Báo cáo khoa học: Engineering toxins for 21st century therapies docx

The meeting was assembled to discuss the design, manufacture and regulatory consid-erations of developing novel therapies that utilize toxin domains, and to discuss the protein engineeri

Engineering toxins for 21st century therapies

John A Chaddock1and K Ravi Acharya2

1 Syntaxin Limited, Abingdon, Oxon, UK

2 Department of Biology and Biochemistry, University of Bath, UK

Introduction

It is a paradox of drug development that nature’s most

powerful toxins can also be the active component of

some of the most effective therapies for a range of

conditions For example, botulinum toxins produced

by the genus Clostridia have the ability to cause

botu-lism and tetanus following exposure to extremely small

doses of protein Conversely, botulinum neurotoxins

(BoNTs, e.g BOTOX, Dysport; see Table 1, [1])

have become the first-line treatment for a variety of

debilitating neuromuscular conditions; for example,

various dystonia, spasmodic torticollis, blepharospasm

and strabismus Native-sourced botulinum products

have also been approved for use in hyperhidrosis and,

most recently, in chronic migraine Similarly, up to 10%

of diphtheria patients die (even if properly treated) and yet components of diphtheria toxin have been used

to create successful new medicines (e.g ONTAK; a CD25-directed cytotoxin) for the treatment of a range

of cancers, such as persistent or recurrent cutaneous T-cell lymphoma

In September 2010, a small, focussed meeting was convened as part of the Royal Society (UK) Interna-tional Seminar series The meeting was assembled to discuss the design, manufacture and regulatory consid-erations of developing novel therapies that utilize toxin domains, and to discuss the protein engineering

Keywords

biotechnology; botulinum neurotoxin;

innovation; therapy; toxin

Correspondence

K R Acharya, Department of Biology and

Biochemistry, University of Bath, Claverton

Down, Bath BA2 7AY, UK

Fax: +44 1225 386779

Tel: +44 1225 386238

E-mail: bsskra@bath.ac.uk

J A Chaddock, Syntaxin Limited, Units

4–10, Barton Lane, Abingdon, Oxon OX14

3YS, UK

Fax: +44 1235 552200

Tel: +44 1235 552115

E-mail: john.chaddock@syntaxin.com

(Received 5 November 2010, revised

21 December 2010, accepted 10 January

2011)

doi:10.1111/j.1742-4658.2011.08013.x

‘Engineering Toxins for 21st Century Therapies’ (9–10 September 2010) was part of the Royal Society International Seminar series held at the Kavli International Centre, UK Participants were assembled from a range

of disciplines (academic, industry, regulatory, public health) to discuss the future potential of toxin-based therapies The meeting explored how the current structural and mechanistic knowledge of toxins could be used

to engineer future toxin-based therapies To date, significant progress has been made in the design of novel recombinant biologics based on domains

of natural toxins, engineered to exhibit advantageous properties The meet-ing concluded, firstly that future product development vitally required the appropriate combination of creativity and innovation that can come from the academic, biotechnology and pharma sectors Second, that continued investigation into understanding the basic science of the toxins and their targets was essential in order to develop new opportunities for the existing products and to create new products with enhanced properties Finally, it was concluded that the clinical potential for development of novel biologics based on toxin domains was evident

Abbreviations

BoNT, botulinum neurotoxin; LC, light chain; TSI, targeted secretion inhibitor.

opportunities that exist for developing new medicines

that harness components of some of nature’s most

potent protein toxins These objectives were

high-lighted by K Ravi Acharya (organizer of the meeting,

University of Bath, UK) in his introduction to the

meeting

The nature of the meeting was such that the majority

of the discussion was related to clostridial neurotoxins

(CNTs), specifically BoNT This is primarily because

BoNTs have emerged from a physician-led

investiga-tional drug in the 1980s to become a multi-billion

dollar product with a range of medical and cosmetic

applications Nevertheless, many of the concepts and

conclusions that were described during the meeting

have relevance to other similar toxins that may have

therapeutic utility The participants spent a

consider-able amount of time evaluating how toxin structure–

function information and understanding their unique

mechanisms of action can provide new opportunities

for the development of therapeutic interventions

Clostridial neurotoxins are multi-domain structures

[2] of  150 kDa that consist of five major structural

elements (Fig 1):

l An N-terminal 50 kDa light chain (LC) domain is

a metalloprotease with specificity for SNARE pro-tein substrate To the C-terminus of the LC is;

l An HN domain ( 50 kDa) that forms a pore in intracellular membranes to effect translocation of the LC into the cytosol The HN domain is cova-lently attached to the LC by;

l A single disulfide bond that is reduced in the cyto-sol as part of the LC translocation event;

l An HCN domain ( 25 kDa) that is C-terminal to the HN domain and is of unknown function The

