Projects_Completed_2012

ICGEB Collaborative Research Programme Projects successfully completed in 2012
ARGENTINA
Title: Sustainable agriculture: Molecular studies on the genetic basis of nodulation preference in
common bean-Rhizobium coevolution
Principal Investigator: O. Mario Aguilar, Instituto de Biotecnología y Biología Molecular (IBBM),
Facultad de Ciencias Exactas, Universidad Nacional de La Plata – CONICET, 1900 La Plata,
Argentina. Tel: +54-221-4229777, Fax: +54-221-4229777, E-mail: aguilar@biol.unlp.edu.ar
ICGEB Contract No.: CRP/08/003
ICGEB Reference No.: CRP/ARG08-03
Abstract: The symbiotic association between Phaseolus vulgaris (common beans) and rhizobio
displays mutual affinity between host and bacteria from the same geographical region of
diversification. The project was addressed to identify genes of the host that are differentially
expressed in response to infection by sympatric strains of R. etli, based in the use of cDNA
analysis, gene expression, reverse genetics. In addition, the molecular signal for nodulation
produced by different lineages of R. etli was examined by chemical analysis of the nodulation
factors. We have shown that a gene coding for a small GTP binding protein, PvRabA2a acts in polar
growth of root hair, and it is also required for reorientation of root hair growth axis during bacterial
infection in the early stages of interaction. In addition, we have identified a gene coding for the C
subunit of the transcription factor NGF-Y, and demonstrated to play a key role in determining
preference for bacteria that occupy nodules. On the other hand, we found no differences in
chemical structure of nodulation factors produced by lineages of R. etli belonging to the
Mesoamerican and Andean gene pool.
Objectives:
We aim to discover and characterise common beans genes involved in the preferential nodulation
towards the definition of molecular indicators of coevolution. For this, we propose to study the
pattern of expression and investigate the role of selected host genes we have identified in our
previous work based on a differential screening. Specifically, we will further characterise common
bean genes coding for a small GTP binding protein Rab11, a transcription factor Hap5, and the
chemical structure of nodulation factors produced by lineages of species Rhizobium etli.
Results Obtained:
Common beans (Phaseolus vulgaris) are the most important legume for direct human consumption.
In countries such as Brazil, Mexico and African countries Burundi, Rwada, Kenya and Tanzania,
represent the major source of protein in the diet. It is the source of protein for more than 500
million people in Africa and Latin America. Latin America produces about 50% of the world
production however in many cases low yields are reported due to different causes that include
pests, fertility of soil, drought. As most legume plants, common beans establish a nitrogen fixing
association with soil bacteria collectively known as rhizobia. This symbiotic interaction results in the
formation of a new organ, the nodule, where the bacteria are allocated, differentiate into
bacteroids and fix atmospheric dinitrogen into reduced forms, which are readily available to sustain
plant growth. To improve our understanding of the basic processes involved in biological nitrogen
fixation by the association common beans-rhizobia can potentially be useful to improve nitrogen
nutrition in this crop. Argentina produces about 300,000 ton per year in the Northwest region of
the country, which about 95% is exported.
This legume is believed to have been domesticated in the Americas, with two centres of
diversification, one of them in Meso-America and the other in the Southern Andes dispersed in
Ecuador, Peru, Bolivia and Argentina.
We have demonstrated nodulation preference between common beans and lineages of R. etli from
the same host region. Our results invited us to speculate that the major genetic pools of P.vulgaris
and rhizobia coevolved independently after geographic separation (Aguilar, et al., PNAS 2004).
By applying a suppressive subtractive hybridisation approach in which cDNA from a Mesoamerican
cultivar inoculated either with the more or the less efficient strain was used as driver and tester, a
set of genes were identified as been involved in the early preferential symbiotic interaction
(Peltzer-Meschini, et al., Mol. Plant. Microb. Interac. 2008). This analysis identified candidate genes
to be further investigation of their roles in the specific interaction with lineages of species
Rhizobium etli, which build up the objectives of this proposal.
The characterisation of PvRabA2a, a common bean cDNA previously isolated as differentially
expressed in root hairs infected with R. etli, which encodes a protein highly similar to small
GTPases of the RabA2 subfamily. This gene is constitutively expressed in roots, particularly in the
elongation zone and in root hairs, where RabA2 protein accumulates in the tip of the expanding
region. The role of this gene in nodulation has been studied in Phaseolus vulgaris composite plants
using a reverse genetic approach. Examination of root morphology in PvRabA2a RNAi plants
revealed that the number and length of the root hairs were severely reduced. Upon inoculation
with R. etli, nodulation was completely impaired and induction of early nodulation genes (ENODs)
was undetectable in silenced hairy roots. Moreover, PvRabA2a RNAi plants failed to induce root hair
deformation and to initiate ITs, indicating that morphological changes that precede bacterial

infection are compromised in these plants. Early molecular responses, as ERN1, ENOD40 and Hap5
induction are abolished in PvRabA2a RNAi roots. We propose that PvRabA2a, which acts in polar
growth of root hair, also is required for reorientation of root hair growth axis during bacterial
infection (Blanco, et al. Plant Cell 2009).
In addition, we identified in Phaseolus vulgaris a C subunit of the heterotrimeric nuclear factor NF-
Y, and demonstrated to play a key role in determining preference for bacteria that occupy nodules.
This gene is part of the signal transduction triggered in Mesoamerican cultivars of Phaseolus
vulgaris in response to Rhizobium etli strains predominant in soils of this domestication center.
Knock-down of NF-YC produces excessive root hairs deformation in response to bacterial infection
and decreases nodule number and occupancy. In addition, over-expression of NF-YC in roots leads
to a more efficient colonisation by less effective bacteria in single inoculation and competence
experiments. NF-YC silencing affected negatively the induction of cell cycle genes, providing a link
between strain perception and cell divisions required for nodule organogenesis. Our results show
that the coevolution between legumes and rhizobia is accompanied by a specific recognition
followed by a more efficient infection process. Understanding molecular mechanisms that
determine the genetic compatibility between legumes and rhizobia is the key to define breeding
programs designed to optimise nitrogen fixation (Zanetti, et al., Plant Cell 2010). Attempts to
identify the interactor of NFYC are in progress.
We have further characterised the interaction between beans from the host center of diversification
and lineage of R. etli and found that that inoculated roots showed earlier response to infection with
cognate strains as compared to alopatric strains. These responses include root hair curling,
formation of infection threads and number of nodule per plant. In order to gain insight on the
differential response, the nodulation factor molecules (NF) produced by strains from the
Mesoamerican center of diversification and from the Andean center of diversification were
analysed. The molecule signal of one out of three Andean R. etli strains (strain 55N1) we examined
showed a chitolipooligosaccharidic molecule that is devoid of the acetyl residue on the fucosyl
moiety at the reducing end of the molecule. The structure of the other Andean strains as well as
the Mesoamerican strains were found to be similar each other. We assessed the competitiveness
for nodule occupancy of Mesoamerican beans by an R. etli nolL mutant derived of the
Mesoamerican strain CE3, which fails to acetylate the fucosyl residue. Several combinations of two
R. etli strains were used in co-inoculation assays. Results indicated that although mutation in gene
nolL affects negatively the competitiveness in co-inoculations with any of the two Mesoamerican R.
etli wild type strains we assayed (CE3 and SC15), it does not affect the large nodule occupancy in
co-inoculation experiments with the alopatric strain R. etli 55N1 from the Southern Andean region.
These results suggested that factors other than NFs could play a role in determining affinity for
nodulation by certain lineages of R. etli. Within this frame, we have performed competition
experiments in bean varieties from both centers of host diversification by using a combination of
strains that included Rhizobium tropici CIAT899, a bean-nodulating genotype other than R. etli.
Nodules of the Mesoamerican cultivar occupied by strains CIAT899 and R. etli 55N1 were found to
be comparable to each other, whereas in the Andean bean cultivar Alubia, occupancy by the
sympatric strain 55N1 clearly overcomes strain CIAT899. This result indicated that competitiveness
is expressed in the host x strain combination of the same centre of genetic diversification.
Results Unforeseen in the Original Project:
The results meet the hypothesis we have proposed in our objectives, and indeed provided
important insight in the biology of symbiotic interaction. What is more important, are the new
avenues for future research, now open, and matter of current investigation in our laboratory, which
brings new challenges for young and future scientists.
Publications:
Blanco, F.A., Peltzer Meschini, E., Zanetti, M.E., Aguilar, O.M. A monomeric G-protein of the
Rab family is required for root hair formation and pre- infection stages in the common bean-
rhizobium symbiotic association. 2009. Plant Cell 21, 2797-2810
Zanetti, M.E., Blanco, F.A., Beker, M.P., Battaglia, M., Aguilar, O.M. A C-subunit of the plant
nuclear factor NF-Y required for rhizobial infection and nodule development affects partner
selection in the common bean-Rhizobium etli symbiosis. 2010. Plant Cell 22, 4142-4157
Mazziotta, L., Reynoso, M., Aguilar, O.M., Blanco F., Zanetti, M.E. Transcriptional and
functional variation of NF-YC1 in genetically diverse accessions of Phaseolus vulgaris during the
symbiotic association with Rhizobium etli. 2012. Plant Biol. (submitted)

BRAZIL
Title: Production of transgenic rabbits for hybridoma-free generation of monoclonal antibodies
Principal Investigator: José Ernesto Belizário, Department of Pharmacology, Institute of
Biomedical Sciences, University of São Paulo, Av. Lineu Prestes, 1524 Cidade Universitária, CEP
05508-900, São Paulo, Brazil. Tel: +55-11-30917318, Fax: +55-11-30917438, E-mail:
jebeliza@usp.br
ICGEB Reference No.: CRP/BRA08-01
Abstract: The Simian Virus 40 (SV40) large T antigen is multifunctional protein with DNA helicase,
RNA helicase and ATPse activities, which contribute to tumourogenesis in rodents and humans. The
Immortomouse mouse strain expresses a mutated large T antigen tsA58 oncogene under the
control of the interferon inducible murine H-2Kb
MHC (Class I antigen) promoter on chromosome
16. We backcrossed H-2Kb
-SV40-tsA58 mice, which is a F1 hybrid of CBA/Ca x C57BL/10 strains
with BALB/c mice to obtain a transgenic colony with unique BALB/c background. We have used two
pre-validated PCR genotype assays that can distinguish between wild-type, hemizygous, and
homozygous animals. Enlargement of thymus is a phenotypic abnormality of immortomouse. We
characterised macroscopically and by immunohistochemistry thymic hyperplasia only in the
offspring hemizygous females (2:5 ratio). Studies are underway to typing T cell (CD4/8)
populations in the thymuses. Moreover, we have routinely observed offspring with undersised
males (1:10 ratio) displaying abnormalities after birth. This BALB/c strain of H-2Kb
-tsA58
immortomouse can be utilised to investigate specific pathological and physiological patterns
associated SV-40 oncogenicity. This transgenic mouse strain will help to isolate immortalising cell
lines growing under the permissive 33˚C temperature including T cell lymphomas and B cell
plasmacytoma capable of producing monoclonal antibodies.
Objectives:
Construction of transgene cassette into the FUGW plasmid, production of transgenic rabbits and
genotype and phenotype of founders.
Results Obtained:
Since the creation of rabbit hybridoma technology, efforts to improve efficiency and stability of
rabbit monoclonal antibody-producing plasmacytoma lines as rabbit antibodies have the best
properties compared to mouse monoclonal antibodies. Aiming to establish the techniques and skills
for creation genetically modified rabbits, along the project we have superovulated does and plated
rabbit embryos in KSOM medium in Petri dishes under mineral oil and analysed the influence of
various conditions and times regarding quality, number and embryo development with and without
manipulation. We observed rabbit embryos to differentiate from zygotes, two-cell, four-cell, eight-
cell, morula, and early blastocyst stage in CO2 incubator at 38o
C. We purified the H-2Kb
-SV40tsA58
transgene using Geneclean III kit (Bio101). Over 200 fertilised zygotes were microinjected and
transferred into recipient New Zealand white rabbits. The live pups from 2 recipient does were
checked for transgene expression using PCR specific primers and no product was observed
suggesting they were negative for the transgene. To address this problem we have been using
FUGW vector construction that expresses GFP report to validate its application to rabbit
transgenesis by lentiviral co-infection method. So far, we did not succeed to obtain live-born rabbit
pups. In parallel, we have done studies for the establishment and characterisation of a primary
rabbit fibroblast cell lines Rb1 and Rb2 expressing both H-2Kb
-SV-40-tsA58 transgene and eGFP
marker gene. We confirmed the expression of SV40tsA58 protein in rabbit fibroblast cell line under
the regulation of H-2Kb
promoter after induction by human IFN-γ at 33o
C permissive temperature.
The fibroblast cell lines will be used in future experiments aiming to evaluate the somatic cell
nuclear transfer technique for producing genetically engineered rabbits. In 2004, Pasqualini and
Arap’s paper in PNAS showed that splenocytes derived from the spleen of hyper-immunised
immortomouse could yield high titers of monoclonal antibodies against a filamentous phase capsid
antigen by growing them at 33o
C in the presence of IFN-γ. This study was carried out with mice
from with CBA/Ca x C57BL/10 F1 strain which have a mixed background and week immunological
response to antigenic stimulation as compared to BALB/c strain. We have successfully
crossbreeding, genotyping and phenotypying mice from offspring of CBA/CaxC57BL/10 hybrid with
BALB/c mice. In functional studies, we will carry out the immunisation of experimental groups of
BALB/c immortomice with specific antigens, check out for antibody titers and collect spleens for in
vitro screening and generation of clonal antibody-producing plasmacytomas under the temperature
of 33o
C in presence of IFN-γ. This new technology may be more reliable and convenient for large-
scale manufacture of antibodies ex vivo.
We had technical problems with the construction of the 6.9 bp containing H-2Kb
-SV40sA58
transgene, which exceeds the estimated size for an insertion fragment into FUGW lentivirus vector.
Most viral vectors can carry no more than 8-10 kb of DNA payload taking in account the cis-acting
sequences from those trans-acting factors that are absolutely required for viral particle production,

