-TcdB construct,
an alkylguanine-DNA alkyltransferase, referred to as AGT,
was appended to the TcdB N-terminus in frame with the
GT coding DNA. Another construct was also prepared in
which the GT was replaced by AGT, but this failed to yield
meaningful amounts of full-size protein apparently because
of proteolysis (data not shown) and was not further pursued.
For the second chimeric toxin, AGT-TcdB-BoNT/A-Hc,
the putative receptor-binding domain (RBD) from AGT-
TcdB was replaced in frame with the well-defined BoNT/A
receptor-binding domain [20, 21] designated BoNT/A-Hc.
The recombinant TcdB chimeric proteins were expressed in
B. megaterium and purified by nickel affinity as previously
described for wild-type TcdB .
3.2. Expressed AGT-TcdB Retains AGT Enzymatic Activity.
Recombinant AGT-TcdB was expressed, and the puri-
fied protein had the expected molecular weight. Western
blots with polyclonal anti-TcdB serum recognized both
the parental TcdB and AGT-TcdB, while AGT antiserum
recognized only the AGT-TcdB (Figure 2). To confirm the
proper folding and function of the AGT fusion partner, the
enzymatic activity of the AGT alkyltransferase was tested.
AGT-TcdB was incubated with BG-biotin which catalyzes
the covalent linkage of biotin to AGT. The efficiency of in
vitro AGT-protein labeling is generally ∼95% according to
the manufacturer (New England Biolabs). Western blotting
with streptavidin demonstrated that AGT-TcdB became
biotinylated following incubation with BG-biotin (Figure 2),
thus demonstrating that the AGT domain in recombinant
AGT-TcdB fusion retained enzymatic activity.
3.3. TcdB Delivers Glucosyltransferase Fusion Protein Cargo to
the Cell Cytosol. Cytosolic delivery of the GT domain is a
requirement for cytotoxicity [22, 23]. Since each of the TcdB
chimeric proteins (Figure 1) contain a fully functional GT
domain, delivery of the AGT/GT fusion cargo into the cell
cytosol could thus be assessed by measuring cell toxicity. In
a dose-response toxicity assay in Vero cells, AGT-TcdB was
found to retain a potency nearly equal to wild-type TcdB
(Figure 3(a)). Cells exposed to the lowest toxic dose of AGT-
TcdB were somewhat slower to become rounded than when
exposed to TcdB indicating that the presence of the AGT
partner may cause a small delay in toxin entry (Figure 3(b)).
Even so, the half maximal effective concentration (EC50) is
virtually identical for both toxin forms. Enzymatic labeling
of AGT-TcdB with BG-biotin did not cause any significant
4Journal of Toxicology
Figure 2: AGT-tag expressed at the N-terminus of TcdB retains enzymatic activity. Recombinant TcdB and AGT-TcdB were expressed and
and by anti-TcdB sera (middle). The positions of molecular weight markers are indicated with arrows. Each of the protein preparations was
subjected to enzymatic reactions with BG-biotin and analyzed for bound biotin by Western blot with streptavidin (right). The data shown
are representative of 2 independent experiments.
Cell rounding (%)
Toxin concentration (ng/mL)
Cell rounding (%)
Figure 3: TcdB with an N-terminal AGT-tag retains cytotoxic potency. (a) AGT-TcdB potency as a function of protein concentration. Vero
cells were treated with serial dilutions of TcdB and AGT-TcdB ± autobiotinylation. The dilution series started at 100ng/mL and continued
with fivefold serial dilutions. The % cell rounding was assessed after 24hr for each concentration tested. (b) AGT-TcdB potency as a function
of time after exposure. Vero cells were exposed to 0.8ng/mL of TcdB or AGT-TcdB, and the % cell rounding was assessed hourly for five
hours and then after 24hr. The data shown in both (a) and (b) are representative of 3 independent experiments.
decrease in toxin potency (Figures 3(a) and 3(b)). We con-
clude that the AGT-GT fusion protein with fully functional
with an efficacy nearly that of GT alone, thus demonstrating
the potential of TcdB-based vectors to function as cytosolic
3.4. Replacing the TcdB Receptor-Binding Domain with the
Equivalent Domain from BoNT/A Increases Neuronal Cell
Toxicity. Recombinant AGT-TcdB-BoNT/A-Hc protein was
expressed from an expression vector in which the TcdB
RBD coding region from AGT-TcdB was replaced by DNA
encoding the BoNT/A RBD, the carboxyl 50kDa portion of
TcdB for improved entry into neuronal cells and reduced
entry into nonneuronal cells. The AGT-TcdB-BoNT/A-Hc
preparation, purified only by nickel affinity, contained a
protein of the predicted molecular weight for the full-size
protein (∼280KDa) as well as some lower-molecular-weight
species that likely result from both protein degradation and
protein contamination (Figure 4(a)). Nevertheless, Western
blot analysis confirmed that the 280kDa AGT-TcdB-
BoNT/A-Hc protein was full size as it stained with both
anti-AGT and anti-BoNT/A-Hc antibodies (Figure 4(b)).