HCN domain is a subdomain of the ‘binding domain’ formed with;

l An HCC domain ( 25 kDa) that exhibits neuronal binding capability via two binding sites

Although the precise ‘shape’ of other bacterial pro-tein toxins, for example diphtheria toxin, can differ markedly from clostridial BoNTs [BoNTs with seven different serotypes (BoNT⁄ A–BoNT ⁄ G)], many have been shown to be similar in concept: the delineation of

a ‘binding’ domain, a ‘translocation’ domain and a

‘catalytic’ domain is common This modular, domain-based structure has enabled drug development scien-tists to implement protein engineering approaches to create novel proteins that harness specific biology of the ‘parent’ toxins The most advanced demonstration

of this concept is exemplified in targeted secretion inhibitors (TSIs); a novel class of biotherapeutics for treating diseases where inappropriate cell secretion is a primary cause Concepts such as the TSI platform and the similar advancements that have been made by sci-entific advancement in this ‘toxin’ field are described in the meeting highlights below

Meeting highlights This meeting brought together leaders in their respec-tive fields and major scientific and conceptual advance-ments are described in the following section There are, however, a number of highlights that require specific mention:

l Academic laboratories and small biotechnology com-panies are rich sources of creativity and entrepreneurial

Fig 1 Tertiary structure of BoNT ⁄ A (pdb code 3BTA [8]).

Table 1 Summary of botulinum toxin products (taken from [1]).

Approvals In over 75 countries In over 65 countries Some EU + USA + CA Some EU, Mexico, Argentina Active substance Type A complex

(900 kDa)

Type A complex (900 kDa)

Type B complex (700 kDa) Type A, free from complexing

proteins (150 kDa)

drive; pharmaceutical industry has the infrastructure to

innovate, i.e apply the new ideas in the market place,

and it is vital that these two approaches to drug

dis-covery and development are brought together

Exam-ples of creativity in the design of toxin fragment-based

therapeutics included next-generation Ontak, TSIs,

BoNT hybrids and BoTIMs, TrapoX

l Investment in basic science research is an essential

requirement for the development of new concepts and

new opportunities Throughout the meeting there were

multiple examples of where opportunities for the

devel-opment of new medicines could emerge from scientific

exploration into fundamental biological mechanisms

l Small, focussed meetings of this type are an

excel-lent forum for cross-discipline discussion and learning

Major achievements

A number of advancements in scientific knowledge or

conceptual thought emerged from the meeting, which

have been collated into four themes:

Understanding protein structure–function can

lead to the development of new medicines

In one of the most advanced examples of developing

engineered toxins for the clinic, John Murphy (Boston

University School of Medicine, USA) noted that

ON-TAK (a diphtheria toxin-interleukin-2 fusion protein

[3]) had successfully completed studies in

steroid-resis-tant graft versus host disease and refractory T-cell

lym-phoma, but commented that the major adverse event is

vascular leak syndrome Murphy provided

experimen-tal evidence for the identification and elimination of

peptide motifs in diphtheria toxin that promote

vascu-lar leak syndrome, which could be the basis of

devel-oping a drug with similar efficacy but a much reduced

adverse event profile, thereby widening the therapeutic

opportunities for use of the diphtheria toxin cell

ablation technology

The advances in structural knowledge in the BoNT

field (as summarized by Subramanyam Swaminathan,

Brookhaven National Laboratory, USA), has led to a

large number of opportunities to manipulate those

domains that nature has brought together as BoNT,

for the creation of new therapeutic opportunities In

particular, Keith Foster (Syntaxin Ltd, UK) described

the development of a platform of new biologicals

based on the LC and HN domain of BoNT termed

TSIs TSIs do not possess the native BoNT binding

domain within their structure and therefore can be

retargeted to any cell of choice by incorporation of an

appropriate ligand (peptide or protein) to a cell surface

marker Although the precise nature of the range of targeting ligands built into TSIs was not disclosed, the TSI approach takes SNARE cleavage beyond the neu-ronal target to alternative cell types In this way, the SNARE cleavage activity of the LC can be redirected

to cells that are secreting mediators that cause disease Foster described the progress of TSIs into phase I clin-ical trials in pain and advanced preclinclin-ical studies in acromegaly

