In vivo neutralization of naturally existing antibodies against linear α (1,3)-galactosidic carbohydrate epitopes by multivalent antigen presentation: A solution for the first hurdle of pig-to-human xenotransplantation

Article abstract of DOI:10.2533/chimia.2010.23

Pig-to-human xenotransplantation of islet cells or of vascularized organs would offer a welcome treatment alternative for the ever-increasing number of patients with end-stage organ failure who are waiting for a suitable allograph. The main hurdle are pre

Full text of DOI:10.2533/chimia.2010.23

From ChemiCal researCh to industrial appliCationsꢀ
CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ1/2ꢀ 23
doi:10.2533/chimia.2010.23ꢀꢀ
Chimiaꢀ64ꢀ(2010)ꢀ23–28ꢀ ©ꢀSchweizerischeꢀChemischeꢀGesellschaft
In vivo Neutralization of Naturally
Existing Antibodies against Linear
a(1,3)-Galactosidic Carbohydrate Epitopes
by Multivalent Antigen Presentation:
A Solution for the First Hurdle of
Pig-to-Human Xenotransplantation
RudolfꢀO.ꢀDuthaler*a,ꢀBeatꢀErnstb,ꢀRetoꢀFischerc,ꢀAndreasꢀG.ꢀKatopodisa,ꢀWillyꢀKinzyd,ꢀ
WolfgangꢀMartererc,ꢀReinholdꢀOehrleine,ꢀMarkusꢀB.ꢀStreiffa,ꢀandꢀGebhardꢀThomaa
DedicatedꢀtoꢀProfessorꢀDanielꢀBellušꢀonꢀtheꢀoccasionꢀofꢀhisꢀ70thꢀbirthday
Abstract:ꢀPig-to-humanꢀxenotransplantationꢀofꢀisletꢀcellsꢀorꢀofꢀvascularizedꢀorgansꢀwouldꢀofferꢀaꢀwelcomeꢀtreat-
mentꢀalternativeꢀforꢀtheꢀever-increasingꢀnumberꢀofꢀpatientsꢀwithꢀend-stageꢀorganꢀfailureꢀwhoꢀareꢀwaitingꢀforꢀaꢀsuit-
ableꢀallograph.ꢀTheꢀmainꢀhurdleꢀareꢀpreexistingꢀantibodies,ꢀmostꢀofꢀwhichꢀareꢀspecificꢀforꢀ‘Linear-B’,ꢀcarbohydrateꢀ
epitopesꢀterminatedꢀbyꢀtheꢀunbranchedꢀGal-a(1,3)Galꢀdisaccharide.ꢀTheseꢀantibodiesꢀareꢀresponsibleꢀforꢀtheꢀ
‘hyper-acuteꢀrejection’ꢀofꢀtheꢀxenograftꢀbyꢀcomplementꢀmediatedꢀhemorrhage.ꢀForꢀdepletionꢀofꢀsuchꢀantibodiesꢀ
weꢀhaveꢀdevelopedꢀanꢀartificialꢀinjectableꢀantigen,ꢀaꢀglycopolymerꢀ(GAS914)ꢀwithꢀaꢀchargeꢀneutralꢀpoly-lysineꢀ
backboneꢀ(degreeꢀofꢀpolymerizationꢀnꢀ=ꢀ1000)ꢀandꢀ25%ꢀofꢀitsꢀsideꢀchainsꢀcoupledꢀtoꢀLinear-B-trisaccharide.ꢀWithꢀ
anꢀaverageꢀmolecularꢀweightꢀofꢀ400ꢀtoꢀ500ꢀkD,ꢀpresentingꢀ250ꢀtrisaccharideꢀepitopesꢀperꢀmolecule,ꢀthisꢀmultiva-
lentꢀarrayꢀbindsꢀanti-aGalꢀantibodiesꢀwithꢀatꢀleastꢀthreeꢀordersꢀofꢀmagnitudeꢀhigherꢀavidityꢀonꢀaꢀper-saccharideꢀ
basisꢀthanꢀtheꢀmonomericꢀepitope.ꢀIn vivoꢀexperimentsꢀwithꢀnon-humanꢀprimatesꢀdocumentedꢀthatꢀratherꢀlowꢀ
dosesꢀ–ꢀ1ꢀtoꢀ5ꢀmg/kgꢀofꢀGAS914ꢀinjectedꢀi.v.ꢀ–ꢀefficientlyꢀreduceꢀtheꢀloadꢀofꢀanti-Linear-Bꢀantibodiesꢀquicklyꢀbyꢀ
atꢀleastꢀ80%.ꢀThisꢀtreatmentꢀcanꢀbeꢀrepeatedꢀwithoutꢀanyꢀsensitizationꢀtoꢀGAS914.ꢀInterestingly,ꢀalthoughꢀtheꢀ
antibodyꢀlevelsꢀstartꢀraisingꢀ12ꢀhꢀafterꢀinjection,ꢀtheyꢀdoꢀnotꢀreachꢀpretreatmentꢀlevels.ꢀTheꢀpolymerꢀisꢀdegradedꢀ
andꢀexcretedꢀwithinꢀhours,ꢀwithꢀaꢀminuteꢀfractionꢀremainingꢀinꢀlymphoidꢀtissueꢀofꢀanti-aGalꢀproducingꢀanimalsꢀ
only,ꢀprobablyꢀbindingꢀtoꢀandꢀinhibitingꢀantibody-producingꢀB-cells.ꢀTheꢀresultsꢀofꢀpig-to-non-humanꢀprimateꢀ
xenotransplantationsꢀestablishedꢀGAS914ꢀasꢀaꢀrelevantꢀtherapeuticꢀoptionꢀforꢀpig-to-humanꢀtransplantationsꢀasꢀ
well.ꢀTheꢀsynthesisꢀofꢀGAS914ꢀwasꢀsuccessfullyꢀscaledꢀupꢀtoꢀkgꢀamountsꢀneededꢀforꢀfirstꢀclinicalꢀstudies.ꢀKeyꢀ
wasꢀtheꢀuseꢀofꢀgalactosylꢀtransferasesꢀandꢀUDP-galactoseꢀforꢀtheꢀsynthesisꢀofꢀtheꢀtrisaccharide.ꢀ
Keywords:ꢀCarbohydrateꢀantigensꢀ·ꢀEnzymaticꢀglycosylationꢀ·ꢀGAS914ꢀ·ꢀGlycopolymerꢀ·ꢀXenotransplantation
The availability of human organs for trans- carbohydrate epitopes, leading to comple-
plantation to treat end-stage organ failure ment activation and destruction of the or-
is limited to about 25% of patients on wait- gan by hemorrhage and thrombosis within
ing lists. This has led to a great interest in hours. These preexisting antibodies origi-
the possibilities of xenotransplantation nate from cross-immunization by bacterial
with the main focus on the pig as donor. flora upon colonization of the intestine