Journal of Toxicology5
Figure 4: Recombinant expression of TcdB with a BoNT/A RBD. Recombinant AGT-TcdB, AGT-TcdB-BoNT/A-Hc, and BoNT/A-Hc
were expressed and purified. (a) Each preparation containing 250ng of protein was analyzed by SDS-PAGE and protein stain. (b) The
preparations were also analyzed by Western blots using anti-AGT serum, anti-TcdB serum, or anti-BoNT/A-Hc mAb as indicated. The
position of molecular weight markers is indicated with arrows.
Next, we tested the cytotoxicity of AGT-TcdB-BoNT/A-
Hc with AGT-TcdB on two neuronal cells lines, Neuro2A
and M17, and on Vero cells, a nonneuronal cell line highly
sensitive to TcdB [24, 25]. The presence of the BoNT/A-Hc
for both neuronal cell lines (Figure 5(a)). The EC50of AGT-
TcdB-BoNT/A-Hc was approximately 25-fold lower than
AGT-TcdB when assessed 24h following toxin exposure.
In contrast, the EC50 of AGT-TcdB-BoNT/A-Hc for Vero
cells was approximately 25-fold higher than AGT-TcdB.
The finding of AGT-TcdB-BoNT/A-Hc toxicity in Vero cells
some toxicity in the absence of the putative RBD [23, 25, 26].
The rounded phenotype of the neuronal cells exposed to
the two toxin forms was indistinguishable, as exemplified
in representative images of Neuro2A cells (Figure 5(b)).
Neuro2A cells exposed for 24h to 0.16ng/mL AGT-TcdB-
of AGT-TcdB caused no observable changes compared to
untreated cells. Even at 4ng/mL, AGT-TcdB induced only
50% rounding of Neuro2A cells (Figure 5(b)).
It is noteworthy that cell rounding was also observed
to occur more rapidly in the two neuroblastoma cell lines
following exposure to AGT-TcdB-BoNT/A-Hc compared
at three different concentrations clearly demonstrates that
it takes about 125-fold more AGT-TcdB to achieve ≥80%
cell rounding in 5hrs than with AGT-TcdB-BoNT/A-Hc
(Figure 6). These results suggest that TcdB enters and intoxi-
cates neuronal cells significantly more rapidly and efficiently
when the toxin contains the BoNT/A RBD in place of the
native TcdB RBD.
BoNT-mediated paralysis is caused by inhibition of neu-
rotransmission in poisoned neuronal cells. This blockage
is induced following delivery of the endopeptidase domain
(light chain) to neurons which then inactivates one or more
cytosolic proteins specifically required for neurotransmitter
release. While biomolecules that inhibit or eliminate the
BoNT light chain have been developed , the in vivo
delivery of these therapeutic agents to the cell cytosol of
intoxicated neurons to promote their recovery remains a
challenge. One viable option for delivery vehicles is to
reengineer clostridial toxins, which are already well evolved
for delivery of their enzymatic cargo to cell cytosol, such that
the toxicity is removed, while the ability to enter cells and
deliver cargo remains intact.
Here, we demonstrate that wild type TcdB can be engi-
neered as a delivery system for selective targeting of neuronal
cells. C. difficile toxins (TcdA and TcdB) have a major
advantage over other clostridial toxins as cytosolic delivery
vehicles for therapeutic biomolecules. These toxins naturally
carry their own protease (cysteine-protease domain) that
enzymatically cleaves and releases their cargo into the
cytosol, eliminating the need to engineer a mechanism that
6Journal of Toxicology
Cell rounding (%)100
Toxin concentration (ng/mL)
Toxin concentration (ng/mL)
Toxin concentration (ng/mL)
Cell rounding (%)
Cell rounding (%)
Figure 5: A TcdB chimera with the BoNT/A receptor-binding domain has increased specificity for neuronal cells. (a) Toxin potency of AGT-
TcdB and AGT-TcdB-BoNT/A-Hc on two neuronal cell lines and Vero cells. Potency was assessed by serial dilutions of purified recombinant
AGT-TcdB-BoNT/A-Hc (?) and AGT-TcdB (◦). The potency was determined on Neuro2A and M17 neuroblastoma cells and Vero cells.