infection, and integration. Transgenic rabbits are much more difficulty to produce than mice. Using
the pronuclear microinjection method, we obtained only three live-born pups from one out of 6
does. The rabbit embryo has a thick mucin layer that forms around the zona pellucida, which
strongly influences implantation in the uterus and pregnancies. We assume that thickness of mucin
layer might be contributed to low rates and efficiency of pregnancies.
Publications:
Belizario, J.E. Immunodeficient mouse models: an overview. 2009. Open Immunol. 2, 79-85
Belizario, J.E. Pathogens and dead cells cooperate with cytokines in activating the innate and
adaptive response. 2011. Eur. J. Inflamm. 9(1), 1-11
Belizario, J.E., Sayami, P., Wolf, P., Strauss, B., Xavier-Neto, J. New routes for transgenesis
of the mouse. 2012. J. App. Genet. 53(3), 295-315

CHILE
Title: Role of the Unfolded Protein Response (UPR) in tumour suppression by caveolin-1
Principal Investigator: Claudio Hetz, Institute of Biomedical Sciences, University of Chile,
Independencia 1027, Santiago, Chile. Tel: +56-2978-6506 (direct), 2978-6876 (assistant), Fax:
+56-2-9786871, E-mail: chetz@hsph.harvard.edu
ICGEB Reference No.: CRP/CHI08-03
Abstract: The mechanisms underlying the ability of caveolin-1 to suppress tumour formation
remains highly controversial. In particular, the possibility that caveolin-1 presence in the
endoplasmic reticulum (ER) may modulate stress responses in this organelle relevant to tumour
formation remains unexplored. Tumour hypoxia causes a robust increase in the expression of
molecular chaperones and other components of an ER stress reaction termed the Unfolded Protein
Response (UPR). Several UPR components are directly implicated in the generation of solid
tumours. In this project, we developed a joint investigation with Dr. Andrew Quest to define the
possible role of the UPR as a downstream target of caveolin-1 in cancer. An in vivo cancer model
was developed using B16F10 mouse melanoma cells that lack caveolin-1 expression. Tumour
progression correlated with the activation of a strong ER stress response, a phenomenon that was
significantly attenuated by caveolin-1 expression. Our collaborative results suggest that caveolin-1
expression suppresses survival pathways triggered by ER stress and hypoxia in solid tumours. Our
results contributed to understand the functional relationship between caveolin-1 expression and
suppression of the UPR, as well as define the possible therapeutic benefits of targeting the UPR
pathway in cancer models in vivo.
Objectives:
(i) To define the role of the UPR in tumour suppression by caveolin-1;
(ii) To assess the effects of caveolin-1 on UPR signalling and cell survival in vitro in experimental
models of ER stress and hypoxia;
(iii) To investigate the correlation between caveolin-1 levels and markers of ER stress in human
tumour samples derived from Chilean cancer patients.
Results Obtained:
In this project we have studied the role of the UPR as a possible target to uncover the known
tumour suppressor activities of caveolin-1. We have performed a series of in vivo experiments to
provide in vivo validation of the major results obtained in cell culture. Our results indicate that
caveolin-1 expression significantly reduces the signalling of two major UPR signalling branches
initiated by the stress sensors IRE1a and PERK both in tumours and ex-tumour cultures. Moreover,
we were able to demonstrate a functional role of caveolin-1 in the suppression of ER stress
signalling in cell culture experiments in models of ER stress and hypoxia.
Publications:
Rojas-Rivera, D., Armisén, R., Colombo, A., Eguiguren, A.L., Martínez, G., Díaz, A.,
Kiviluoto, S., Rodríguez, D., Patron, M., Rizzuto, R., Bultynck, G., Concha, M.L., Sierralta,
J., Stutzin, A., Hetz C. TMBIM3/GRINA is a novel Unfolded Protein Response (UPR) target gene
that controls apoptosis through the modulation of ER calcium homeostasis. 2011. Cell Death Differ.
(in press)
Valenzuela, V., Collyer, E., Armentano, D., Parsons, G., Court, F.A., Hetz, C. Activation of
the Unfolded Protein Response enhances motor recovery after spinal cord injury. 2012. Cell Death
Dis. (in press)
Barrientos, S.A., Martinez, N.W., Yoo, S., Jara, J.S., Zamorano, S., Hetz, C., Twiss, J.L.,
Alvarez, J., Court, F.A. Axonal degeneration is mediated by the mitochondrial permeability
transition pore. 2011. J. Neurosci. 31, 966-978
Castillo, K., Rojas-Rivera, D., Lisbona ,F., Caballero, B., Nassif, M., Court, F., Schuck, S.,
Ibar, C., Walter, P., Sierralta, J., Glavic, A., Hetz, C. BAX inhibitor-1 regulates autophagy by
controlling the IRE1a/JNK branch of the unfolded protein response. 2011. EMBO J. 30(21), 4465-
4478
Hetz, C., Martino, F., Rodriguez, D., Glimcher, L.H. The Unfolded Protein Response:
Integrating stress signals through the stress sensor IRE1α. 2011. Physiol. Rev. 91, 1219-1243
Klein, A., Mosqueira, M., Martinez, G., Robledo, F., González, M., Caballero, B., Cancino,
G.I., Alvarez, A.R., Hetz, C., Zanlungo, S. Lack of activation of the Unfolded Protein Response
(UPR) in mouse and cellular models of Niemann-Pick type C disease. 2011. Neurodegener. Dis.
8(3), 124-128
Matus, S., Glimcher, L.H., Hetz, C. Protein folding stress in neurodegenerative diseases: a
glimpse into the ER. 2011. Curr. Op. Cell Biol. 23, 239-252
Nassif, M., Hetz, C. Targeting autophagy in ALS: a complex mission. 2011. Autophagy 7(4), 450-
453
Rodriguez, D., Rojas-Rivera, D., Hetz, C. Integrating stress signals at the endoplasmic
reticulum: The BCL-2 protein family rheostat. 2011. Biochim. Biophys. Acta 1813(4), 564-574

Torres, M., Encina, G., Soto, C., Hetz, C. Abnormal calcium homeostasis and protein folding
stress at the ER: a common factor in familial and infectious Prion disorders. 2011. Commun.
Integr. Biol. 4, 258-261
Vidal, R., Caballero, B., Couve, A., Hetz, C. Converging pathways in the occurrence of
endoplasmic reticulum (ER) stress in Huntington’s disease. 2011. Curr. Mol. Med. 11, 1-12
Woehlbier, U., Hetz, C. Modulating stress responses by the UPRosome: A matter of life and
death. 2011. Trends Biochem. Sci. 36, 329-337
Gupta, S., Deepti, A., Deegan, S., Lisbona, F., Hetz, C., Samali, A. HSP72 protects cells from
ER stress-induced apoptosis via enhancement of IRE1-XBP1 signalling through a physical
interaction. 2010. Plos Biol. 8(7), e1000410
Nassif, M., Matus, S., Castillo, K., Hetz, C. Amyotrophic lateral sclerosis pathogenesis: a
journey through the secretory pathway. 2010. Antioxid. Redox. Signal. 13, 1955-1989
Torres, M., Matus, S., Arminsen, R., Stutzin, A., Soto, C., Hetz, C. Prion protein misfolding
affects calcium homeostasis and sensitises cells to endoplasmic reticulum stress. 2010. Plos One
5(12), e15658
Hetz, C. The UPR as a survival factor of cancer cells: More than folding proteins? 2009. Leuk. Res.
33(7), 880-882
Hetz, C., Glimcher, L. Fine tuning of the Unfolded Protein Response: Assembling the IRE1
interactome. 2009. Mol. Cell 35, 551-561
Hetz, C., Thielen, P., Matus, S., Nassif, M., Court, F., Cuervo, A.M., Martinez, G., Kiffin, R.,
Brown, R.H., Glimcher, L. XBP-1 deficiency in the nervous system protects against amyotrophic
lateral sclerosis by increasing autophagy. 2009. Genes Dev. 23, 2294–2306 (Comment in: Genes
Dev. 2009. 23(19), 2253-2259)

CHINA
Title: Genomic sequence diversity and oncogenic potential of human papillomavirus type 58
variants across the world
Principal Investigator: Paul Kay-Sheung Chan, Department of Microbiology, Faculty of Medicine,
The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong
Special Administrative Region, China. Tel: +86-852-2632-3333, Fax: +86-852-2647-3227, E-mail:
paulkschan@cuhk.edu.hk
ICGEB Reference No.: CRP/CHN08-03
Abstract: This project analysed the sequence variation, phylogeny and geographical distribution of
human papillomavirus (HPV) 58 variants from a global perspective. It assessed the epidemiological
risk association and characterised the in vitro oncogenic properties of E7 proteins derived from
naturally occurring HPV58 variants. The data are crucial for designing diagnostic assays and
vaccines, and are instrumental for further studies on HPV58, a variant with strong
ethnogeographical predilection in attribution to cervical cancer.
Objectives:
Ultimate goal is to achieve a better control on human papillomavirus (HPV) 58-related cervical
neoplasia which constitutes a substantial health burden in East Asia and parts of Latin America.
(i) Comprehensive analysis on genomic sequences of HPV58 isolates collected worldwide;
(ii) Elucidate the oncogenicity of representative HPV58 strains.
Results Obtained:
We examined 401 HPV58 isolates collected from 15 countries/cities, with the full length of L1, LCR,
E2, E4, E5, E6 and E7 gene regions accounting for ~57% (4416-4462 bp) of the whole viral
genome sequenced.
(i) Lineage classification: A lineage classification system was established based on phylogenetic
analysis of the concatenated E6-E7-E2-E5-L1-LCR sequences. HPV58 variants could be classified
into 4 distinct lineages designated as A, B, C and D. In order to facilitate large-scale
epidemiological studies, we have identified 3 short fragments that can serve as surrogate targets
for constructing phylogenetic trees for lineage identification.
(ii) Geographical distribution of lineages: The distribution of HPV58 variant lineages were found to
vary worldwide. Lineage A was significantly less frequent in Africa (49.3%) than other regions
(85.7%-95.8%). Lineage C was significantly more frequent in Africa than Asia, America or Europe
(P ≤ 0.001 for each comparison).
(iii) Sequence variation of PCR primer binding sites on L1: We analysed the sequence variation of
HPV58 L1 regions correspond to the binding sites of five commonly used consensus primer sets. As
a result, 19 mismatches were found within the primer binding sites of 5 sets of commonly used
primers. GP5+/6+ had the most, whereas PGMY09/11 had the least number of mismatches.
(iv) Sequence variation of E6 and E7 ORF: Although the degree of nucleotide sequence variation
among HPV58 variants was small (maximum 2% for E6, 3% for E7), the variants could be grouped
into 4 clades that were identifiable by distinct signatures of sequence variations. These molecular
signatures are not only convenient markers for large-scale epidemiological studies, the
substitutions themselves may carry biological implications.
(v) Geographical distribution of E6 and E7 variants: The distribution of E6/E7 clades varied
geographically. Africa had significantly more E6 clade 3, but significantly less E6 clade 1 than other
regions. E7 clades 2 and 3 co-circulated in Asia, whereas clades 2 and 4 co-circulated in Africa. E7
clade 3 was more frequent in Asia, and clade 4 was more frequent in Africa. E6 WW01 was the
predominant variant detected in Asia (78%), Europe (91%) and Americas (79%), but not in Africa
(47%). E6 WW02 was significantly more frequent in Africa compared to other regions. E7 WW01
was the most common variant found across the four regions, and with a significantly higher
prevalence in Europe (77%) compared to Asia (42%) and Africa (47%). The other variants E7
WW02 / 03 / 04 showed a more distinct geographical predilection.
(vi) Epidemiological risk association: E7 clade 2, E7 WW01, E7 G41R and G63D were associated
with a lower risk; whereas E7 clade 3, E7 WW02, E7 T20I and G63S were associated with a higher
risk for cervical neoplasia. Similar associations were reproduced for Hong Kong and Korean which
provided sufficient sample for subgroup analysis.
(vii) Transformation potential of E7 proteins: Results of epidemiological analyses indicated that the
most important amino acid substitutions were T20I and G63S located in the E7 ORF. E7 proteins
expressed from plasmids containing the prototype and three naturally occurring variants, 58E7-V1
(with nucleotide changes of C632T [T20I], T744G [silent] and G760A [G63S]), 58E7-V2 (G694A
[G41R], T744G [silent] and G761A [G63D]) and 58E7-V3 (T744G [silent], A793G [T74A], C798T
[D76E], C801A [D76E] and C840T [silent]) were compared. HPV58 E7-V1 showed a marked
increase in ability to transform BRK cells compare to prototype; whereas the other two variants
were similar to prototype.