Although organs from the larger non-hu- after birth. The predominant antigen is re-
*Correspondence:ꢀDr.ꢀR.ꢀO.ꢀDuthalera
man primates would pose the least immu- lated to a carbohydrate epitope found on
nogenic barrier, ethical considerations as proteins and lipids of all mammals except
well as the high risk of cross-species virus primates.[2] This epitope – also called ‘Lin-
transmission preclude such an option.[1] ear B’ (1) – is terminated with an a(1,3)-
On the other hand, the immune reaction to- galactosidic linkage on an unbranched ga-
wards a pig-to-human xenograft includes lactose, i.e. the a(1,2)-linked fucose of hu-
all branches of the immune system, result- man blood group epitopes (2, 3) is missing
ing in hyper-acute rejection, acute vascular (Fig. 1). Such variations in glycosylation
rejection, and – after a period of accom- patterns are the result of species-specific
modation – cellular rejection as well as differences in glycosyl transferase genes.
chronic rejection. The first hurdle – the The fraction of preformed antibodies di-
hyper-acute rejection – is initiated within rected to these antigens widely differs be-
minutes upon contact with monoreactive tween individuals, and can reach 3% of all
naturally existing antibodies directed to preexisting antibodies.
Tel.:ꢀ+ꢀ41ꢀ61ꢀ324ꢀ31ꢀ38
E-mail:ꢀRudolf.duthaler@novartis.com
aNovartisꢀInstitutesꢀofꢀBiomedicalꢀResearchꢀ(NIBR)
NovartisꢀPharmaꢀAG
P.O.ꢀBox,ꢀCH-4002ꢀBasel
bPharmacenter,ꢀUniversityꢀofꢀBasel
Klingelbergstrasseꢀ50
CH-4056ꢀBasel
cTechnicalꢀResearchꢀandꢀDevelopmentꢀ(TRD)
NovartisꢀPharmaꢀAG
P.O.ꢀBox,ꢀCH-4002ꢀBasel
dF.ꢀHoffmann-LaꢀRocheꢀAG
HumanꢀResourcesꢀ–ꢀPSHR
Grenzacherstrasseꢀ124
CH-4070ꢀBasel
eBASF,ꢀTheꢀChemicalꢀCompany
P.O.ꢀBox,ꢀCH-4002ꢀBasel

24ꢀ CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ1/2ꢀ
From ChemiCal researCh to industrial appliCations
glycol scaffolds,[9] and with poly-acryl am-
ide conjugates[10] were made.
tion. One way to overcome this hurdle is
to breed transgenic animals expressing
factors regulating complement activation
such as ‘hDAF’, human decay accelerat-
ing factor,[1a,3] or by depleting the xeno-
reactive antibodies using extracorporeal
plasmapheresis with affinity columns.[4]
Another strategy is to ‘inhibit’ these nat-
urally existing antibodies by injection of
carbohydrate antigen.[5] Such monovalent
epitopes have, however, to compete with a
multivalent presentation on cell surfaces,
e.g. of endothelial cells on blood vessels.
High concentrations of at least 1 mM have
to be reached to compete for the xenoanti-
bodies, which are divalent receptors in the
case of the IgG subclass, and decavalent for
the pentameric IgM subclass, which plays
a major role in hyper-acute rejection. The
well-established principle of multivalent
potentiation of weak carbohydrate–protein
interactions[6] should also be operative for
anti-carbohydrate antibodies. For this pur-
pose attempts with high molecular weight
oligosaccharides from pig-stomach mu-
cins,[7] with serum albumin conjugates,[8]
with multivalent arrays on oligo-ethylene
OH
OH
O
OR
O
Our poly-lysine based system, which
had already been successfully applied
for potent E-selectin ligands,[11] would be
well-suited for a multivalent presentation
of Linear-B. This scaffold has several ad-
vantages over many of the other approach-
es for multivalency. Poly-lysine is com-
mercially available with different degrees
of polymerization. After derivatization
a charge-neutral and hydrophilic amide
linked polymer results, which has a low
propensity for immunogenicity and is de-
gradable, as opposed to poly-acrylates or
poly-ethylene glycols. On the other hand,
albumin conjugates[8] and other proteins
may be immunogenic, the ethylene glycol
oligomer carries only eight carbohydrate
epitopes,[9] and poly-acryl amide with a
lipophilic backbone is not degradable, and
the molecular weights have upper limits.