Fivefold serial dilutions of the proteins were added to medium, and cells were monitored for toxicity by assessing the % cell rounding
after 24hr. Data shown are representative of 4 independent experiments. (b) Microscopic images of Neuro2A cells exposed to AGT-TcdB
or AGT-TcdB-BoNT/A-Hc. Representative images are shown of Neuro2A cells exposed for 24hr to 0.16ng/mL or 4ng/mL of either AGT-
TcdB-BoNT/A-Hc or AGT-TcdB, respectively.
permits this release. For example, it has been reported that
the GT domain of TcdA can be removed and replaced with
luciferase to generate a functional delivery vehicle capable
of translocating luciferase to the cytosol of target cells .
Previous studies have shown that BoNTs can also be adapted
as delivery vehicles, but unlike Tcds, BoNTs do not have
an autocatalytic domain to release the light chain into the
cell cytoplasm. Thus, in constructs in which the toxic light
chain was removed, it was necessary to insert a linker
that promotes a disulfide bond between the cargo and the
translocation domain to make possible cargo release into the
cytosol following disulfide bridge reduction in the endosome
[28, 29]. Such an approach may be less efficient or ineffective
for some therapeutic cargo.
In this work, we have engineered expression vectors
for two chimeric TcdB proteins in which a functional GT
domain remains in place. The strategy was to test the
TcdB ability to deliver cargo to the cytosol by measuring
the cytotoxic potency of the two chimeric proteins in
comparison to wild-type TcdB. This eliminates the need
for microscopic or fractionation methods to distinguish
between endosomal and cytosolic cargo. In AGT-TcdB, an
alkylguanine-DNA alkyltransferase, referred to as AGT, was
appended to the GT domain of wild-type TcdB. Our results
show that AGT was at least partially functional, since it
capable of intoxicating Vero cells nearly as efficiently as wild
type TcdB. Thus, we infer that AGT and GT domain were
delivered together to the cell cytosol, and that adding a cargo
to the N-terminus of the toxin did not interfere substantially
with GT domain translocation and activity. Earlier work had
shown that gluthatione-S-transferase (GST) could also be
appended to the wild-type TcdB and detected in the cytosol
of intoxicated cells, but the fusion toxin was not fully active,
and there was no data as to the efficiency of delivery or
that the GST fusion protein retained its enzymatic activity
Next, we tested whether TcdB could be engineered for
selective neuronal toxicity by exchanging the TcdB receptor-
binding domain (RBD) on AGT-TcdB with the equivalent
RBD from the neuronal-specific BoNT/A toxin, generating
AGT-TcdB-BoNT/A-Hc. Our results showed clearly that
Journal of Toxicology7
Cell rounding (%)
Cell rounding (%)
Figure 6: A TcdB chimera with the BoNT/A receptor-binding domain more rapidly elicits Neuro2A cell rounding. Neuro2A cells were
exposed to three different concentrations (20ng/mL, 4ng/mL, and 0.16ng/mL) of AGT-TcdB-BoNT/A-Hc or AGT-TcdB and the % cell
rounding was assessed hourly for 5hr. Data shown are representative of 2 independent experiments.
neuronal cells were at least 25-fold more sensitive to the toxic
effects of AGT-TcdB-BoNT/A-Hc than AGT-TcdB, and signs
of intoxication appeared more rapidly. The results imply
that it may be possible to engineer TcdB with specificity for
almost any cell type by replacing the RBD with a RBD that
binds to an appropriate receptor expressed on the target cell
population. It is interesting to speculate that TcdB-BoNT/A-
Hc chimeras may also be capable of transcytosis through
endothelial cells in the same manner as native BoNT can
feature may permit delivery of therapeutic agents via oral or
Although AGT-TcdB-BoNT/A-Hc showed enhanced
toxicity for neuronal cells, this chimera retained some tox-
icity for nonneuronal cells. Previous work has also reported
putative RBD and suggested that TcdB internalization into
host cells may be mediated by additional receptor-binding
regions that are outside of the CROP domain of the toxin
[25, 26]. Indeed, a recent report  has shown that deletion
of the C-terminal region (residues 1500–1851) of putative
translocation domain (TD) reduces cell toxicity without
greatly affecting GT translocation and function. The authors
conclude that this portion of the putative TD likely contains
additional RBD properties. The results thus indicate that
deletion or replacement of residues contributing to receptor
binding within the C-terminus of TcdB putative TD will
likely reduce or eliminate non-neuronal cell intoxication by
TcdB-BoNT/A-Hc without loss of its capability to deliver
cargo to neuronal cells.