In contrast to the domain replacement path taken

by Syntaxin Ltd, Andreas Rummel (Medizinische Ho-chschule, Hannover, Germany) described how the study of the mechanism of binding by the native neu-rotoxins had led to the development of TrapoX, a BoNT⁄ A-based protein that incorporates the latest understanding of binding to increase the potency of BoNT⁄ A Rummel reported that TrapoX (which has a specific mutation of the HC binding site) has more than three-fold increased potency and could, therefore, lead to lower therapeutic dosages of BoNT In a sec-ond example of engineering the native BoNT structure, Oliver Dolly (Dublin City University, Ireland) reported the construction of ‘BoTIMs’ (full-length BoNTs incorporating catalytic-inactive LC⁄ A), which were recombinantly fused to LC⁄ E domains to create a hybrid construct that utilized components within the

LC⁄ A element to extend the intracellular persistence of the LC⁄ E and therefore the duration of action of

LC⁄ E-induced SNAP-25 cleavage Dolly proposed that the LC⁄ E-induced cleavage of SNAP-25 would

be advantageous for specific conditions, for example pain Finally, studies by Joseph Barbieri (Medical College of Wisconsin at Milwaukee, USA) have led to the identification of a new binding loop in BoNT⁄ C and⁄ D and the observation that TeNT, BoNT ⁄ C and BoNT⁄ D enter cortical neurons via activity-independent endocytosis

Advances in the understanding of toxin fragment translocation

A leader in the field of membrane protein transloca-tion, Mauricio Montal (University of California, San Diego, CA, USA), discussed the impact of the BoNT

HC domain on the pH dependency of translocation The established dogma states that low pH is essential for HN domain insertion into the endosomal mem-brane in order to form the pore for LC translocation Using a precise membrane conductance assay he noted that no pH gradient was required for LHA (a frag-ment of BoNT⁄ A comprising the LC and the HN domain) translocation, whereas BoNT⁄ A required

a pH of  5 for efficient LC translocation Montal

hypothesized that the HC domain acts as a chaperone

for the LC; restricting membrane insertion until

local-ized into an acidic endosome Because this effect was

seen in simple lipid bilayers, it is probably a structural

effect rather than one based on binding Montal also

confirmed the importance of the disulfide bond to

ensure LC translocation through the HNpore

In the field of diphtheria toxin, Murphy described

data that indicated binding of diphtheria toxin to the

COPI complex via KXKXX sequence motifs in the

transmembrane domain of diphtheria toxin He

hypothesized that this binding event provides the drive

for the translocation process of the diphtheria toxin

C-domain across the intracellular vesicle membrane,

and made the staggering observation that the same

KXKXX signals for this process are present in anthrax

lethal factor and elongation factor The discussion on

this topic addressed the question of whether there is a

common motif within bacterial protein toxins that

facilitates the translocation process by binding to host

cell machinery Murphy reported that a detailed

under-standing of the amino acid requirements for the

trans-location event had led to an experimentally testable

opportunity to deliver nucleic acids into the cytosol for

cell modulation Murphy also noted the role of Grp78

and hypothesized that such a vesicle-located unfoldase

may be involved in unfolding the catalytic domains of

bacterial protein toxins prior to translocation

Although immunogenicity was not a key topic of

the meeting, Jim Marks (University of California and

San Francisco General Hospital, USA) noted that

antibodies raised to the HN domain of BoNT were

effective inhibitors of function and Montal noted that

one specific antibody raised to a conserved epitope

across serotypes did inhibit membrane insertion of the

HNdomain

Understanding the biology of toxin action can

provide new ideas for medicine development

Thierry Galli (INSERM, France) reported that

BoNT⁄ D inhibited fast endocytosis in a primary dorsal

root ganglion neuronal culture system, thereby

impli-cating VAMP2 to have a role in endocytosis Also,

expression of TeNT in epithelial cells (containing

VAMP3) prevents the recycling of integrins Hence,

BoNTs have a role in the disruption of endocytosis

and endocytic mechanisms, not just exocytosis

George Oyler (Synaptic Research LLC, USA)

described the development of a designer VHH E3

ubiquitin ligase that is able to degrade LC⁄ A The

VHH structure is the antigen binding fragment of

cam-elid heavy chain antibodies If suitably delivered to

BoNT-intoxicated neurons, such a protein would have the potential to degrade the ordinarily persistent LC⁄ A and thereby accelerate recovery from BoNT⁄ A poison-ing However, the potential for this approach is more widespread and could provide a mechanism for the delivery of ubiquitin ligases for the modification of cellular function per se