To prepare poly-lysine conjugates a
Linear-B oligosaccharide with a 3-ami-
nopropyl aglycon had to be synthesized.
A purely chemical synthesis using for the
most part conventional transformations is
O
H
O
O
HO
O
HNAc
OH
HO
HO
OH
OH
1
"Linear B type II"
OH
HO
Fucose
OH
O
H3C
OH
O
O
OR
O
O
H
O
O
X
HNAc
OH
HO
HO
OH
O
OH
2
3
X: NHAc bloodgroup A
X: OH bloodgroup B
Fig.ꢀ1.ꢀStructuresꢀofꢀLinear-BꢀandꢀHumanꢀ
BloodgroupꢀCarbohydrateꢀEpitopes.
The hyper-acute rejection and to a
large part also the acute vascular rejection
are caused by anti-Linear-B antibodies
and the connected complement activa-
OAc
O
1) HO(CH2)3NHZ/Hg(CN)2
O
O
Ph
X
O(CH2)3NHZ
NTCP
AcO
O
Ac
O
O
O
H
N
2) NaOMe/MeOH
O
Cl
Cl
8
(37%, 3 steps)
3) PhCH(OMe)2/pTsOH
5
6
X: OAc
X: Br
HBr/AcOH
Cl
Na[CNBH3]/HCl-Et2O
THF/MS 4A
Cl
OBn
OBz
OBz
O
Et3SiOTf
Et2O
O(CH2)3NHZ
NTCP
HO
O
HO
H
SEt
SEt
NH
O
O
OBz
O
OBz
BnO
BnO
+
+
BnO
BnO
BnO
BnO
OAc
OAc
O
CCl3
O
OBn
O
10 (90%)
9
(64%)
11
OBn
DMTST/MS 4A
MeCl2
(1.5 equivalents)
12 (1.5 equivalents)
1) H2N(CH2)2NH2/EtOH
2) Ac2O/pyridine
3) LiOH, MeOH/H2O, 60 oC
OBn
OBz
O
O(CH2)3NHZ
NTCP
O
O
H
O
O
OBz
BnO
4) 20% Pd(OH)2-C/H2
BnO
BnO
OAc
O
OBn
13 (79%)
OH
OH
O
O(CH2)3NH2
O
O
O
O
OH
HO
O
HNAc
H
HO
HO
OH
4
(72%, 4 steps)
OH
Schemeꢀ1.ꢀChemicalꢀsynthesisꢀofꢀLinear-BꢀtypeꢀIIꢀtrisaccharideꢀ4ꢀwithꢀanꢀamino-propylꢀaglycon.

From ChemiCal researCh to industrial appliCationsꢀ
CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ1/2ꢀ 25
anol, acetylation, and deprotection finally
gives the trisaccharide 4. The conjugation
of Linear-B-propanolamine 4 to poly-
lysine is described in Scheme 2. Reaction
of 4 with thio-butyrolactone gives thiol 16.
Commercially available poly-(l)-lysine
hydrobromide batches 14 with average de-
gree of polymerization n around 40, 250,
or 1000 are then converted to the DMF-
soluble per-chloroacetamides 15, which in
turn are coupled with different equivalent
amounts x of thiolated trisaccharide 16 as
described previously.[11] The final Linear-B
conjugates 17 are obtained by capping the
remaining chloroacetamide groups with
racemic thioglycerol.[15] The composition
of these polymeric materials is given by
the specification of the starting batch of 14
OH
O
OH
H
N
O
O(CH2)3NH2
O
O
H
O
O
(CH2)4
HO
O
HNAc
OH
H3N+
HO
HO
OH
Br
14
poly-L-Lysine (Sigma)
4
n
OH
1) Dowex (OH) 2) p-TsOH/H2O
3) (ClCH2CO)2O/Lutidine/DMF
Et3N
Dithio-threitol
DMF, 90 oC, 18h
O
S
H
N
O
O
H
N
OH
OH
O
(CH2)4
HN
O
O
O
O
HS
O
H
HO
O
HNAc
OH
HO
HO
OH
O
16 (74%)
Cl
15 (84%)
OH
n
1) X equivalents
DBU/DMF, RT, 40 min.
1
(vendor), and by H-NMR measurements
O
O
H
N
H
at elevated temperatures. Assessment by
integration of selected signals, assigned to
either backbone, thioglycerol, or trisaccha-
ride, closely matched the stoichiometry of
reactants (cf. ref. [11]).
2) Thioglycerol Excess
Et3N/DMF, RT, 18h
N
(CH2)4
(CH2)4
HN
HN
O
O
3) Ultrafiltration
S
S
(CH2)3
CH2
These polymers were then tested in
vitro for binding avidity with anti-aGal
antibodies, and in a functional assay for
inhibiting complement dependent hemo-
lysis of pig erythrocytes by human serum.
[16] In binding assays competition for anti-
Linear-B antibodies (pooled human AB-
serum) between antigenic carbohydrate
coated on well plates and the soluble an-
tigen test compounds is measured with an
ELISA (enzyme-linked immunosorbent as-
say) format, selective either for the divalent
IgG or the decavalent IgM subtypes. The
results shown in Table 1 list inhibitory IC
concentrations based on equivalent weigh5t0,
i.e. trisaccharide concentration, rather
than on molecular weight of the polymer.