In addition to engineering TcdB with AGT fused to the
amino terminus of the GT domain, we also prepared a TcdB
this fusion protein apparently produced an unstable fusion
protein, resulting in very low yields of purified full-size
protein. One possibility is that the removal of GT resulted
in an altered conformation of the protein that resulted in
activation of the cysteine protease domain which then led
to autoproteolysis. The results suggest it may be important
to retain the GT domain in order to maintain stability of the
the GT enzymatic activity to render the glucosyltransferase
toxicity. Towards this goal, we recently produced an atoxic
and safe TcdB, by introducing two point mutations in the
GT domain (Haiying Wang and Hanping Feng, unpublished
data). We expect that TcdB-based vehicles with its cargo
fused to the atoxic GT domain will have better stability and
perhaps lead to a product having more native conformation,
thus resulting in a more efficient endosomal uptake and
cargo translocation to the cytosol than TcdB vehicles lacking
the GT domain. Thus, in future studies, we are opting to
append foreign coding DNA to a mutated TcdB-BoNT/A-Hc
rather than to replace the GT domain.
In summary, this study strongly suggests it will be
possible to engineer TcdB as a cytosolic delivery vehicle
of therapeutic cargo to neuronal cells, and perhaps other
cell types, by redirecting its cellular binding specificities.
As an example, we recently developed a small biomolecule
that specifically inhibits BoNT/A protease and promotes its
rapid degradation , and a mutated TcdB containing the
8Journal of Toxicology
BoNT/A-Hc may permit specific delivery of this therapeutic
cargo to neurons as an antidote for botulism.
The authors thank Weijia Nie and Amelie Debrock for tech-
BoNT/A-Hc, Dr. Jean Mukherjee for the anti-BoNT/A-Hc
mAb, and Dr. Rebekka Biedendieck for the MS941 strain.
They also thank Dr. Saul Tzipori for his early and continuing
support of this project and for helpful discussions. This
project was funded with federal funds from the NIAID,
NIH, DHHS, under Contract no. N01-AI-30050, and awards
R21AI088489 and R01AI088748. The content is solely the
responsibility of the authors and does not necessarily
and Infectious Diseases or the National Institutes of Health.
 T. Jank and K. Aktories, “Structure and mode of action of
clostridial glucosylating toxins: the ABCD model,” Trends in
Microbiology, vol. 16, no. 5, pp. 222–229, 2008.
 M. Egerer, T. Giesemann, T. Jank, K. J. Fullner Satchell, and K.
A and B depends on cysteine protease activity,” Journal of
Biological Chemistry, vol. 282, no. 35, pp. 25314–25321, 2007.
 T. Giesemann, M. Egerer, T. Jank, and K. Aktories, “Processing
vol. 57, no. 6, pp. 690–696, 2008.
 M. Egerer, T. Giesemann, C. Herrmann, and K. Aktorles,
“Autocatalytic processing of Clostridium difficile toxin B:
binding of inositol hexakisphosphate,” Journal of Biological
Chemistry, vol. 284, no. 6, pp. 3389–3395, 2009.
 D. E. Voth and J. D. Ballard, “Clostridium difficile toxins:
Reviews, vol. 18, no. 2, pp. 247–263, 2005.
 X. He, X. Sun, J. Wang et al., “Antibody-enhanced, Fc γ
receptor-mediated endocytosis of Clostridium difficile toxin
A,” Infection and Immunity, vol. 77, no. 6, pp. 2294–2303,
 J. Steele, H. Feng, N. Parry, and S. Tzipori, “Piglet models
of acute or chronic Clostridium difficile illness,” Journal of
Infectious Diseases, vol. 201, no. 3, pp. 428–434, 2010.
 G. Yang, B. Zhou, J. Wang et al., “Expression of recombinant
Clostridium difficile toxin A and B in Bacillus megaterium,”
BMC Microbiology, vol. 8, article 192, 2008.
 L. L. Simpson, “Identification of the major steps in botulinum
toxin action,” Annual Review of Pharmacology and Toxicology,
vol. 44, pp. 167–193, 2004.
tages and limitations of individual botulinum neurotoxins,”
Trends in Neurosciences, vol. 28, no. 8, pp. 446–452, 2005.
 C. Verderio, O. Rossetto, C. Grumelli, C. Frassoni, C. Monte-
cucco, and M. Matteoli, “Entering neurons: botulinum toxins
and synaptic vesicle recycling,” EMBO Reports, vol. 7, no. 10,
pp. 995–999, 2006.