Giampietro Schiavo (Cancer Research UK London Research Institute, UK) described elegant studies into axonal transport of CNTs and proposed the use of labelled CNTs as diagnostic markers for neuronal dis-ease, e.g motor neurone disdis-ease, utilizing live body imaging to assess the speed of axonal transport Praveen Anand (Imperial College London, UK) described the wide range of toxins currently employed

in providing medical benefit (e.g capsaicin, resinifero-toxin, ziconotide, CNTs, chemotherapeutic agents) By understanding the expression of SV2A in neuronal samples from patients, Anand was able to identify potential and preferential sensory targets for BoNT therapy, including painful nerve injury, inflammatory bowel disease and irritable bowel syndrome

Development of new medicines requires a new mindset, a new model

Melanie Lee (Syntaxin Ltd, UK) opened the meeting with a discussion of the rise and fall of the fully inte-grated pharmaceutical company model, pharma’s resis-tance to change and blindness to entrepreneurial opportunities Acknowledging that the pharmaceutical industry faced many challenges to their pipeline and profitability (research and development, patent expiry, generics, high attrition rates), Lee commented that the pharmaceutical industry has evolved away from prod-uct innovation to focused process innovation and yet the pipelines required creativity, which in turn requires overcoming the major hurdles of new knowledge advancement William Habig (former member of the Food and Drug Administration, USA) observed that there was generally a lower success rate for small com-panies⁄ first applicants to the Food and Drug Adminis-tration and added that only 30% of approved pharmaceuticals recover the cost of their development Murphy also noted the negative impact on product optimization when a small biotechnology company develops innovative medicines and funds are limited for extensive basic science investment

Emerging trends and future directions One of the clear outputs of the meeting was an acknowledgement that functional domains from toxins

can form the basis for new medicine development.

Engineered toxins take advantage of the natural

selec-tion process that has already created protein structures

with specific functions that can modulate intracellular

events In the final session of the meeting, John

Chad-dock (Syntaxin Ltd, UK) captured the immediate

opinion of the participants regarding the major themes

that had emerged from the meeting These flipchart

scribbles are reproduced in Table 2 and represent a

snapshot of the opinions at the time

In preparing this meeting summary, many of the

ideas and thoughts for the future have been

crystal-lized into trends and future direction statements

(Table 2) The general consensus of the meeting was

that there is an opportunity for ‘science-driven

evolu-tion’, i.e using in-depth knowledge of the fine details

of toxin structure–function in combination with a

sig-nificantly enhanced understanding of the biology of

the target cell For example, in the botulinum toxin

opportunities space, the participants appreciated the

importance of understanding the diversity of the

neu-ronal system and the detail of biological processes

therein For this to occur, there is an absolute

require-ment for continued basic scientific research There is

also the need to combine the talents of the academic

and entrepreneurial biotechnology sector with the

resources of pharma that are necessary to innovate

Mechanistically, targeted toxins were known to have

multiple medical applications based on their ability to

affect: (a) secretion⁄ exocytosis ⁄ endocytosis through

SNARE cleavage, (b) cell viability through modifica-tion of essential cellular pathways, (c) general cellular modulation by facilitating delivery of protein⁄ nucleic acid cargo It was agreed that targeted toxins could have diagnostic potential As noted above, naturally evolved toxins are a good framework for the design of cell modulation technologies, as they often target molecular mechanisms at the heart of disease condi-tions A challenge to the community is to change the

‘reputation’ of clostridial (and other) toxins from ‘dan-gerous’ to be accepted as unique and effective medi-cines and diagnostic tools for the future (Fig 2)

The wider context Although not a topic that was specifically discussed at the meeting, it is useful to appreciate the wider con-text of the use of toxins and toxin fragments for the development of medical products, vaccines etc In addition to the toxins described within the meeting, other researchers have, for a number of years, devel-oped novel molecule-based ribosome inactivating pro-teins (such as ricin, saporin and Shiga toxin) and ADP-ribosylating bacterial toxins such as Pseudomo-nas exotoxin All of these approaches are linked by their desire to utilize warheads that lead to cell death The most prevalent target for such novel molecules has been for the treatment of cancer and proliferative diseases For example, the literature is rich with exam-ples of fragments of Pseudomonas exotoxin targeted

to cancer cells via a range of targeting ligands and antibodies [4] Such molecules have demonstrated some significant success in preclinical studies and have

Table 2 Captured participant opinions on seminar themes.