Thus, these equivalent weights are only
dependent on the fraction x of lysine resi-
dues coupled to saccharide, determined
by NMR, and are independent of the de-
gree of polymerization n, determined by
the vendor (SIGMA) by physical methods
(size exclusion chromatography coupled to
low angle laser light scattering). Polydis-
persities also specified by the vendor vary
between 1.1 and 1.2 (cf. also refs [11b,c]).
It is evident that multivalent arrays 17 in-
hibit anti-Linear-B antibodies better than
monovalent trisaccharide 4 (entry 1), and
for most cases also better than the human
serum albumin (HSA) conjugate tested
(entry 7). For IgM-binding and inhibition
of cytotoxicity the potency increases with
the degree of polymerization n: 36 (entry
2), 250 (entry 4), and 1050 (entry 6). Also
25% loading x is better than either 10% or
60% (entries 3, 4, 5). For the IgG subtype
a less distinct but similar trend can be seen.
For the best composition of 17 (entry 6) a
potency gain of several orders of magnitude
(2 to 6), compared to monovalent 4, is ob-
served. Further experiments led to the con-
CHOH
CH2OH
HN
O
Linear-B-(CH2)3
17
(1-x)n
xn
Schemeꢀ2.ꢀPreparationꢀofꢀLinear-Bꢀpoly-(l)-lysineꢀconjugatesꢀ17.
Tableꢀ1.ꢀIn vitroꢀDataꢀofꢀpoly-(l)-lysineꢀLinear-BꢀtypeꢀIIꢀtrisaccharideꢀ4 conjugatesꢀ17
Entry Compound
MW
(average) per
Molec.
# Lin-B
Equiv.
Weight
IC50
IgG
(μM)
IC50
IgM
(μM)
IC50
Hemolysis
(μM)
1
2
Trisaccharideꢀ4
602
1
602
0.7
390.0 397.0
Polymerꢀ17
15.5ꢀkD
~ꢀ9
1657
0.02
1.2
1.0
n:ꢀ36ꢀx:ꢀ0.26
3
4
5
6
7
Polymerꢀ17ꢀꢀ
n:ꢀ250ꢀx:ꢀ0.10
83.9ꢀkD
~ꢀ25
~ꢀ62
3356
1700
0.005 6.0
0.11
0.02
Polymerꢀ17ꢀꢀ
n:ꢀ250ꢀx:ꢀ0.25
106.3ꢀkD
158.4ꢀkD
446.3ꢀkD
0.003 0.45
Polymerꢀ17ꢀꢀ
n:ꢀ250ꢀx:ꢀ0.60
~ꢀ150
~ꢀ263
~ꢀ11
1056
0.2
0.005 0.08
Polymerꢀ17ꢀꢀ
n:ꢀ1050ꢀx:ꢀ0.25
1700
0.007 0.002 0.004
0.7
HSA-Conjugateꢀ ~ꢀ70ꢀkD
~ꢀ6000
(Dextra)
depicted in Scheme 1. The per-acetylated
tetrachloro-phthalimide-protected glucos-
amine derivative 5[12] is converted to the
anomeric bromide 6. Helferich-glycosy-
lation of N-benzyloxycarbonyl-protected
propanolamine 7, deacetylation and trans-
acetalization affords the 4,6-benzal deriva-
tive 8, which is reduced to the 6-O-benzyl
protected glucosamine 9. By virtue of the
bulky tetrachloro-phthalimido residue the
3-hydroxy group need not be protected
for the regioselective glycosylation of
the 4-hydroxy group with the Gal-a(1,3)
Gal disaccharide 10. This intermediate is
obtained by glycosylation of the selec-
tively protected S-ethyl-1-thiogalactose
derivative 11[13] with trichloro-acetimidate
12.[14] The crucial glycosylation of 9 with
glycosyl donor 10 is mediated by activa-
tion with dimethyl-thiomethyl-sulfonium
trifluoromethyl-sulfonate (DMTST) af-
fording the protected trisaccharide 13 in
79% yield. Cleavage of the tetrachloro-
phthalimide with ethylene diamine in eth-

26ꢀ CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ1/2ꢀ
From ChemiCal researCh to industrial appliCations
clusion that the configuration of lysine and
thioglycerol, as well as the spacer length
between carbohydrate and backbone play a
less significant role (data not shown).
β(1,4)Gal-Transferase (milk)
BSA, CIAP, MgCl2-6H2O, MnCl2-4H2O
OH
OH
OH
H2O, pH 5.5 (acetate buffer)
O
O
O(CH2)3NHZ
O(CH2)3NHZ
HO
HO
H
O
H
O
O
O
First in vivo experiments confirmed
the superiority of high molecular weight
poly-lysine conjugates 17. Therefore the
polymer with 25% carbohydrate loading
(x: 0.25) and a degree of polymerization n
between 900 to 1200 entered development
with the code GAS914. A major technical
hurdle turned out to be the chemical syn-
thesis of trisaccharide 4 (Scheme 1). In a
Prep Lab-setting the 24-step sequence with
six chromatographic separations could be
scaled up to 20 g of 4. Still, some toxic,
smelly, or touchy reagents, such as DMTST
(see above), and the amount of waste es-
timated to 10 tonnes per kg of trisaccha-
ride would make this route a daunting task.
After some attempts at optimization[17] it
was decided to resort to enzymatic gly-
cosylations with glycosyl transferases
and sugar nucleotides as glycosyl donors.
The advantages of total stereo- and regi-
oselectivity without the need of protect-
ing groups has to be balanced against cost
and availability of enzymes and activated
sugars, difficulties of reaction control, re-
producibility, and process scale up. A re-
search procedure starting with Z-protected
1(3-aminopropyl)-N-acetyl-glucosamine
18[18] had been elaborated,[19] but needed
to be optimized for robustness, and toxic
buffer (Na-cacodylate) should be replaced.