 J. J. Schmidt, R. G. Stafford, and K. A. Bostian, “Type A
botulinum neurotoxin proteolytic activity: development of
competitive inhibitors and implications for substrate speci-
ficity at the S1’ binding subsite,” FEBS Letters, vol. 435, no. 1,
pp. 61–64, 1998.
 J. M. Tremblay, C. L. Kuo, C. Abeijon et al., “Camelid single
agents and inhibitors of Clostridium botulinum neurotoxin
(BoNT) proteases,” Toxicon, vol. 56, no. 6, pp. 990–998, 2010.
neurotoxin by binding to the non-catalytic α-exosite binding
region,” Journal of Molecular Biology, vol. 397, no. 4, pp. 1106–
 Y. C. Tsai, R. Maditz, C. L. Kuo et al., “Targeting botulinum
neurotoxin persistence by the ubiquitin-proteasome system,”
Proceedings of the National Academy of Sciences of the United
States of America, vol. 107, no. 38, pp. 16554–16559, 2010.
 C. L. Kuo, G. A. Oyler, and C. B. Shoemaker, “Accelerated neu-
targeted ubiquitination,” PLoS ONE, vol. 6, no. 5, 2011.
 A. Keppler, S. Gendreizig, T. Gronemeyer, H. Pick, H. Vogel,
and K. Johnsson, “A general method for the covalent labeling
of fusion proteins with small molecules in vivo,” Nature
Biotechnology, vol. 21, no. 1, pp. 86–89, 2003.
 H. F. LaPenotiere, M. A. Clayton, and J. L. Middlebrook,
“Expression of a large, nontoxic fragment of botulinum
neurotoxin serotype A and its use as an immunogen,” Toxicon,
vol. 33, no. 10, pp. 1383–1386, 1995.
 M. C. Shoshan, T. Bergman, M. Thelestam, and I. Florin,
“Dithiothreitolgenerates anactivated 250,000mol.wtformof
Clostridium difficile toxin B,” Toxicon, vol. 31, no. 7, pp. 845–
 G. Lalli, J. Herreros, S. L. Osborne, C. Montecucco, O.
Rossetto, and G. Schiavo, “Functional characterisation of
of Cell Science, vol. 112, no. 16, pp. 2715–2724, 1999.
 J. B. Park and L. L. Simpson, “Inhalational poisoning by
botulinum toxin and inhalation vaccination with its heavy-
chain component,” Infection and Immunity, vol. 71, no. 3, pp.
 G. Pfeifer, J. Schirmer, J. Leemhuis et al., “Cellular uptake of
Clostridium difficile toxin B. Translocation of the N-terminal
Biological Chemistry, vol. 278, no. 45, pp. 44535–44541, 2003.
R. Benz, and K. Aktories, “Structural determinants for
membrane insertion, pore formation and translocation of
Clostridium difficile toxin B,” Molecular Microbiology, vol. 79,
no. 6, pp. 1643–1654, 2011.
 A. Sundriyal, A. K. Roberts, R. Ling, J. McGlashan, C. C.
Shone, and K. R. Acharya, “Expression, purification and cell
cytotoxicity of actin-modifying binary toxin from Clostridium
difficile,” Protein Expression and Purification, vol. 74, no. 1, pp.
 T. Dingle, S. Wee, G. L. Mulvey et al., “Functional properties
of the carboxy-terminal host cell-binding domains of the two
toxins, TcdA and TcdB, expressed by Clostridium difficile,”
Glycobiology, vol. 18, no. 9, pp. 698–706, 2008.
 L. A. Barroso, J. S. Moncrief, D. M. Lyerly, and T. D. Wilkins,
“Mutagenesis of the Clostridium difficile toxin B gene and
4, pp. 297–303, 1994.
 S. M. Kern and A. L. Feig, “Adaptation of Clostridium difficile
toxin A for use as a protein translocation system,” Biochemical
and Biophysical Research Communications, vol. 405, no. 4, pp.
Journal of Toxicology9 Download full-text
 P. Zhang, R. Ray, B. R. Singh, D. Li, M. Adler, and P. Ray, “An
efficient drug delivery vehicle for botulism countermeasure,”
BMC Pharmacology, vol. 9, article 12, 2009.
 M. Ho, L.-H. Chang, M. Pires-Alves et al., “Recombinant
botulinum neurotoxin A heavy chain-based delivery vehicles
for neuronal cell targeting,” Protein Engineering, Design and
Selection, vol. 24, no. 3, pp. 247–253, 2011.
the botulinum neurotoxin molecule that govern binding and
transcytosis across polarized human intestinal epithelial cells,”
Journal of Pharmacology and Experimental Therapeutics, vol.
310, no. 2, pp. 633–641, 2004.