(viii) Binding with proteasome subunits: Proteosome subunit S2 shows a negative interaction with
all the E7 proteins. S4 interacted efficiently with the prototype E7 but less efficiently with the three
variants. S5a showed the strongest interaction with E7, especially with 58E7-V1. S7 and S8 also
interacted efficiently with HPV58 E7 proteins.
(ix) pRb degradation: pRB was more efficiently degraded by HPV58 E7-V1 protein in vivo compared
to prototype.
(x) Suppression of cellular senescence: The HPV58 E7-V1 protein was found to be most efficient in
suppressing the cellular senescence, which was in concordance with the results obtained from the
pRB degradation assay.
Publications:
Chan, P.K., Luk, A.C., Park, J.S., Smith-McCune, K.K., Palefsky, J.M., Konno, Y.,
Giovannelli, L., Coutlée, F., Hibbitts, S., Chu, T.Y., Settheetham-Ishida, W., Picconi, M.A.,
Ferrera, A., De Marco, F., Woo, Y.L., Raiol, T., Piña-Sánchez, P., Cheung, J.L., Bae, J.H.,
Chirenje, M.Z., Magure, T., Moscicki, A.B., Fiander, A.N., Di Stefano, R., Cheung, T.H., Yu,
M.M., Tsui, S.K., Pim, D., Banks, L. Identification of human papillomavirus type 58 lineages and
the distribution worldwide. 2011. J. Infect. Dis. 203, 1565-1573
Chan, P.K., Cheung, J., Luk, A., Ho, W. Sequence variations of human papillomavirus type 58
E2, E4 and E5 genes across the world. In: 26th
International Papillomavirus Conference & Clinical
and Public Health Workshop, Montreal, Canada, 3-8 July 2010. Abstract No. P-119
Cheung, J., Luk, A.C., Ho, W.C., Chan, P.K. Sequence variation and lineage classification system
for HPV58: A worldwide perspective. In: 26th
International Papillomavirus Conference & Clinical and
Public Health Workshop, Montreal, Canada, 3-8 July 2010. Session: Basic Science 6, Abstract No.
345
Ho, W.C., Cheung, J.L., Luk, A.C., Chan, P.K. Sequence variation of Human Papillomavirus Type
58 E7 across the world. In: 26th
International Papillomavirus Conference & Clinical and Public
Health Workshop, Montreal, Canada, 3-8 July 2010. Abstract No. P-125
Chan P.K.S. Worldwide distribution of human papillomavirus 58 variant lineages. In: The
Christophe Merieus Conference – Trends in Virology, Annecy, France, 17-19 January 2010
Cheung, J., Luk, A., Chan, P.K. Mismatch of commonly used primers for HPV58 sequences
collected worldwide. In: 27th
International Papillomavirus Conference and Clinical Workshop, Berlin,
17-22 September 2011, Abstract No. P-32.23

COLOMBIA
Title: Identification of pathogenicity factors in Batrachochytrium dendrobatidis by in silico
comparative genomics
Principal Investigator: Silvia Restrepo Restrepo, Mycology and Plant Pathology Laboratory,
Biological Sciences Department, Faculty of Sciences, Universidad de Los Andes, Cra 1 No 18A-10,
Bogota, Colombia. Tel: +57-1-3394949, ext. 2777, Fax: +57-1-3394949, ext. 2817, E-mail:
srestrep@uniandes.edu.co
ICGEB Reference No.: CRP/COL09-01
Abstract: Global amphibian diversity is at great risk, and Chytridiomycosis is one of the main
causes linked to this massive extinction. To develop future control strategies for this threat is
paramount to understand what molecular determinants mediate the infection process and allow
Batrachochytrium dendrobatidis (Bd) to cause disease. We developed a comparative genomics
strategy using both available sequenced Bd strains and ESTs libraries of 44 organisms (9 parasites,
9 non-pathogenic Fungi, 26 pathogenic fungi) to independently determine potential pathogenicity
determinants. Among the candidates recovered, five are particularly promising. To test these
pathogenicity hypotheses, a controlled infection system with tolerant host skin (Rheobates
plamatus) was developed in which we assessed expression patterns of three of the candidates at
early stages of infection (0, 12, 24, and 36hpi). For these three genes, we found experimental
evidence of strong variations in transcript in very early stages in the establishment of infection (0-
12hpi), suggesting a possible role in the initial interactions between Bd and its host. On the other
hand, due to the importance of the studying of Bd in Colombia as a way to understand the
epidemiology and potentially the biology of the disease, here we also did the genetic and
morphological characterisation of the first Colombian isolate of Bd (strain EV001).
Objectives:
(i) Prediction of putative pathogenicity factors by comparative genomics;
(ii) Isolation and characterisation of a Colombian strain of Batrachochytrium dendrobatidis;
(iii) Development of a co-culture system of Bd and host skin;
(iv) Determination of the expression patterns of the putative pathogenicity genes.
Results Obtained:
We developed a comparative genomics strategy using both available sequenced Bd strains and
ESTs libraries of 44 organisms (9 parasites, 9 non-pathogenic Fungi, 26 pathogenic fungi) to
independently determine potential pathogenicity determinants based on two criteria: 1) Proteins
shared between Bd, parasites and pathogenic fungi, 2) Proteins shared only with pathogenic fungi.
Around 20% of the predicted proteins were shared with pathogenic fungi. Of these, only about 3%
were absent in non-pathogenic, but more that 60% of these were also shared with Parasites.
Among these and of special interest we obtained a complex with putative Ascorbic acid-dependant
ferric reductase activity; a receptor possibly involved in lipophilic ligands detoxification (LMBRL); a
Serine peptidase (POP1) possibly related to skin defense peptide degradation; a Metazoa-like
possibly secreted PHACTR protein and a PhsF homolog maybe obtained by horisontal transfer.
These candidates provide lights towards understanding Bd pathogenesis and future control
strategies.
As part of the required analysis to complement and test the pathogenicity hypothesis obtained
through the bioinformatic approach, we developed the first controlled system for the study of this
infection. The system consists of a co-culture system in which frog skin microorgans are inoculated
with Bd zoospores. Based on extant expression data, we suggest possible reference genes for
further expression analyses and tested three of them experimentally. The transcript profile of the
early stages of infection (0, 12, 24, and 36hpi) of tolerant host (Rheobates palmatus) was followed
for three of pathogenicity factors (POP1, LMBRL and PhzF) and for a melanin synthesis protein. For
these three genes, we found experimental evidence of strong variations in transcript in very early
stages in the establishment of infection (0-12hpi), suggesting a possible role in the initial
interactions between Bd and its host.
In Colombia, which hosts the second most diverse Amphibian fauna in the world, the introduction
of Bd seems to be very recent and is hypothesised to be from two of the three suggested waves of
spread of Bd for South America. Due to the importance of the studying of Bd in Colombia as a way
to understand the epidemiology and potentially the biology of the disease, here we present the
genetic and morphological characterisation of the first Colombian isolate of Bd (strain EV001). Our
aim was to locate the Colombian gene pool in the global context and to gather initial evidence to
test the hypothesis of introduction of the disease to Colombia. For this we used Light and Scanning
electron microscopy and an MLST approach using 17 markers previously developed, providing 20
loci for further analysis, given that intralocus recombination was identified for three loci. Although
the morphological characteristics are congruent with the general features of Bd, we found broad
distance between our strain and the only extant South American isolate. Our results suggest a
frequency of recombination that might be higher than the observed for strains described previously

and provide evidence, although circumstantial, for the multiple introduction or sex for Northern
South America.
The comparative genomic exploratory analyses provided other proteins that although were not
recovered directly by our approach, are of great interest in understanding pathogenesis and the
evolutionary history of early diverging Fungi. Among these there are a Polyphenol oxidase (PPO1)
involved in melanin synthesis and a possible light-sensing rhodopsin. We were able to determine
the expression pattern of PPO1 using the co-culture system and found very low variation in mRNA
levels along the studies time frame. Research on this proteins and their corresponding family is
ongoing, includes collaboration with renowned members of the field (see supplementary research).
The comparative genomic exploratory analyses provided other proteins that although were not
recovered directly by our approach, are of great interest in understanding pathogenesis and the
evolutionary history of early diverging Fungi. Among these there are proteins involved in melanin
synthesis and members of the rhodopsin family We were able to determine the expression pattern
of the melanin synthesis protein using the co-culture system and found very low variation in mRNA
levels along the studied time frame. Research on these proteins and their corresponding families is
ongoing. Also, due to the training internship and conferences one of the researchers attended, we
were able to collaborate in research in which we applied approaches developed during this project
to another system (see Parallel molecular evolution in an herbivore community).
Publications:
Zhen, Y., Aardema, M.L., Medina, E.M., Schumer, M., Andolfatto, P. Parallel molecular
evolution in an herbivore community. 2012. Science (to be resubmitted)
Medina, E.M., Flechas, S.V., Sarmiento, C., Cárdenas, M., Crawford, A.J., Amézquita, A.,
Restrepo, S. Highly recombinant Colombian strain of Batrachochytrium dendrobatidis. 2011.
Ecohealth (to be resubmitted)
Medina, E.M., Ahrendt, S., Restrepo, S., Stajich, J.E. Because they were blind but now they
see!: animal-like light-sensing rhodopsins in early diverging Fungi. 2012. (in preparation)
Medina, E.M., Restrepo, S., Stajich, J.E. Melanin-mediated stress resistance in the amphibian-
killing pathogen Batrachochytrium dendrobatidis. 2012. (in preparation)