Also a risk assessment concerning proteins
of ruminant origin – bovine serum albumin,
β(1,4)Gal-transferase from cow milk, calf
intestine alkaline phosphatase – was need-
ed. As shown in Scheme 3, a first galacto-
sylation of 18 with commercially available
β(1,4)Gal-transferase from cow milk gave
N-acetyl-lactosamine 19 in 61–89% yield.
The pH 7.5 cacodylate buffer was success-
fully replaced by acetate buffering at pH
5.5, a measure which at the same time al-
leviated substrate inhibition. Addition of
Mg2+ solved the issue of excess of cofac-
tor Mn2+, which otherwise was depleted
by precipitation of its phosphate. Equally
successful was the optimization and scale
up of the second galactosylation with re-
combinant a(1,3)Gal transferase giving the
Z-protected trisaccharide 20 in 89% yield.
Crucial here was the quality of the com-
mercial enzyme in respect of minimized
β-galactosidase activity, which leads back
to monosaccharide 18. Again the arsenic
buffer could be replaced by a pH 6.5 ac-
etate buffer. In this case addition of Mg2+
was avoided, as it appeared to be connect-
ed with the unwanted glycosidase activity.
Besides inorganic phosphates uridine 21
is the only byproduct of these enzymatic
glycosylations. It is formed by phosphate
cleavage with CIAP (calf intestine alkaline
phosphatase) from uridine diphosphate, an
HNAc
HNAc
OH
OH
18 (50%, 4 steps)
19 (61 - 89%)
HO
OH
O
O
HO
α(1,3)Gal-T. (recombinant)
O
ONa
O
ONa
NH
H
O
O
BSA, CIAP, MnCl2-4H2O
P
P
N
O
O
H2O, pH 6.5 (acetate buffer)
O
O
UDP-Gal
(Yamasa Corp.)
OH
OH
O
OH
OH
NH
O
HO
O(CH2)3NHR
O
O
H
O
N
O
O
OH
HO
O
HNAc
Mn3[PO4]2
Mg3[PO4]2
HO
HO
OH
+
O
OH
OH OH
21
20 (89%)
R: Z
R: H
H2, 10% Pd-C
H2O/AcOH
4
(95%)
Schemeꢀ3.ꢀChemo-enzymaticꢀsynthesisꢀofꢀLinear-BꢀtypeꢀIIꢀtrisaccharideꢀ4ꢀwithꢀanꢀamino-propylꢀ
aglycon.
gus monkeys and baboons – both spe-
cies with preexisting natural antibodies
against the Linear-B epitope. As shown
in Fig. 2 (results of one animal) 1 mg/kg
i.v. injections into Cynomolgus monkeys
on days 1, 4, and 7 resulted in immediate
disappearance of circulating antigen spe-
cific antibodies, and also quenched the
complement dependent hemolytic activ-
ity of treated Cynomolgus monkey serum
on pig erythrocytes. Antibody levels and
serum cytotoxicity slowly recovered after
12 h, but never reached pretreatment lev-
els, even after 28 days. When the same
animals were treated again at days 105,
108, and 111, the same depletion/inhibi-
tion was observed, notably without any
immune response, i.e. without sensitiza-
tion to GAS914.[16] For pharmacokinetic
studies GAS914 was radiolabelled with
14C, using 14C-chloroacetic acid for the
preparation (cf. Scheme 2). In vivo experi-
ments with –/-Gal-T knockout and wild
type mice, rats, as well as with Rhesus
monkeys showed a biphasic elimination
of drug with 95% of radioactivity ex-
creted within minutes in the first phase.
The second slower phase was species
dependent with a half life of 30 min for
mice and 70 h for Rhesus monkeys. In
the case of animals without anti-Linear-B
specific antibodies – wild type mice and
rats – the radioactivity was completely
cleared by degradation in the liver and
further degradation and excretion by the
kidneys. For animals with preformed anti-
Gal specific antibodies – Rhesus monkeys
and Gal-T knockout mice – clearance was
essentially identical, but a minute residual
level of radiolabel was retained in lym-
phoid tissue, spleen and lymph nodes.
inhibitor of galactosyl transferases. The
intermediates 19 and 20 were purified by
reversed phase chromatography, and a
crystallization from ethanol/water yield-
ing products of >98% purity. A membrane
filtration before hydrogenolytic cleavage
to the target Linear-B propanolamine 4 en-
sured removal of any larger entities such as
endotoxins, prions, and viral particles. The
deprotected 3-aminopropyl-trisaccharide 4
wasalsoamenabletopurificationbyrecrys-
tallization from ethanol/water. The remain-
ing steps to the glycopolymer 17 (GAS914)
followed the research procedure (Scheme
2) with some adjustments. The thiolated
oligosaccharide 16 could be isolated by
crystallization from the concentrated reac-
tion mixture, thereby alleviating the separa-
tion from high boiling thio-butyrolactone.
This thiol was protected from oxidation
to disulfide by the addition of antioxidant.
Isolation of chloro-acetylated poly-(l)-ly-
sine 15 and the final glycopolymer 17 by
precipitation was a major hurdle. Precipi-
tates were finally obtained in reproducible
quality by direct precipitation in a stirrable
pressure filter under high dilution, and re-
dissolving of the product-cake directly on
the filter. For the final purification by tan-
gential flow ultra filtration compatible filter
material had to be evaluated. Characteriza-
tion of the product after lyophilization was
done with NMR, gel permeation chroma-
tography, and in vitro biochemical assays
for biological properties. The optimized
procedure thus allowed the preparation of
kg-quantities of drug substance of suffi-
cient quality to prepare injectable solutions
for first clinical studies.