The LC protease in the cell is not well understood and could have

a range of properties that are poorly understood and poorly

predicted from simple in vitro experiments

The role of the glycolipids and glycoproteins is greater than

previously understood Also, we need tools to better understand

glyco-contributions

It is difficult to translate ideas into products

Targeting ubiquitin-like molecules is a therapeutic opportunity

There is an ability to engineer proteins based on structural

knowledge: science-driven evolution

The neuronal system is highly diverse and there are many

unknowns

There is a need for new tools to enable better understanding of

biological systems

There is an appreciation of the increased number of applications for

nonserotype A-based products

Custom-designed molecules are a possibility, based on a better

understanding of an individual’s specific needs and the

opportunities to tailor the products Opportunities for selected

subpopulation treatment (at reduced cost)

Interaction between the toxin and the target cell is an important

understanding

Fig 2 Key elements of developing toxin-based therapies.

given much hope to the possibility of targeted cellular

ablation technology being implemented within the

clinical setting However, despite over 20 years of

investigative studies, no such molecules comprising

Pseudomonas domains have yet transitioned through

clinical trials to the clinic Concerns have been raised

over antibody formation, hepatic toxicity and

nonspe-cific side-effects, such as vascular leak syndrome

Indeed, attempts to overcome potential immunological

responses to targeted toxins have led to the

develop-ment of agents based on human protein warheads, for

example proapoptotic proteins or RNase [5]

Alterna-tively, protein engineering solutions have been used to

create modified targeted toxins with reduced

immuno-genicity [6]

Some of the pioneering work towards the

implemen-tation of cell ablation technology in the clinic arose

through the use of ricin and domain fragments of ricin

targeted to specific cells by virtue of conjugation to an

appropriate antibody Ricin, and the

ribosome-inacti-vating class, kills cells by preventing protein synthesis

Ricin-based approaches were promoted as exemplars

of the ‘immunotoxin’ concept and, although very

effec-tive in the preclinical setting, did not translate into

widespread use in the clinic The main issues faced

included vascular leak syndrome (as described for the

ONTAK strategy earlier) and hepatic toxicity An

alternative strategy for the use of cell ablation

technol-ogy has been taken with the development of a

conju-gate of substance P-saporin [7] Using the substance P

peptide to target NK1 receptors on the extracellular

face of the cell membrane of pain-sensing neurons,

such molecules are being developed to ablate specific

neuronal populations Such an approach has the

potential to inhibit the symptoms of cancer pain, for

example in patient populations that have become

resis-tant to morphine and other opiates

After 30 years of research and development, where

is the next clinical targeted-toxin product? Clearly,

advancements have been made with the Pseudomonas

exotoxin and diphtheria toxin platforms that give hope

to successful completion of clinical studies and the expansion of the targeted toxin strategy In the botu-linum field, TSIs have entered the clinic, bringing a new, nonablation strategy that will complement the various cell-kill strategies Retargeting cell modifying proteins to target cells is entering an exciting phase of development

Acknowledgements KRA is supported by the Royal Society (UK) through

an Industry Fellowship and is grateful to the Royal Society for sponsoring this international seminar He also wishes to acknowledge Syntaxin Ltd (UK) for providing additional travel support for the partici-pants

References

1 Carruthers A & Carruthers J (2008) Botulinum toxin products overview Skin Therapy Lett 13, 1–4

2 Turton K, Chaddock JA & Acharya KR (2002) Botulinum and tetanus neurotoxins: structure, function and therapeutic utility Trends Biochem Sci 27, 552–558

3 Foss FM (2000) DAB(389)IL-2 (ONTAK): a novel fusion toxin therapy for lymphoma Clin Lymphoma 1, 110–116

4 Kreitman RJ (2006) Immunotoxins for targeted cancer therapy AAPS J 8, E532–E551

5 Mathew M & Verma RS (2009) Humanized immunotox-ins – a new generation of immunotoximmunotox-ins Cancer Sci 100, 1359–1365

6 Nagata S & Pastan I (2009) Removal of B cell epitopes

as a practical approach for reducing the immunogenicity

of foreign protein-based therapeutics Adv Drug Deliv Rev 61, 977–985

7 Wiley RG & Lappi DA (2003) Targeted toxins in pain Adv Drug Deliv Rev 55, 1043–1054

8 Lacy DB, Tepp W, Cohen AC, DasGupta BR & Stevens

RC (1998) Crystal structure of botulinum neurotoxin type A and implications for toxicity Nat Struct Biol 5, 898–902