CROATIA
Title: Mechanisms of aminoglycoside resistance by ribosomal RNA methylation
Principal Investigator: Gordana Maravić Vlahoviček, Department of Biochemistry and Molecular
Biology, Faculty of Pharmacy and Biochemistry, University of Zagreb, Ante Kovačića 1, Zagreb,
Croatia. Tel: +385-1-639-4448, Fax: +385-1-639-4400, E-mail: gordana@pharma.hr
ICGEB Reference No.: CRP/CRO08-02
Abstract: Increasing and improper use of antibiotics in humans, animals, and agriculture has
resulted in emergence and spreading of multiply resistant bacterial pathogens. As a consequence,
many infectious diseases are increasingly difficult to treat, making the antibiotic resistance problem
one of the top priorities of public health. This project investigates the scarcely explored mechanism
of resistance to aminoglycoside antibiotics, found in both aminoglycoside producers and in growing
number of clinical strains. This type of resistance is generated through the action of
methyltransferases from the Arm and Kam family, which methylate the aminoglycoside interaction
site in 16S rRNA (Arm - m7
G1405; Kam - m1
A1408) using the S-adenosyl-methionine as a methyl
donor. Research objectives are directed towards the functional characterisation of the Arm and
Kam enzymes and their RNA substrate. Obtained results provide a notable insight into molecular
basis of aminoglycoside resistance, which will directly lead the future research initiatives towards
the finding of specific inhibitor of the Arm and Kam enzymes and thus contribute to the worldwide
efforts to combat antibiotic resistance in bacterial pathogens.
Objectives:
(i) to define the core motif of Arm and Kam RNA substrate;
(ii) to define RNA-binding modules for both enzyme families;
(iii) to propose the reaction mechanism of 16S rRNA:m7
G and rRNA: m1
A methylation.
All individual objectives of the project have been met and accomplished. Obtained results have
been published in four papers and two additional papers are currently in preparation.
Results Obtained:
In order to analyse the binding pattern of Arm and Kam enzymes to their cognate rRNA substrate,
we did the mutagenesis of the ribosomal A-site within the helix 44 of 16S rRNA using E. coli strains
with a single rRNA operon. We analysed the impact of the mutations on the function of all
members of Arm and Kam family by determining the minimal inhibitory concentration of kanamycin
and analysing the target nucleotide methylation with primer extension. These results enabled us to
define the core motif of Arm and Kam RNA substrate and to discover the differences among the
individual representatives of both families, as well as to improve the model of enzyme interaction
with the ribosomal subunit 30S.
We carried out extensive structural and functional analysis of the Sgm enzyme, as well as
sequence analysis of entire Arm family and related RNA methyltransferases (MTases). We showed
that the Sgm methyltransferase confers resistance to 4,6-disubstituted deoxystreptamine
aminoglycosides by introducing the 16S rRNA modification m7
G1405 within the ribosomal A site.
This region of Escherichia coli 16S rRNA already contains several methylated nucleotides including
m4
Cm1402 and m5
C1407. Modification at m5
C1407 by the methyltransferase RsmF is impeded as
Sgm gains access to its adjacent G1405 target on the 30S ribosomal subunit. An Sgm mutant
(G135A), which is impaired in S-adenosylmethionine binding and confers lower resistance, is less
able to interfere with RsmF methylation on the 30S subunit. The two methylations at 16S rRNA
nucleotide m4
Cm1402 are unaffected by both the wild-type and the mutant versions of Sgm. The
data indicate that interplay between resistance methyltransferases and the cell’s own indigenous
methyltransferases can play an important role in determining resistance levels confirmed the
guanine 1405 to be the methylation site of Sgm methyltransferase. We then solved the structure of
Sgm in complex with cofactors S-adenosyl-methionine and S-adenosyl-homocysteine at 2.0 and
2.1 Å resolution, respectively. We also analysed a number of mutants by isothermal titration
calorimetry, performed the rRNA footprinting by chemical probing of Sgm-30S subunit complex and
protein-substrate docking. We proposed the mechanism of methylation of G1405 by Sgm and
compared it with other m7
G methyltransferases, revealing a surprising diversity of active sites and
binding modes for the same basic reaction of RNA modification. The results of our analyses provide
a stepping stone for the design of inhibitors against the medically important Arm family of MTases.
In particular, the guanosine-binding site identified in the Sgm structure (and by extension, in
homologous structures of ArmA and RmtB MTases) can be used as receptor for structure-based
virtual screening.
Furthermore, we conducted the extensive structural and functional analysis of the NpmA enzyme,
as well as sequence analysis of entire Kam family and related RNA MTases. We solved the structure
of NpmA enzyme and of its complex with cofactors S-adenosyl-methionine and S-adenosyl-
homocysteine at 2.4, 2.7 and 1.68 resolution, respectively. We also analysed a number of mutants
by isothermal titration calorimetry, methylation test in vitro, studied their ability to generate
kanamycin resistance in vivo and performed the rRNA footprinting by chemical probing of Sgm-30S

subunit complex and protein-substrate docking. We proposed the mechanism of methylation of
A1408 by NpmA and compared it with other m1
A methyltransferases. Helices 24, 42 and 44 were
found to be the main NpmA-binding site. Both experimental and theoretical analyses suggest that
NpmA flips out the target nucleotide A1408 to carry out the methylation. Obtained results will
assist in the development of specific NpmA inhibitors that could restore the potential of
aminoglycoside antibiotics.
In our studies of the Sgm methyltransferase, where we confirmed the guanine 1405 to be the
methylation site of Sgm methyltransferase, we discovered the interconnection of Sgm enzyme with
the housekeeping methyltransferase RsmF.
We extended our research to specific application of the MTases from the Arm and Kam family in
the studies of bacterial quorum sensing. We investigated whether tobramycin sensing/response at
sub-inhibitory concentrations was affecting the two independent AHL QS systems in and
environmental Pseudomonas aeruginosa strain. It was established that sub-inhibitory
concentrations of tobramycin inhibited the RhlI/R system by reducing levels of C4-HSL production.
This effect was not due to a decrease of rhlI transcription and required tobramycin-ribosome
interaction, as confirmed by the action of RmtC, RmtA and NpmA MTases.
Networking:
The project enabled and promoted intensive collaboration with the following groups: (i) Dr. Janusz
Bujnicki (International Institute of Molecular and Cell Biology, Warsaw, Poland), (ii) Dr. Jayaraman
Sivaraman (National University of Singapore), (iii) Dr. Stephen Douthwaite (University of Southern
Denmark, Odense, Denmark) and (vi) Dr. Vittorio Venturi (ICGEB, Trieste, Italy).
Publications:
Čubrilo, S., Babić, F., Douthwaite, S., Maravić Vlahoviček, G. The aminoglycoside resistance
methyltransferase Sgm impedes RsmF methylation at an adjacent rRNA nucleotide in the ribosomal
A site. 2009. RNA. 15, 1492-1497
Babić, F., Venturi, V., Maravić Vlahoviček, G. Tobramycin at subinhibitory concentration
inhibits the RhlI/R quorum sensing system in a Pseudomonas aeruginosa environmental isolate.
2010. BMC Infect. Dis. 10,148
Husain, N., Tkaczuk, K., Tulsidas, S., Kaminska, K., Čubrilo, S., Maravić Vlahoviček, G.,
Bujnicki, J., Sivaraman, J. Structural basis for the methylation of G1405 in 16S rRNA by
aminoglycoside resistance methyltransferase Sgm from an antibiotic producer: A diversity of active
sites in m7
G methyltransferases. 2010. Nucleic Acids Res. 38(12), 4120-4132
Husain, N., Obranić, S., Koscinski, L., Seetharaman, J., Babić, F., Bujnicki, J., Maravić
Vlahoviček, G., Sivaraman, J. Structural basis for the methylation of A1408 in 16S rRNA by a
panaminoglycoside resistance methyltransferase NpmA from a clinical isolate. 2011. Nucleic Acids
Res. 39(5), 1903-1918

EGYPT
Title: plg-1 gene expressing propionicin by lactic starters in dairying
Principal Investigator: Mahmoud Khairy Tahoun, Department of Dairy Science and Technology,
Faculty of Agriculture, Alexandria University, Aflatoon Street, El-Shatby 21545, Alexandria, Egypt.
Tel: +20-3-5925405, Fax: +20-3-5922780, E-mail: samehgene@yahoo.com
ICGEB Reference No.: CRP/EGY07-03
Abstract: The expression of propionicin PLG-1 by lactic acid starter culture is the target of the
present research program, hence propionicin PLG-1 expressed by plg-1 gene of Propionibacterium
thoenii P127 was reported to inhibit several food-borne pathogens such as Campylobacter jejuni,
Escherichia coli, Pseudomonas fluorescens, Pseudomonas aeruginosa, Vibrio parahaemolyticus,
Trichoderma reesi, Fusarium tricinctum, Candida lipolytica, Aspergillus spp., Yersinia enterocolitica
and a strain of Corynebacterium sp. Such program depends on plg-1 gene transformation to
Lactococcus lactis subsp. lactis MG1613 and Lactobacillus plantarum TF103 as models of dairy
starters. This approach can be accomplished using pFG200, pEM76, pINT125 and pLEB590 vectors
derived from lactococcal DNA. The multicopy vector pFG200 contains the replicon fragment of
pCT1138, the amber suppressor supD from Lactococcus strain NJ1 as a food-grade selectable
marker, and a versatile polylinker. The vector pEM76 is based on an integrase (Int), a resolve-
invertase (β- recombinase) and their respective target sites (attP – attB and six, respectively). It
contains heterologous origin and antibiotic resistance markers surrounded by two directly oriented
six sites, a multiple cloning site, the int gene and the attP site of phage A2. pINT125 is a food-
grade vector system that allows stable integration of multiple plasmid copies into the chromosome
of Lactococcus lactis. The vector consists of the plus origin of replication (Ori+
) of the lactococcal
plasmid pWV01, the sucrose genes of the lactic acid bacterium Pediococcus pentosaceus PPE1.0 as
a selectable marker, a multiple-cloning site, and a lactococcal DNA fragment of a well-
characterised chromosomal region. Later single-crossover integration system could be a simple
procedure for the engineering of stable food-grade strains carrying multiple copies of a gene of
interest. pLEB590 is a food-grade cloning vector for lactic acid bacteria that constructed using the
nisin immunity gene nisI as a selection marker. It is constructed entirely of lactococcal DNA
containing the pSH71 replicon, the nisI gene, and the constitutive promoter P45 for nisI
expression. Such food-grade expression system has potential for expression of foreign genes in
lactic acid bacteria in order to construct improved starter bacteria for food applications.
Propionibacterium thoenii P127 (propionicin PLG-1 producer strain) was screened for its
antimicrobial activity against a wide range of microorganisms including, 14 Escherichia coli strains,
7 Propionibacterium strains, 3 yeast strains, 3 Lactococcus strains, 2 Lactobacillus strains, and one
strain of each of Bacillus sp., Clostridium sp., Streptococcus sp., Oenococcus sp., Listeria sp.,
Salmonella sp. and Aspergillus sp. The results indicated inhibition of above microbial strains with
different degrees. Above findings were found to contradict the results obtained by other
investigators. Therefore, total crude protein of the cell-free supernatant was partially purified and
retested again against above strains. The Results obtained indicated that total crude protein
strongly inhibited the growth of all Escherichia coli, Listeria sp., Salmonella sp. and yeasts strains
investigated. However, it moderate affected the growth of all propionibacteria strains expect
Propionibacterium freudenreichii and weakly inhibited the growth of all lactic acid bacterial strains.
However it showed negative effect on the growth of Aspergillus niger, Bacillus subtilis DB100 host,
and Clostridium sporogenes DSM1446. Therefore, the total crude protein extract was exposed to
SDS-PAGE and the band containing propionicin PLG-1 (~10 kDa) was extracted and retested
against all culture strains investigated at the beginning and revealed similar data to the total crude
protein extract. This was followed by biochemical tests to confirm the characterisation of the
propionicin PLG-1. On the genetic level, plg-1 gene expressing propionicin PLG-1 was isolated using
specific bioinformatically designed primers and its sample was sent oversees to be sequenced for
the first time. The resulted PCR product was visualised on DNA agarose gel (1.5%) and migrated
beside 1 KB DNA marker. After gel electrophoresis, the PCR product size was calculated to be ~
500 bp. Such product was isolated from the gel and purified and transferred to different lactic acid
bacterial strains (Lactococcus lactis subsp. lactis MG1613 and Lactobacillus plantarum TF103 as
models of dairy starters) using pLEB590 (as the best examined lactic acid bacterial vector selected
from a collection of different vectors including pFG200, pEM76, pINT125 and pLEB590 vectors
originated from lactococcal DNA) to give the modified vector pLEBPLG-1. The lactic acid bacterial
transformants showed strong antimicrobial activity against Escherichia coli DH5α, Listeria
monocytogenes 18116, and Salmonella entirica 25566. Such lactic acid bacterial transformants can
be used in dairy industry to control above food-borne pathogens that are largely distributed in
worm climates and to feed school children in the poor countries where epidemic diseases and
diarrhea prevails.
Objectives:

(i) Screening for antimicrobial activity of Propionibacterium thoenii P127 against some
microorganisms including bacteria, yeast and fungi;
(ii) Propionicin PLG-1 isolation, partially purification and characterisation;
(iii) Isolation of Propionibacterium thoenii P127 genomic DNA using different method;
(iv) Isolation of plg-1 gene of P. thoenii P127. Isolation of several cloning vectors for plg-1 transfer
into some strains of lactic acid bacteria;
(v) Digestion of PCR product using appropriate restriction endonucleases. Digestion of pLEB590
cloning vector with the same restriction endonucleases;
(vi) Ligation of digested plg-1 gene and digested pLEB590 cloning vector to construct the new
vector pLEBPLG-1;
(vii) Electroporation of pLEBPLG-1 to Lactococcus lactis subsp. lactis MG1613as well as
Lactobacillus plantarum TF103;
(viii) Examine the new transformants for their antimicrobial activity against some food-borne
pathogens. Examine the use of the transformants for their capacity to affect the growth of above
food-borne pathogens in milk environment.
Results Obtained:
(i) Propionicin PLG-1 of Propionibacterium thoenii P127 was isolated and partially purified and
characterised;
(ii) Such bacteriocin strongly inhibited different microorganisms including the pathogens
Escherichia coli DH5α, Listeria monocytogenes 18116, Salmonella entirica 25566;
(iii) Isolation of genomic DNA of P. thoenii P127 using different isolation methods;
(iv) Designing different primers for the isolation of plg-1 gene using different bioinformatics
programs;
(v) PCR product and the cloning vector pLEB590 were digested using appropriate restriction
endonucleases. Ligation of digested PCR product containing plg-1 gene and digested pLEB590
cloning vector in order to construct the new vector pLEBPLG-1 (3600 bp);
(vi) Electroporation of pLEBPLG-1 to Lactococcus lactis subsp. lactis MG1613 as well as
Lactobacillus plantarum TF103 as models of lactic acid bacteria was successfully performed;
(vii) The new lactic acid bacterial transformants showed strong antimicrobial activity against above
pathogens;
(viii) The use of above new transformants in milk environment and their capacity to affect the
growth of above food-borne pathogens specially in milk was successfully performed.
Publications:
Tahoun, M.K., Mohamed, S.E. Propionicin PLG-1, screening and partial characterisation of the
protein. 2010. Int. J. Biotech. Biochem. 6, 325–338
Mohamed, S.E., Tahoun, M.K. Heterologous expression of pctA gene expressing propionicin T1
by some lactic acid bacterial strains using pLEB590. 2011. J. Agric. Biol. Sci. (in press)
Tahoun, M.K., Mohamed, S.E. Heterologous expression of pctA gene expressing propionicin T1
by some lactic acid bacterial strains using pINT125. 2011. Adv. Biosci. Biotechnol. (in press)

HUNGARY
Title: Assessing the protein folding-aggregation paradigm with NMR and molecular modeling
Principal Investigator: András Perczel, Eötvös Loránd University, Institute of Chemistry,
Pázmány Péter sétány 1/A, H-1117 Budapest, Hungary. Tel: +36-1-3722500, ext. 1653, Fax:
+36-1-3722620, E-mail: perczel@chem.elte.hu, Website:
www.chem.elte.hu/departments/protnmr
ICGEB Reference No.: CRP/HUN08-03
Abstract: We have investigated issues related to protein folding and aggregation. Our model
systems comprised (mini)proteins and peptides of general and special biological interest,
investigated at atomic level by CD and NMR spectroscopy, compiled with theoretical works on
molecular modeling, quantum chemical calculations and bioinformatics. We have completed the
rational design of selected Trp-cage miniproteins and have further optimised their stability. The
variants were selected to exhibit either more stable structures, inducible aggregation or
aggregation-prone behavior. Their structure, dynamics and folding was investigated by homo- and
heteronuclear NMR and theoretical methods. Furthermore, additional model systems were studied
such as intrinsically disordered proteins or IDPs (PP25, calpastatin fragments, highly charged
sequences, etc.) as well as protein domains of variable sizes (CCP and DLC). In addition,
supramolecular assemblies of both alpha- and beta-peptides were studied. Our present results are
hoped to contribute to the general understanding of protein folding and stability, as well as to the
specific description of several biologically important molecular systems spelled out above.
Objectives:
Our aims were to investigate several of the principles involved in determining the structure and
dynamics of folded and intrinsically disordered natural proteins and foldamers, as well as to gain
insight into the structure-dynamic-function relationships of selected biomolecules at the atomic
level.
Results Obtained:
(A) Secondary structure elements and foldamers
Quantum chemical calculations contributed to the theoretical aspects of structure analyses. On one
hand, the collagen triple-helix model was studied and its stabilisation and elasticity was reported to
be due to the presence of bridging water molecules (Pálfi and Perczel, 2010). On the other hand,
the structure and stability of sheet-like conformers of beta-peptides were studied and novel
foldamers were reported, which secondary structural elements are structural analogs of beta-
pleated sheets of proteins and are enzymatically resistant biomaterials (Beke, et al., 2009).
NMR is a powerful tool for analysing proteins especially when well folded. However, traditional
techniques based on NOEs, J-coupling constants, H-bonds etc. fail to perform well for highly
dynamic proteins, especially for IDPs. In these cases secondary chemical shifts are of great help to
decipher information on the backbone fold (if any). These Δδ-type experimental techniques
however require more computational efforts. Thus, additional ab initio NMR chemical shift
calculations were completed on major types of backbone folds and subsequently compared to the
corresponding experimental counterparts retrieved from selected protein databases and adjusted
carefully. We have found acceptable/good agreement for the most common backbone conformers
(e.g. helices, β-sheets, PPII). Experimental and computed ΔδHa
- ΔδCa
, ΔδHa
- ΔδCb
, and ΔδCa
- ΔδCb
correlation maps provided qualitatively realistic pictures, yielding help to estimate the dominant
backbone fold of even dynamic proteins from CSIs (Czajlik, et al., 2011).
Using first principles calculations, we have characterised the hydration layers of collagen in detail.
Our results suggest that the water binding places on the surface of the triple helix can provide
explanation on how an almost liquid-like hydration environment exists between the closely packed
tropocollagens. It seems that these water reservoirs or buffers can provide space for ‘‘hole
conduction’’ of water molecules and thus contribute to the elasticity of collagen (Pálfi and Perczel,
2010).
We have conducted a matrix isolation conformational study on model beta peptides, N-acetyl-3-
aminopropionic acid-N′-methylamide (Ac-β-HGly-NHMe) and N-acetyl-3-aminobutanoic acid-N′-
methylamide (Ac-β-HAla-NHMe). The IR spectra recorded suggest that although β-peptides having
an “extra” backbone torsion should be more flexible than α-peptides, fewer backbone conformers
are favorable based on their relative energies. Thus, from a larger conformational arsenal, only a
lower number of backbone conformers can emerge, which possibly had a fundamental effect on
their applicability during prebiotic evolution (Beke, et al., 2009).
Foldamers and the replacement of α-amino acid residues by β- or γ-residue types is a fascinating
and modern approach to stabilise peptide backbones. In order to predict the most probable
secondary structures of β-peptides, a systematic QM modeling project was launched early on in our
laboratory. More recently, calculations were carried out on artificial β-peptides in order to deeply
understand their conformational behavior. A systematic study on the effect of chirality on ‘‘apolar’’
sheet stability was completed on disubstituted [HCO- β-Ala- β2,3
-hAla- β-Ala-NH2]2 model peptides

calculated at the M05-2X/6-311++G(d,p)//M05-2X/6-31G(d) and B3LYP/6-
311++G(d,p)//B3LYP/6-31G(d) levels of theory, both in vacuum and in polar and apolar solvents.
In addition, both types of ‘‘apolar’’ sheets were investigated; the one with two strands of identical
(AA) and enantiomeric (AB) backbone structure. Our results show that not only heterochirally
disubstituted sheets have great preference for sheet formation (ΔG ~ -11 kcal mol-1
), but
unexpectedly ‘‘homochiral disubstitution’’ itself could result in a stable fold (ΔG ~ -5 kcal mol-1
).
Thus, it does not necessarily disrupt the sheet structure. The predictive nature of the effect of
chirality on sheet stability offers an easy-to-use summary on how to design β-peptide sheet
structures. It was found that heterochirally disubstituted models (“lego elements”) are the best
candidates for sheet formation, if the two strands are substituted in a way to create identical
torsional angle sets on the two backbones for an ideal hydrogen-bonding pattern. With adequately
selected side chains, homochirally disubtituted derivatives may also form sheet structures, and the
position of methyl groups would prevent the assembly of more than two strands making it ideal to
create hairpin folds (Pohl, et al., 2012, Pohl, et al., in press).
(B) Miniproteins
We have proceeded with the design, production and characterisation of Trp-cage miniprotein
variants. One of the concepts applied is to generate longer variants of the model protein Tc5b
according to its ‘parent molecule’, exendin 4. We have prepared a series of Tc5b variants elongated
at its N-terminus by one residue at each step. Detailed characterisation (NMR measurements,
structure calculations etc.) of these molecules is in progress. In addition, other Tc5b variants have
been characterised using state-of-the-art dynamic structure calculation methods (MUMO protocol
as implemented in GROMACS). Evaluation of the results is in progress. Additionally, we started to
redesign a Trp-cage miniprotein by one or two mutations to obtain a genuine IDP. Based on
rational considerations and extensive bioinformatical investigations, the biotechnological production
of selected sequences has been started.
One of the variants, Tc5b-D9Aad has been extensively analysed with CD spectroscopy using CCA+,
a program developed in our group. Convolution analysis revealed that the CD spectrum this variant
can be adequately described by two components, one corresponding to the ordered and the other
to the disordered state (Jákli and Perczel, 2009).
We have significantly proceeded with the design and analysis of novel Tc5b variants. We have
detailed dynamical data for selected variants at different temperatures, for which the calculation
and analysis of dynamically restrained structural ensembles is in progress. In addition, we have
several double mutants at hand that behave like intrinsically unfolded/dynamic proteins. Detailed
characterisation of these variants and the design and preparation novel ones is under way. Our
present data promise understanding of the role of interaction networks within Tc5b from a dynamic
perspective.
We have continued our extended research on miniproteins by investigating the folding and stability
role of a salt-bridge (Asp9
–Arg16
) within the molecule. The point mutations identified that both of
the above side-chains are also part of a dense interaction network composed of various types of
nature. The fold stabilising effects were judged by electronic circular dichroism, ECD, and NMR data
of these mutants at different pHs. Our present results indicate that the folding of Trp-cage
miniproteins is a complex process via an intermediate state related to the salt-bridge stabilisation.
In this molecular framework, Arg16
stabilises more than Asp9
does because of its higher degree of
3D-fold cooperation. The Xxx9
↔Yyy16
salt-bridge is not an isolated entity of this fold; rather it is
an integrated part of a complex interaction network (Rovó et al. 2011).
(C) Modular folded protein(s)
(C.i) Small serine protease inhibitors: Using NOE and backbone S2
data characteristic of ps-ns
time-scale motions, we have generated dynamically restrained protein structural ensembles of
small serine protease inhibitors and investigated the relation of the conformer sets to the available
crystal structures of the enzyme-bound forms of the inhibitors. We found that the dynamically
restrained conformer sets reproduce most NMR experimental data better than structures obtained
using “conventional” structure calculation methods, Moreover, they also include conformers highly
similar to those observed in the enzyme complexes, indicating that enzyme binding of small, 35-
residue inhibitors is actually a dynamic conformer selection process on the nanosecond-timescale.
Thus, fast timescale dynamics enables the association rate to be solely diffusion-controlled just like
in the rigid-body model traditionally used to describe the mechanism of canonical serine protease
inhibitors (Gáspári, et al., 2010a).
(C.ii) Complement control modules: We have focused on the dynamical determination of protein
molecules especially those of the two CCP modules from human C1r of innate immunity by NMR
spectroscopy. We have also identified the flexible parts of the CCP1-CCP2 module pair that might
be responsible for interactions with third molecules and those regions whose flexibility could be
modulated by partner binding to the neighboring module. Furthermore, we showed that the two
separated modules interact with each other, which observation might have consequences regarding
the solution-state dimerisation of full C1r molecules and thus can contribute to further
understanding the details of C1 activation (Láng, et al., 2010a, 2010b).
(C.iii) Modules in protein evolution: We have participated in providing an overview of the SBASE
database and server, capable of identifying domains in modular proteins from sequence. SBASE is