In vivo experiments were first done
with non-human primates – Cynomol-

From ChemiCal researCh to industrial appliCationsꢀ
CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ1/2ꢀ 27
should prevent thrombotic microangiopa-
thy, causing the majority of organ losses.
[25] Yet another complication might be in-
flammation and loss of pig-xenografts by
infection with Human cytomegalovirus.[26]
Other options are tolerance induction
through chimeric bone marrow cells, so
far restricted to transplantations in early in-
fancy, before establishment of full immune
competency, or organogenesis from de-
veloping animal organ primordials rather
than from Human embryonic stem cells.[27]
Under these circumstances the clinical
development of GAS914 for eliminat-
ing xeno-antibodies was discontinued. It
should, however, be noted, that the poly-
lysine backbone of GAS914[11,15] offers an
excellent base for other antigen-specific
therapies in antibody mediated diseases, be
it as injectables or as ligands for immune-
aphaeresis.
anti-Gal IgM
anti-Gal IgG
2
2400
2000
1600
1200
800
400
0
haemolytic anti-pig
antibody (HAPAb)
1.5
1
0.5
0
Time (day. hr)
Fig.ꢀ2.ꢀAnti-GalꢀantibodyꢀtitersꢀandꢀcytotoxicityꢀlevelsꢀofꢀaꢀsingleꢀCynomolgusꢀmonkeyꢀ(Y119)ꢀ
receivingꢀ1ꢀmg/kgꢀi.v.-dosesꢀofꢀGAS914ꢀatꢀdaysꢀ1,ꢀ4,ꢀandꢀ7.ꢀPlottedꢀareꢀantibodyꢀtitersꢀ(IgGꢀandꢀ
IgM)ꢀrelativeꢀtoꢀstandardꢀhumanꢀserumꢀ(1.0),ꢀandꢀhemolyticꢀactivityꢀ(pigꢀerythrocytes)ꢀrelativeꢀtoꢀ
standardꢀhumanꢀserumꢀ(1000).
Received: December 18, 2009
[1] a) ‘Xenotransplantation’, Eds. D. K. C. Cooper,
E. Kemp, K. Reemtsma, D. J. G. White,
Springer Verlag, Heidelberg, 1991; b) M.
Cascalho, J. L. Platt, Immunity 2001, 14, 437;
c) H.-J. Schuurman, R. N. Pierson III, Frontiers
in Bioscience 2008, 13, 204.
[2] a) U. Galili, B. A. Macher, J. Buhler, S. B.
Shohet, J. Exp. Med. 1985, 162, 573; b) U.
Galili, S. B. Shohet, E. Kobrin, C. L. M. Stults,
B. A. Macher, J. Biol. Chem. 1988, 263, 17755;
c) E. Yuriew, M. Agostino, W. Farrugia, D.
Christiansen, M. S. Sandrin, P. A. Ramsland,
Expert Opin. Biol. Ther. 2009, 9, 1017.
With the aid of monoclonal antibodies
raised against the poly-(l)-lysine back-
bone of GAS914 it could be shown with
immuno-histochemical methods that the
radioactivity remaining in lymphoid tis-
sue was associated with intact GAS914
and also co-staining with B-cell regions.
onstrated that GAS914 injections consis-
tently removed at least 80% of preexist-
ing natural anti-Linear-B antibodies, thus
offering a relevant therapeutic option for
xenotransplantation in general.[21] Even
better results can be expected, when, in
addition to GAS914 with Linear-B type II
carbohydrate epitopes, analogous glyco-
polymers with the related Linear-B type
VI epitopes would be applied as well. Yet
another principle to present carbohydrate
ligands in multivalent arrays is self-assem-
bly of small entities, e.g. lipids or other
bipolar compounds aggregating to form
vesicles or liposomes, mimics of cell sur-
faces. We recently discovered that small
glyco-dendrimers, based on an aromatic
scaffold, also form aggregates, which bind
anti-Linear-B antibodies with high avidity,
comparable with GAS914.[22]
In the past years interest in pig-to-hu-
man xenotransplantation has to some ex-
tent diminished mainly because of the risk
of transmitting infections by the Porcine
endogenous retrovirus (PERV).[23] Then
the successful production of a(1,3)-ga-
lactosyl transferase doubly knockout pigs
by nuclear transfer cloning[24] was another
leap forward towards clinical xenotrans-
plantation. Currently the interest is focused
on transplantation of pig islet cells to treat
diabetes, or to bridge the waiting time for
an allograft by 2–6 months with a pig heart
transplant, thereby replacing mechanical
devices currently in use. From recent pig-
to-primate transplantations with a(1,3)
Gal-T knockout animals it became evident
that further genetic modifications would be
necessary to prolong survival of xenografts
beyond the six months currently achieved.
Along these lines anti-coagulant genes
[16]
It might therefore be speculated that
such collocation indicates binding of
GAS914 to B-cells, and suppression of
anti-Gal antibody production, resulting
in the observed long term lowering of an-
tibody levels.[16] During these treatments
no adverse side effects such as immune
complex mediated glomerulonephritis
were observed, and complement activa-
tion remained very low. To our knowl-
edge, GAS914 is the most potent inhibitor
of anti-aGalactosidic carbohydrate anti-
bodies reported. In the case of oligo-eth-
ylene glycol conjugates 50 mg/kg doses
have to be injected for similar effects,[9]
and poly-acryl amide conjugates appear
to be even less efficient, especially for the
IgG-type antibodies.[10] Furthermore the
lipophilic backbone of poly-acryl amide
poses a high risk for immunogenicity, and
biodegradation is difficult at best.
Polymers analogous to GAS914, also
with other Gal-(1,3)Gal carbohydrate
epitopes – disaccharide and pentasaccha-
ride – have been prepared, with an addi-
tional low percentage of lysine side chains
coupled to a reactive linker (amino group)
for immobilization on solid supports such
as Sepharose®. Similar to other immuno
adsorbents,[4] columns filled with such
material were less efficient for deplet-
ing serum from anti-Linear-B antibodies
than GAS914 injections.[20] Subsequent
pig-to-primate xenotransplantations dem-
[3] E. Cozzi, F. Bhatti, M. Schmoekel, G. Chavez,
K. G. Smith, A. Zaidi, J. R. Bradley, S. Thiru,
M. Goddard, C. Vial, D. Ostlie, J. Wallwork, D.
J. White, P. J. Friend, Transplantation 2000, 70,
15.
[4] a) D. Lambrights, P. Van Calster, Y. Xu, M.
Awwad, F. A. Neethling, T. Kozlowsky, A.
Foley, A. Watts, S. J. Chae, J. Fishman, A. D.
Thall, M. E. White-Scharf, D. H. Sachs, D. K.
C. Cooper, Xenotransplantation 1998, 5, 274;
b) A. Watts, A. Foley, M. Awwad, S. Treter, G.
Oravec, L. Buhler, I. P. Alwayn, T. Kozlowsky,
D. Lambrights, S. Gojo, M. Basker, M. E.
White-Scharf, D. Andrews, D. H. Sachs, D. K.
C. Cooper, Xenotransplantation 2000, 7, 181.
[5] a) U. Galili, K. L. Matta, Transplantation
1996, 62, 256; b) P. M. Simon, F. A. Neethling,
S. Taniguchi, P. L. Goode, D. Zopf, W. W.
Hancock, D. K. C. Cooper, Transplantation
1998, 65, 346.
[6] N. Jayaraman, Chemical Society Reviews 2009,
38, 3463; and literature cited therein.
[7] S. F. Li, F. A. Neethling, S. Taniguchi, J.-C.Yeh,
T. Kobayashi,Y.Ye, E. Koren, R. D. Cummings,
D. K. C. Cooper, Transplantation 1996, 62,
1324.
[8] a) R. G. Warner, WO Patent Appl. No. WO
97/07823, 1997; b) K. Teranishi, B. Gallackner,
L. Buhler, C. Knosalla, L. Correa, J. D. Down,
M. E. White-Scharf, D. H. Sachs, M. Awwad,
D. K. C. Cooper, Transplantation 2002, 73,
129.
[9] a) A. Schwarz, T. A. Davis, L. E. Diamond, J.
S. Logan, G. W. Byrne, WO Patent Appl. No.
WO 99/52561, 1999; b) L. E. Diamond, G. W.
Byrne, A. Schwarz, T. A. Davis, D. H. Adams,
J. S. Logan, Transplantation 2002, 73, 1780.
[10] a) J.-Q. Wang, X. Chen, W. Zhang, S. Zacharek,
Y. Chen, P. G. Wang, J. Am. Chem. Soc. 1999,
121, 8174; b) R. Rieben, N. V. Bovin, E. Y.

28ꢀ CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ1/2ꢀ
From ChemiCal researCh to industrial appliCations
Korchagina, R. Oriol, N. E. Nifant’ev, D. E.
Tsvetkov, M. R. Daha, P. J. Mohacsi, D. H.
Joziasse, Glycobiology 2000, 10, 141.
Transplantation Proceedings 2002, 34, 2757;
b) R. Zhong, Y. Luo, H. Yang, B. Garcia, A.
Ghanekar, P. Luke, S. Chakrabarti, G. Lajoie,
M. J. Phillips, A. G. Katopodis, R. O. Duthaler,
M. Cattral, W. Wall, A. Jevnikar, M. Bailey, G.
A. Levy, D. R. Grant, Transplantation 2003, 75,
10; c) T. T. Lam, B. Hausen, K. Boeke-Purkis,
R. Paniagua, M. Lau, L. Hook, G. Berry, J.
Higgins, R. O. Duthaler, A. G. Katopodis, R.
Robbins, B. Reitz, D. Borie, H.-J. Schuurman,
R. E. Morris, Xenotransplantation 2004, 11,
517; d) K. Kuwaki, C. Knosalla, K. Moran,
A. Alt, A. G. Katopodis, R. O. Duthaler, H.-
J. Schuurman, M. Awwad, D. K. C.Cooper,
Xenotransplantation 2004, 11, 210.
[11] a) G. Thoma, R. Duthaler, B. Ernst, J. L.
Magnani, J. T. Patton, WO PatentAppl. No. WO
97/19105, 1997; b) G. Thoma, J. L. Magnani,
R. Oehrlein, B. Ernst, F. Schwarzenbach, R. O.
Duthaler, J. Am. Chem. Soc. 1997, 119, 7414; c)
G. Thoma, J. T. Patton, J. L. Magnani, B. Ernst,
R. Oehrlein, R. O. Duthaler, J. Am. Chem. Soc.
1999, 121, 5919; d) G. Thoma, R. O. Duthaler,
J. L. Magnani, J. T. Patton, J. Am. Chem. Soc.
2001, 123, 10113.
[12] J. C. Castro-Palomino, R. R. Schmidt,
Tetrahedron Lett. 1995, 36, 5343.