of particular interest in our current and planned studies on protein evolution as its remarkable
feature is its capability to detect atypical members of domain families, which exhibit only a
moderate similarity to most other members (Dhir, et al., 2010).
(C.iv) dUTPase: We have solved the 1.8 Ǻ-resolution structure of a myocbacterial dUTPase enzyme
with and Asp/Asn mutation. Our results show that the role of this residue is not in Mg2+
coordination, as previously thought, rather contributing to other types of intra- and intersubunit
contacts. Our results contribute to the understanding of elaborate residue-residue interaction
networks in this protein family (Takács, et al., 2010).
(D) Intrinsically disordered proteins
To explore the evolutionary significance of protein disorder and aggregation, we have started to
generate random sequences systematically and to select those with no detectable similarity to
known proteins but which are predicted to be well-folded or on the contrary, could be expected to
be genuine IDPs. This work is in progress.
We have also conducted a model study to assess the extent and relevance of cross-predictions
between folded and unfolded sequences. As a model for the folded state, we have chosen coiled-
coil sequences as these adopt a well-defined structured state and are relatively straightforward to
predict based on their regularity. Based on our benchmark tests and survey involving the Swiss-
Prot database, we propose the simultaneous use of the programs Coils and IUPred to achieve
acceptable prediction accuracy and minimise the extent of cross-predictions. The relevance of
observed cross-predictions might be that disordered sequences can adopt coiled-coil conformation
relatively easily during protein evolution. (Szappanos, et al., 2010).
A newly recognised protein structural motif, the charged single alpha-helix (CSAH) shows
similarities both to coiled coils and intrinsically disordered segments. We have investigated these
segments in more detail, set up a web server for their detection (http://csahserver.chem.elte.hu)
and provided a discussion on their possible evolutionary transitions in relation to coiled coils
(Gáspári, et al., 2012, Gáspári and Nyitray, 2011).
Structure elucidation by NMR was completed for several of the envisaged proteins. The main goal
of these studies is to provide a link between atomic resolution structural information and biological
function. Concerning the intrinsically disordered proteins (IDPs) the most recent target of us was
dehydrin, termed as ERD14, from Arabidopsis thaliana, which protein is highly expressed in
somatic cells upon dehydration elicited by water stress, high salinity or cold. The backbone 1
H, 15
N
and 13
C resonance assignment of the 95% of the 185 amino acid-long protein was completed. Its
secondary chemical shifts as well as NMR relaxation data were determined and showed that ERD14
is indeed a fully disordered protein under near native conditions, with very short regions of
somewhat restricted motion and about 5–15% helical propensity. These results suggested that
ERD14 may have partially preformed segmental elements for functional interaction with its
partner(s) (Ágoston, et al., 2011).
(E) Analysis of protein structural ensembles reflecting experimentally determined dynamics
Our previous work on the use of structural ensembles reflecting the internal dynamics of proteins
on different time scales is further elaborated by implementing a web server for the evaluation of
such ensembles. The CoNSEnsX (Consistency of NMR-derived Structural Ensembles with
eXperimental data) method is freely available at http://consensx.chem.elte.hu (Ángyán, et al.,
2010). We have also performed a pilot study to analyse the effect of using dynamically restrained
ensembles (instead of single conformers) for structural bioinformatics analysis (CX, DPX, PRIDE
and PRIDE-NMR methods). All these investigations prompted us to invoke the idea termed
Dynamics–Structure–Activity Relationships (DSARs) of proteins (Gáspári, et al., 2010b, Gáspári
and Perczel, 2010).
Dai, X.Y., Gu, X.C., Luo, M., Zheng, X.F. Protein expression, crystallisation and preliminary X-ray
crystallographic studies on HSCARG from Homo sapiens. 2006. Protein Pept. Lett. 13, 955-957
Publications:
Beke, T., Somlai, C.S., Magyarfalvi, G., Perczel, A., Tarczay, G.Y. Chiral and achiral
fundamental conformational building units of β-peptides: a matrix isolation conformational study
on Ac- β -HGly-NHMe and Ac- β -HAla-NHMe. 2009. J. Phys. Chem. B., 113, 7918-7926
Jákli, I., Perczel, A. The inherent flexibility of peptides and protein fragments quantitised by CD
in conjunction with CCA+. 2009. J. Pept. Sci. 15, 738-752
Ángyán, A.F., Szappanos, B., Perczel, A., Gáspári, Z. CoNSEnsX: an ensemble view of protein
structures and NMR-derived experimental data. 2010. BMC Struct. Biol. 10, 39
Dhir, S., Pacurar, M., Franklin, D., Gáspári, Z., Kertész-Farkas, A., Kocsor, A., Eisenhaber,
A., Pongor, S. Detecting atypical examples of known domain types by sequence similarity
searching: the SBASE domain library approach. 2010. Curr. Prot. Pept. Sci. 11, 515-549
Gáspári, Z., Perczel, A. Protein dynamics as reported by NMR. 2010. In: Annual reports on NMR
spectroscopy. Academic Press 71, 35-75
Gáspári, Z., Várnai, P., Szappanos, B., Perczel, A. Reconciling the lock-and-key and dynamic
views of canonical serine protease inhibition. 2010a. FEBS Lett. 584, 203-206
Gáspári, Z., Ángyán, A.F., Dhir, S., Franklin, D., Perczel, A., Pintar, A., Pongor, S. Probing
dynamic protein ensembles with atomic proximity measures. 2010b. Curr. Prot. Pept. Sci. 11, 515-
522

Láng, A., Szilágyi, K., Major, B., Gál, P., Závodszky, P., Perczel, A. Intermodule cooperativity
in the structure and dynamics of consecutive complement control modules in human C1r:
structural biology. 2010a. FEBS J. 277, 3986-3998
Láng, A., Major, B., Szilágyi, K., Gáspári, Z., Gál, P., Závodszky, P., Perczel, A. Interaction
between separated consecutive complement control modules of human C1r: implications for
dimerisation of the full-length protease. 2010b. FEBS Lett. 584, 4565-4569
Pálfi, V.K., Perczel, A. Stability of the hydration layer of tropocollagen: a QM study. 2010. J.
Comput. Chem. 31, 764-777
Pohl, G., Beke, T., Csizmadia, I.G., Perczel, A. Extended apolar beta-peptide foldamers: the
role of axis chirality on beta-peptide sheet stability. 2010. J. Phys. Chem. B. 114, 9338-9348
Szappanos, B., Süveges, D., Nyitray, L., Perczel, A., Gáspári, Z. Folded-unfolded cross-
predictions and protein evolution: the case study of coiled-coils. 2010. FEBS Lett. 584, 1623-1627
Takács, E., Nagy, G., Leveles, I., Harmat, V., Lopata, A., Tóth, J., Vértessy, B.G. Direct
contacts between conserved motifs of different subunits provide major contribution to active site
organisation in human and mycobacterial dUTPases. 2010. FEBS Lett. 584, 3047-3054
Ágoston, B.S., Kovács, D., Tompa, P., Perczel, A. Full backbone assignment and dynamics of
the intrinsically disordered dehydrin ERD14. 2011. Biomol. NMR Assign. 5, 189-193
Czajlik, A., Hudáky, I., Perczel, A. Local protein backbone folds determined by calculated NMR
chemical shifts. 2011. J. Comput. Chem. 32, 3362-3382
Gáspári, Z., Nyitray, L. Coiled coils as possible models of protein structural evolution. 2011.
BioMol. Concepts 2, 199-210
Rovó, P., Farkas, V., Hegyi, O., Szolomajer-Csikós, O., Tóth, G.K., Perczel, A. Cooperativity
network of Trp-cage miniproteins: probing salt-bridges. 2011. J. Pept. Sci. 17, 610-619
Gáspári, Z., Süveges, D., Perczel, A., Nyitray, L., Tóth, G. Charged single alpha-helices in
proteomes revealed by a consensus prediction approach. 2012. Biochem. Biophys. Acta 1824,
637-646
Pohl, G., Jakli, I., Csizmadia, I.G., Papp, D., Matıas, G.F., Perczel, A. The role of entropy in
initialising the aggregation of peptides: a first principle study on oligopeptide oligomerisation.
2012. Phys. Chem. Chem. Phys. 14, 1507–1516
Pohl, G., Beke-Somfai, T., Csizmadia, I.G., Perczel, A. Exploiting diverse stereochemistry of β-
amino acids: toward a rational design of sheet-forming β-peptide systems. 2012. Amino Acids (in
press)

MEXICO
Title: Signal transduction by the BarA/UvrY two component system
Principal Investigator: Dimitrios Georgellis, Lab 226N, Molecular Genetics Department,
Universidad Nacional Autónoma de México, Instituto de Fisiología Celular, Circuito Exterior S/N,
Col. Copilco, México D.F., Mexico. Tel: +52-55-56225738, Fax: +52-55-56225611, E-mail:
dimitris@ifc.unam.mx
ICGEB Reference No.: CRP/MEX08-02
Abstract: The animal and plant pathogens of the gamma subdivision of Proteobacteria cause vast
amounts of agricultural damage and human disease. These pathogens include members of the
genera Pseudomonas, Erwinia, Escherichia, Vibrio, and Salmonella. The individual species among
these genera are very diverse, as they include free-living species, symbionts, plant pathogens, and
animal pathogens. However, they also have striking similarities. For instance, in all five genera the
BarA/UvrY two-component system (TCS) has been shown to be required for virulence.
Despite this realisation, two observations motivate us to propose that the primary function of the
BarA/UvrY TCS has not yet been discovered. Firstly, this system is found in both pathogens and
non-pathogens. Secondly, although uvrY is encoded within an evolutionarily conserved region of
the various genomes, the virulence genes that it regulates are specific to each pathogen. This
strongly suggests that uvrY was present in these genomes before the acquisition of the virulence
genes that it now controls, and therefore the regulation of these genes must be a relatively new
function for the UvrY orthologs. Hence, UvrY must have a more ancient and evolutionarily
conserved function(s) that remains to be discovered.
We propose to probe for the structural requirements of BarA for signal reception, to elucidate the
routes of phosphoryl-group transfer for signal transmission and signal decay, to search for the
member genes of the UvrY regulon, to identify the conserved functions of the various UvrY
orthologs, and to screen libraries of small molecules for potential antibacterials that specifically
target TCSs. To address these questions we will primarily use E. coli as the model organism, and a
wide array of biochemical and genetic techniques.
Objectives:
(i) Probe for the physiological signals that stimulate BarA: We intend to use biochemical and
genetic analyses in our searches for the physiological signal(s) that stimulate BarA;
(ii) Probing for the structural requirements of BarA for signal reception;
(iii) Structure function relationships in the BarA/UvrY two component system: Biochemical and
genetic analyses will be undertaken to probe for the structural requirements of BarA for signal
reception, and to elucidate the routes of phosphoryl-group transfer for signal transmission and
decay;
(iv) Search for BarA/UvrY controlled genes: For identifying genes that belong to the UvrY regulon we
plan to use a modified protocol of the chromatin immuno-precipitation (ChIP) genome based
search. In order to optimise the output of this method the ChIP will be combined with a microarray
(ChIP on chip);
(v) Conserved functions of the various ortholog systems: The UvrY binding site on the promoter of
csrB, csrC, and other genes identified under (d) will be located by foot-printing assays, validated
by site directed mutagenesis, and a consensus sequence will be derived. Finally, the derived motif
will be used for searches of potential UvrY controlled promoters on different genomes;
(vi) Antibacterial agents that inhibit signal transduction systems: Two-component systems are
considered potential targets for antibacterials. Therefore, we plan to use two independent methods
to screen libraries of small molecules with the intention of identifying antimicrobial agents that
specifically target these systems.
Results Obtained:
The BarA/UvrY two-component system activates transcription of CsrB and CsrC noncoding RNAs,
which act by sequestering the RNA-binding global regulatory protein CsrA. Previously, we
demonstrated that pH lower than 5.5 provides an environment that does not allow activation of the
BarA/UvrY signaling pathway (Mondragón, V., et al. 2006. J. Bacteriol. 188, 8303-8306). Based on
this finding and using genetic and microbial physiology approaches we searched for the
physiological signal for the BarA sensor kinase. We discovered that the key metabolites, formate,
acetate as well as other aliphatic short chain fatty acids (propionate butyrate, valerate and
caproate) act as stimuli that activate the BarA/UvrY two-component system. We also showed that
acetate might stimulate UvrY directly via the formation of acetyl-P, thereby linking BarA/UvrY
signaling as another target of the “acetate switch”. Finally, we demonstrate that acetate signalling
via the BarA/UvrY system is physiologically relevant since extracellular acetate accumulation in the
growth medium correlated with BarA/UvrY activation. Thus, our results link post-transcriptional
regulation by the Csr system to the metabolic state of the cell (Gonzalez Chavez, R., et al. 2010. J.
Bacteriol. 192, 2009-2012).

Furthermore, in two subsequent reports, we demonstrate that the BarA/UvrY TCS is integrated in a
complex regulatory cascade involving the Csr and stringent response global regulatory systems,
and also HilD, which controls expression of the Salmonella SPI-1 and SPI-2 virulence regulons
(Edwards, A.N., et al. 2011. Mol. Microbiol. 80, 1561-1580; Martínez, L.C., et al. 2011. Mol.
Microbiol. 80, 1637-1656).
Publications:
Alvarez, A., Georgellis, D. In vitro and in vivo analysis of the ArcB/A redox signaling pathway.
2010. Methods Enzymol. 471, 205-228
Gonzalez Chavez, R., Alvarez, A., Romeo, T., Georgellis, D. The physiological stimulus for the
BarA sensor kinase. 2010. J. Bacteriol. 192, 2009-2012
Barba-Ostria, C., Lledías, F., Georgellis, D. The Neurospora crassa histidine kinase DCR-1 is
required for normal sexual and asexual development, carotenogenesis, and circadian rhythmicity.
2011. Eukaryot. Cell 10, 1733-1739
Edwards, A.N., Patterson-Fortin, L.M., Vakulskas, C.A., Mercante, J.W., Potrykus, K.,
Vinella, D., Camacho, M.I., Fields, J.A., Thompson, S.A., Georgellis, D., Cashel, M.,
Babitzke, P., Romeo, T. Circuitry linking the Csr and stringent response global regulatory
systems. 2011. Mol. Microbiol. 80, 1561-1580
Martínez, L.C., Yakhnin, H., Camacho, M.I., Georgellis, D., Babitzke, P., Puente, J.L.,
Bustamante, V.H. Integration of a complex regulatory cascade involving the SirA/BarA and Csr
global regulatory systems that controls expression of the Salmonella SPI-1 and SPI-2 virulence
regulons through HilD. 2011. Mol. Microbiol. 80, 1637-1656
Luévano-Martínez, L.A., Barba-Ostria, C., Georgellis, D., Rial, E., Uribe-Carvajal, S. A
critical tyrosine residue of the mitochondrial oxaliacetate carrier determines its uncoupling protein
(UCP)-like function in yeast. 2012. Biochemical J. (in press)

MEXICO
Title: Human health and melon biotechnology
Principal Investigator: Miguel A. Gómez Lim, CINVESTAV-Irapuato, Km 9.6 Carretera Irapuato-
León, Irapuato, Guanajuato, Mexico. Tel: +52-462-6239679, E-mail: mgomez@ira.cinvestav.mx
ICGEB Reference No.: CRP/MEX09-01
Abstract: Many epidemiology studies have shown that there exists a relationship between dietary
habits and disease risk, indicating that the type of food consumed has a direct impact on health.
Biofortification, the process of enriching the nutrient content of crops as they grow, provides a
sustainable solution to worldwide malnutrition, because other methods, such as diversifying
people’s diets or providing dietary supplements, have proved impractical, especially in developing
countries. Carotenoids provide essential phytonutrients, such as provitamin A and confer many
health benefits, such as reduction in incidence of certain diseases, including cancer, cardiovascular
diseases, and age-related eye diseases. Developing carotenoid-enriched food offers a sustainable
avenue for improving human nutrition and health. Melon fruit is one of the most important fruits of
the world. They contain a cohort of antioxidant enzymes. In this project we have transformed
melon with two genes that increase the contents of carotenoids. Five lines were analysed by
Southern blot, northern blot and by GUS staining. The molecular assays performed indicate that
they contain and express both genes.
Objectives:
(i) Construction of vectors for expression of CHYB and Or genes under the control of the
CaMV promoter;
(ii) Expression of the constructs from in transgenic melon plants.
Results Obtained:
We have identified five melon lines that contain and express both genes.

PAKISTAN
Title: Genetic basis of hereditary hearing impairment
Principal Investigator: Sheikh Riazuddin, Centre for Molecular Biology, 87-West Canal Bank
Road, Lahore-53700, Pakistan & Iqbal Medical College, Jinnah Hospital Complex, Lahore-54550,
Pakistan. Tel: +92-42-35161710, Fax: +92-42-5164155, E-mail: riaz@lhr.comsats.net.pk;
riazuddin@aimrc.org
ICGEB Reference No.: CRP/PAK08-01
Abstract: The prevalence of genetic deafness has been estimated on a worldwide basis as 1 per
1000; however, in Pakistan and Middle East, where consanguinity is a social requirement, the
prevalence is 40-50% higher than the world average. Mapping of loci and identification of genes
are helpful in understanding the functioning of the inner ear, where sound waves are converted
into electrical pulses, which are translated in the brain as hearing perception. This is accomplished
by a mosaic of supporting cells and sensory hair cells, which are interconnected by uniquely
complex tight junction proteins such as Tricellulin.
One hundred large consanguineous families were identified and enrolled from ethnically distinct
areas of Pakistan bordering Afghanistan and Iran. DNAs isolated from 1250 blood samples were
genetically analysed to identify 25 new/novel mutations, and five new genetic loci. A deaf mouse
model was developed to study the malfunctioning of tight junction protein, tricellulin. Clinical
examination of the deaf mouse revealed that mutated tricellulin gene affects the functioning of
salivary gland, thyroid follicles, heart functioning, epithelia and inner hair cells in the cochlea.
Hearing impairment can lead to impaired ability to develop speech, language and effective
communication skills. Identification of the auditory gene, characterisation of pathogenic mutations
in these genes and genotype/ phenotype correlation in the deaf mouse is expected to elucidate the
molecular basis of recessive deafness caused by malfunctioning of the inner ear. This information
will be helpful in developing preventive strategies.
Objectives:
Aim 1: Search for new loci and pathogenic mutations associated with human hearing impairment
(1.i) Family identification and enrolment;
(1.ii) Clinical evaluation;
(1.iii) Sample collection;
(1.iv) Genotyping to find the linkage to known loci;
(1.v) Mutations identified;
(1.vi) Genome wide linkage analysis/high through put sequencing.
Aim 2: Generation of DFNB49 gene knockin and knockout transgenic mouse model and elucidation
of the role of tricellulin in hearing impairment
(2.i) DFNB49 knock-in construct;
(2.ii) DFNB49 conditional knockout construct;
(2.iii) Electroporation of the above constructs into 129vsJES cells of the mouse;
(2.iv) Selection of positive ES cells clones;
(2.v) Microinjection of positive ES cells into blastocysts to produce chimeric animals;
(2.vi) Evaluation of the hearing and balance function in Tric R497X mice.
Results Obtained:
One hundred large consanguineous families were identified and enrolled in these studies. All
members were clinically evaluated to determine if the deafness is syndromic or non-syndromic.
Blood samples (1250) were collected from all affected individuals, unaffected siblings, parents and
grandparents, if alive. Further, 300 normal random samples were collected from the same
population for comparative purposes. Genes were sequenced for mutational analysis and 25 new
mutations were discovered. Ten selected unlinked families were screened by whole genome scan
and high throughput sequencing, resulting in the discovery of five new loci.
DFNB49 Knock-in and conditional knock-out constructs were developed and used in raising
Tricellulin deaf mice. The said two constructs were electroporated into specific mouse cells. Positive
clones were selected and introduced by microinjection into blastosists to produce chimeric animals.
The resultant animals were examined for the growth of heart, spleen, liver and kidney, which were
all found heavier than those wild type animals. The level of total protein, phosphorous, albumin
and triglycerides were also higher in heterozygotes. Enlarged nuclei were seen in a large number of
myocytes, as a consequence, the animal suffered from myocardial hypertrophy.
The project went smoothly. There was no serious constraint. All the family members were fully
cooperative and enrolled voluntarily. They all signed consent forms approved by the institutional
Ethical Review Board.
Publications:
Anwar, S., et al. SLC26A4 mutation spectrum associated with DFNB4 deafness and Pendred's
Syndrome in Pakistanis. 2009. J. Hum. Genet. 54(5), 266-270

Rehman, A.U., et al. Targeted capture and next-generation sequencing identifies C9orf75,
encoding taperin, as the mutated gene in nonsyndromic deafness DFNB79. 2010. Am. J. Hum.
Gent. 86, 378-388
Borck, G., et al. Loss-of-function mutations of ILDR1 cause autosomal-recessive hearing
impairment DFNB42. 2011. Am. J. Hum. Genet. 88(2), 127-137
Rehman, A.U., et al. Mutations of GIPC3 cause nonsyndromic hearing loss NB72 but not DFNB81
that also maps to chromosome 19p. 2011. Hum. Genet. 130(6), 759-765
Waryah, A.M., et al. Molecular and clinical studies of X-linked deafness among Pakistani families.
2011. J. Hum. Genet. 56(7), 534-540
Ali, R.A., et al. DFNB86, a novel autosomal recessive non-syndromic deafness locus on
chromosome 16p13.3. Clin. Genet. (in press)
Nayak, G.D., Belyantseva, I.A., Yousaf, R., Edelmann, S., Lee, S., Van Itallie, C.M., Rafeeq,
M., Sinha, G., Anderson, J.M., Forge, A., Frolenkov, G.I., Riazuddin, S. Tricellulin is required
to maintain the barrier function of the reticular lamina and protect the cochlear hair cells from
potassium toxicity (in preparation)
Nayak, G.D., Rafeeq, M., Trincot, C.E., Khan, S.N., Riazuddin, S., Ahmed, Z.M., Riazuddin,
S. Genotype-phenotype correlation in DFNB49 families with TRIC mutations (in preparation)

PERU
Title: An ELISA assay for cysticercosis based on a recombinant cathepsin-L
Principal Investigator: Mirko Zimic, Laboratorio de Bioinformática y Biología Molecular,
Departamento de Ciencias Celulares y Moleculares, Facultad de Ciencias y Filosofía, Universidad
Peruana Cayetano Heredia, Av. Honorio Delgado 430, SMP Lima, Peru. Tel: +51-1-3190000, ext.
2604, E-mail: mirko.zimic@upch.pe
ICGEB Reference No.: CRP/PER08-02
Abstract: Human neurocysticercosis (NCC), caused by the cestode Taenia solium, is responsible
for a significant amount of neurological morbidity and epilepsy in developing countries. The disease
remains highly endemic in many areas, despite several efforts and interventions to control it.
Cathepsin L-like proteases are secreted by several parasites including Taenia solium. The
mechanism used by T. solium oncospheres to degrade and penetrate the intestine and infect the
host is incompletely understood. It is assumed that intestinal degradation is driven by the
proteolytic activity of enzymes secreted by the oncosphere. Blocking the proteolytic activity by an
antibody response would prevent the oncosphere penetration and further infection. Serine and
cysteine proteases including chymotrypsin, trypsin, elastase, and cathepsin L, are secreted by T.
solium and Taenia saginata oncospheres when cultured in vitro, being potential vaccine candidates.
However, the purification of a sufficient quantity of proteases secreted by oncospheres to conduct a
vaccine trial is costly and lengthy.
A 53/25kDa cathepsin L-like fraction partially purified from T. solium cyst fluid was described
previously by our group as an important antigen for immunodiagnostics in a Western blot and in an
ELISA.
We initially explored the functional role of the 53/25 kDa cathepsin L-like fraction in the
cysticercus. Immunohistochemistry and confocal microscopy were used to localise the 53/25 kDa
cathepsin L-like fraction in the brain tissue of naturally infected swine. The effect of this fraction on
the proliferation of human peripheral blood mononuclear cells and its capacity to degrade human
IgG and bovine serum albumin were tested.
The 53/25 kDa cathepsin L-like fraction was detected in the internal face of the cyst membrane
and was seen to migrate to the external face during progression of inflammation. The 53/25 kDa
cathepsin L-like fraction did not induce proliferation of human peripheral blood mononuclear cells
and was unable to degrade human IgG or BSA.
We confirmed that the 53/25 kDa cathepsin L-like fraction purified from the cysticercus fluid is also
present in the T. solium oncosphere and is secreted by the cysticercus. This protein fraction was
tested for its ability to protect pigs against an oral challenge with T. solium oncospheres in a
vaccine trial. IgG antibodies against the 53/25 kDa cathepsin L-like protein fraction were elicited in
the vaccinated animals but did not confer total protection (i.e., total absence of viable parasites in
the pig carcass after a necropsy) although we found a significant trend to reduced number of cysts
in the vaccinated group.
The 53/25 kDa cathepsin-L-like protein fraction purified from the cysticercus fluid performed well in
immunodiagnostics based on Western immunoblots or standard ELISA. These tests, however,
depend on the availability of infected pigs as a source of antigens. In addition, a Western blot is a
lengthy and complicated test not able to be performed in rural settings. A simple, cheap and fast
diagnostic assay that is suitable for use in field conditions is highly desired.
We tested the 53/25-kDa antigen in a dot-ELISA that could be employed for mass screenings
under field conditions. The assay was standardised and tested not only against sera from a large
group of NCC cases but also against sera from patients with other common parasitic infections, so
that sensitivity and specificity could be assessed.
For NCC, the assay gave better sensitivity in the detection of individuals with extraparenchymal
cysts (94.4%–100%) or multiple parenchymal cysts (74.6%–80.0%) than in the detection of
individuals with single parenchymal cysts (29.4%–45.1%). The assay also showed a high
specificity for NCC (99.0%–100%), with a very low level of cross-reactivity with other parasitic
infections. The dot-ELISA developed in this study is a highly specific, simple, cheap and rapid test
for NCC that could be used under field conditions, even in the low-resource settings that are
common in developing countries.
We analysed the composition and stability of the 53/25 kDa cathepsin L-like purified fraction using
mass spectrometry. Treatment with type-O and type-N glycosidases and lectin blots were used to
analyse the carbohydrate content. A total of 165 serum samples from patients with
neurocysticercosis were evaluated to ascertain the sensitivity of the subunits for diagnosis, and
103 serum samples from negative controls were evaluated to estimate the specificity. Ninety-nine
sera from patients with other helminth infections were included to determine cross-reactions.
The 53/25 kDa fraction was shown to be a protein complex comprising four subunits with type-N
glycosylation, and to include several lectin-type carbohydrates. The proteomic analysis revealed
that the four subunits comigrated with and showed motifs similar to the 8 kDa lectin-lentil