[13] P. J. Garegg, S. Oscarson, Carbohydrate Res. [22] a) G. Thoma, A. G. Katopodis, N. H. Voelker,
1985, 136, 207.
[14] B. Wegmann, R. R. Schmidt, J. Carbohydrate
Chem. 1987, 6, 357.
[15] R. Duthaler, A. Katopodis, W. Kinzy, R.
Oehrlein, G. Thoma, WO Patent Appl. No. WO
98/47915, 1998.
R. O. Duthaler, M. B. Streiff, Angew. Chem.
2002, 114, 3327; Angew. Chem. Int. Ed. 2002,
41, 3195; b) G. Thoma, M. B. Streiff, A. G.
Katopodis, R. O. Duthaler, N. H. Voelcker, C.
Ehrhardt, C. Masson, Chemistry, A European
Journal, 2006, 12, 99.
[16] A. Katopodis, R. G. Warner, R. O. Duthaler, [23] a) D. Butler, M. Wadman, S. Lehrman, Q.
M. B. Streiff, A. Bruelisauer, O. Kretz,
B. Dorobeck, E. Persohn, H. Andres, A.
Schweitzer, G. Thoma, W. Kinzy, V. F. J.
Quesniaux, E. Cozzi, H. F. S. Davis, R. Mañez,
D. White, J. Clin. Invest. 2002, 110, 1869.
[17] S. Hanessian, O. M. Saavedra, V. Mascitti, W.
Marterer, R. Oehrlein, C.-P. Mak, Tetrahedron
2001, 57, 3267.
Schiermeier, Nature 1998, 391, 322; b) F. H.
Bach, H. V. Fineberg, Nature 1998, 391, 326;
c) R. A. Weiss, Nature 1998, 391, 327; d) K.
Paradis, G. Langford, Z. Long, W. Heneine, P.
Sandstrom, W. M. Switzer, L. E. Chapman, C.
Lockey, D. Onions, The XEN 111 Study Group,
E. Otto, Science 1999, 285, 1236; e) L. J. W. van
der Laan, C. Lockey, B. C. Griffith, F. S. Frasier,
C. A. Wilson, D. E. Onions, B. J. Hering, Z.
Long, E. Otto, B. E. Torbett, D. R. Salomon,
Nature 2000, 407, 90.
[18] E. Kamst, K. Zegelaar-Jaarsveld, G. A. van der
Marel, J. H. van Boom, B. J. J. Lugtenberg, H.
P. Spaink, Carboh. Res. 1999, 321, 176.
[19] a) J. Fang, J. Li, X. Chen, Y. Zhang, J. Wang, [24] a) L. Lai, D. Kolber-Simonds, K.-W. Park, H.-T.
Z. Guo, W. Zhang, L. Yu, K. Brew, P. G.
Wang, J. Am. Chem. Soc. 1998, 120, 6635;
b) N. Brinkmann, M. Malissard, M. Ramuz,
U. Römer, T. Schumacher, E. G. Berger, L.
Elling, C. Wandrey, A. Liese, Bioorg. Med.
Chem. Lett. 2001, 11, 2503; c) R. Oehrlein, G.
Baisch, unpublished results, cf. also G. Baisch,
R. Oehrlein, Angew. Chem. 1996, 108, 1949,
Angew. Chem. Int. Ed. 1996, 35, 1812.
Cheong, J. L. Greenstein, G. S. Im, M. Samuel,
A. Bonk, A. Rieke, B. N. Day, C. N. Murphy, D.
B. Carter, R. J. Hawley, R. S. Prather, Science
2002, 295, 1089; b) Y. Dai, T. D. Vaught, J.
Boone, S.-H. Chen, C. J. Phelps, S. Ball, J. A.
Monahan, P. M. Jobst, K. J. McCreath, A. E.
Lamborn, J. L. Cowell-Lucero, K. D. Wells, A.
Colman, I. A. Polejaeva, D. L. Ayares, Nature
Biotechnology 2002, 20, 251.
[20] a) R. Mañez, F. Crespo, E. Gonzales,A. Centeno, [25] a) Z. Ibrahim, M. Ezzelarab, R. Kormus, D. K.
A. Juffe, F. Arnal, E. Cozzi, D. J. G. White, R.
Duthaler, W. Kinzy, R. Oehrlein, A. Katopodis,
Transplantation Proceedings 2000, 32, 888; b)
R. Mañez, N. Domenech, A. Centeno, E. Lopez-
Pelaez, F. Crespo, A. Juffe, R. O. Duthaler, A.
G. Katopodis, Xenotransplantation 2004, 11,
408.
C. Cooper, Xenotransplantation 2005, 12, 168;
b) D. K. C. Cooper, A. Dorling, R. N. Pierson
III, M. Rees, J. Seebach, M. Yazair, H. Ohdan,
M. Awwad, D. Ayares, Transplantation 2007,
84, 1; c) R. N. Pierson III, J. Amer. Med. Assoc
(JAMA) 2009, 301, 967.
[26] M. Ghielmetti, A.-L. Millard, L. Haeberli, W.
Bosshard, J. D. Seebach, M. K. J. Schneider, N.
J. Mueller, Transplantation 2009, 87, 1792.
[21] a) K. Teranishi, I. P. J. Alwayn, L. Buhler, B.
Gollackner, C. Knosalla, L. Correa, J. D. Down,
M. E. White-Scharf, D. H. Sachs, R. Duthaler, [27] M. R. Hammermann, Transplant Immunology
A. Katopodis, M. Awwad, D. K. C. Cooper, 2005, 15, 1.