glycoproteins used in the EITB from the USA Centres of Disease Control (CDC). The sensitivity of
each of the subunits individually as well as of the combinations was lower than that of the native
complex.
The highly antigenic 53 kDa band, associated with cathepsin L-like activity, is a protein complex
that comprises four subunits with type-N glycosylation that show motifs of the 8 kDa family
antigens. The lack of evidence of a known proteolytic domain in the proteomic analysis, suggests
the possibility of the existence of a novel proteolitic domain associated to the 53/25 kDa fraction.
None of the single subunits attained the high sensitivity of the native complex, suggesting the
presence of quaternary structural epitopes.
Although the mechanism of infection is not completely understood, it is likely driven by proteolytic
activity that degrades the intestinal wall to facilitate oncosphere penetration and further infection.
We analysed the publicly available T. solium EST/DNA library and identified two contigs comprising
a full-length cDNA fragment very similar to Echinococcus granulosus Ag5 protein. The T. solium
cDNA sequence included a proteolytic trypsin-like-domain in the C-terminal region, and a
thrombospondin type-1 adherence-domain in the N-terminal region. Both the trypsin-like and
adherence domains were expressed independently as recombinant proteins in bacterial systems.
TsAg5 showed marginal trypsin- like activity and high sequence similarity to Ag5. The purified
antigens were tested in a Western immunoblot assay to diagnose human neurocysticercosis. The
sensitivity of the trypsin-like-domain was 96.36% in patients infected with extraparenchymal cysts,
75.44% in patients infected with multiple cysts, and 39.62% in patients with a single cyst.
Specificity was 76.70%. The thrombospondin type-1 adherence-domain was not specific for
neurocysticercosis.
Objectives:
The goal of this project is to further analyse the 53/25 kDa protein fraction with cathepsin L-like
activity purified from T. solium cysticercus fluid. This antigen will be used to standardise an
Enzyme Linked Immunosorbent Assay (ELISA), for immunodiagnostics of human cysticercosis.
(i) To clone, express and purify an active recombinant cathepsin L included in the 53/25 kDa
protein fraction, from cysticercus stage mRNA. Recombinant proteins will be expressed in
prokaryote cells.
(ii) To standardise an ELISA assay for human cysticercosis based on the recombinant antigen, and
to estimate the sensitivity, specificity and positive/negative predictive values of this test.
Results Obtained:
The 53/25 kDa cathepsin L-like fraction was shown to be a protein complex comprised of four
subunits with cysteine proteinase activity with type-N glycosylation, and several lectin-type
carbohydrates. The proteomic analysis revealed that the four showed motifs similar to the 8 kDa
lectin-lentil glycoproteins used in the CDC EITB. This fraction was detected mainly in the internal
face of the cyst and did not induce proliferation of human PBMC. We confirmed that the 53/25 kDa
cathepsin L-like fraction was also present in the T. solium oncosphere and was secreted by the
cysticercus. Although this fraction elicited IgG antibodies in vaccinated pigs, it did not confer total
protection. However, a significant trend to reduced number of cysts in the vaccinated group was
observed. The 53/25 kDa cathepsin-L-like protein fraction purified from the cysticercus fluid
performed well in immunodiagnostics based on a dot-ELISA. Other proteolytic antigens, including a
cathepsin L-like and a trypsin-like protease have been cloned and expressed recombinantly, and
are being tested in immunoassays.
We confirmed that the 53/25 kDa antigen is actually a protein complex comprising four subunits
that are likely to be the 8kDa LLGP family proteins of the CDC EITB assay.
Publications:
Piña, R., Gutiérrez, A.H., Gilman, R.H., Rueda, D., Sifuentes, C., Flores, M., Sheen, P.,
Rodriguez, S., García, H.H., Zimic, M. A dot-ELISA using a partially purified cathepsin-L-like
protein fraction from Taenia solium cysticerci, for the diagnosis of human neurocysticercosis. 2011.
Ann. Trop. Med. Parasitol. 105(4), 311-318
Rueda, A., Sifuentes, C., Gilman, R.H., Gutiérrez, A.H., Piña, R., Chile, N., Carrasco, S.,
Larson, S., Mayta, H., Verástegui, M., Rodriguez, S., Gutiérrez-Correa, M., García, H.H.,
Sheen, P., Zimic, M. TsAg5, a Taenia solium cysticercus protein with a marginal trypsin-like
activity in the diagnosis of human neurocysticercosis. 2011. Mol. Biochem. Parasitol. 180(2), 115-
119
Zimic, M., Pajuelo, M., Gilman, R.H., Gutiérrez, A.H., Rueda, L.D., Flores, M., Chile, N.,
Verástegui, M., Gonzalez, A., García, H.H., Sheen, P.; for the Cysticercosis Working Group in
Peru. The highly antigenic 53/25kDa Taenia solium protein fraction with cathepsin-L like activity is
present in the oncosphere/cysticercus and induces non-protective IgG antibodies in pigs. 2012.
Vet. Immunol. Immunopathol. 145(1-2), 171-178

POLAND
Title: TFE as the 3D-fold inducer. Application for domain 4 of E. coli RNAP σ70
Principal Investigator: Jarosław Poznański, Department of Biophysics, Institute of Biochemistry
and Biophysics, Polish Academy of Sciences, Pawińskiego 5a, 02-106 Warszawa, Poland. Tel: +48-
22-5925783, Fax: +48-22-5922190, E-mail: jarek@ibb.waw.pl
ICGEB Reference No.: CRP/POL08-03
Abstract: Domain 4 of E. coli (σ70
4) RNA polymerase holoenzyme (RNAP) is responsible for the
recognition of the -35 element of cognate promoters during the transcription initiation. Its
structure in a free form has not been yet determined, and the only accessible data come from the
homologous domains of thermophilic bacteria, and for σ70
4 from the two complexes with its natural
cofactors. We have found that our construct of σ70
4 (rECσ70
4) is almost insoluble in the neutral
aqueous solution, but in acidic conditions (pH ~ 2.8) it displays propensities towards a low-
populated structural arrangement that generally agrees with the known structures of σ70
4
homologues (Biochemistry 2003, 42, 13438). We have also proved that 2,2,2-trifluoroethanol
(TFE), which is a well known protein-folding inducer preferentially stabilising α-helical protein
structures, significantly increases the protein solubility and induces not only the helical secondary
structure, but also the canonical HLHTH fold, which we have successfully determined by the
combination of NMR data and molecular modelling. Screening for solution composition combined
with the detailed analysis of the collected 15
N relaxation data enabled us to propose the system
suitable for the study of the dynamics of the DNA-recognising HLHTH motif in a solution. However,
the limited solubility of the engineered protein sample at neutral pH enables the acquisition of basic
2D heteronuclear NMR spectra, at the moment it precludes NMR determination of the protein
structure or detailed relaxation study. In the progress, we have additionally demonstrated how
heteronuclear NMR spectroscopy supplemented by molecular dynamics methods can be applied for
the structural studies of the separated protein domains, which lost their structure when cut from
the whole protein context.
Objectives:
The main objectives of this project were:
(i) Determination of the stability of the TFE-induced fold using CD-monitored unfolding in various
solvent systems;
(ii) Estimation of the local mobility of the protein NMR with the aid of 15
N relaxation estimated for
protein in rationally-selected solvents;
(iii) Determination of at least low/resolution structure of the protein;
(iv) To propose protein sequence and solvent composition that enables the determination of
protein solution structure;
(v) Comparison of the determined solution structure of E. coli σ70
4 with those of the homologous
subdomains of thermophilic bacteria of a known structure.
Results Obtained:
An Intensive Circular Dichroism (CD) study demonstrated that the aqueous solution of TFE
significantly increases both the contents of the helical structure of rECσ70
4 and its thermal stability,
but also decreases the free energy of protein unfolding, thus, acting simultaneously as a structure
inducer and denaturant. The wide screening pointed at some promising compositions of solution
that may be applicable for structural studies. Thus, rECσ70
4 in the presence of 1.8M MgSO4 in
10mM sodium cacodylate buffer at pH~7 remains a good starting point for solvent optimisation
directed at protein crystallisation. 100mM NDSB-195 at pH ~7 was found to be a good folding
agent, and the free energy difference between the folded and unfolded states was estimated to
equal 1(0.3) kcal/mol. This solvent system could be used in an NMR application, although, besides
the acquisition of the basic heteronuclear HSQC spectra, the limited protein solubility (50 µM
achieved for the best system) precluded at the moment the application of any NMR structural
and/or relaxation studies.
We have demonstrated that TFE does not act as a non-specific helical structure inducer or
denaturant, but also that for the aqueous solution of rECσ70
4 the stepwise addition of TFE causes, in
its 10% content, an almost cooperative reorganisation of secondary structure of the limited set of
residues. Interestingly, their location within the protein sequence coincides with the location of
non-helical elements in the structures of homologous proteins. Complete NMR studies were carried
out for the aqueous solution of protein at pH ~4.4 in the presence of either 10 or 30% of TFE. In
lower TFE content, the TFE-induced helical structures were found to be loosely interacting, and on
the basis of 15
N relaxation and CSI data (chemical shift index), only conformational preferences
and an internal motion of the individual residues in protein could have been modeled (Proteins
2010, 78, 754). In high-TFE solution, the NMR studies concerned both the structure and the
internal dynamics of the protein. Using the limited set of distance restrains deduced from 3D
NOESY spectra, extended by the additional set of structural constraints obtained from CSI data
with the use of TALOS+ software, we succeeded in the determination of the low-resolution solution

structure of the DNA-recognising HLHTH motif (manuscript in preparation). The experimental data
enabled the estimation of protein backbone arrangement with the accuracy better than 2Å.
However, it should be stressed that, according to 15
N relaxation data, the proteins experience
conformational exchange, on the µs to ms timescale, between the proposed folded state and an
ensemble of unfolded forms, and only the structure of the folded component could be determined.
The backbone architecture of rECσ70
4 that is built of the three helices is similar to that of the
HLHTH motif in the closest homologue of Thermotoga maritima (TMσA
4). RMSD difference between
these two subunits, calculated for the CA backbone atoms, is high (11Å), although for the
framework of H1 and H3 helices it is reduced to 1.5 Å. This confirms the high functional homology
of these two regions. Moreover, the H1 and H3 helices in both proteins have the same
arrangement. However, the helix H1 of rECσ70
4 is slightly longer than that of TM σA
4. The H2
region, which is involved in DNA sequence recognition, displays differences in the relative
orientation in both proteins. The loop separating H1 and H2, as well as the short turn between H2
and H3, are longer in rECσ70
4, and their arrangement is visibly different from that observed in
TMσA
4. Only H3 region seems to be of a similar length. At the moment, on the basis of the
performed comparative analysis of HLHTH motifs, it should be pointed out that secondary structure
elements are not of an identical length, while the general organisation and secondary structure
distribution along the sequence remain the same, however, the promoter sequence preferences
may be different due to divergence in spatial location of the H2 helix.
The detailed analysis of the structural data followed by relaxation analysis showed that the large N-
terminal part of rECσ70
4 displays substantially increased conformational flexibility. We have tested
several oligopeptides corresponding to the rationally chosen C-terminal parts of ECσ70
4. The most
promising results were obtained for a 64-residue fragment Gly550-Asp613 of the whole E. coli
sigma factor, although the solubility of the protein sample in non-TFE systems was, even at
moderately acidic pH, limited to 50 µM. However, C-terminal His6-tag improves protein solubility.
Despite the fact that the actual solubility is very limited, the application of a combination of salts
and some low-mass protein stabilisers may help in preparing a sample for NMR structural studies.
All the rationally chosen protein fragments displayed extremely low solubility at neutral pH. The
application of protein stabilisers resulted in a very limited improvement of protein solubility. The
complex formed with DNA precipitates.
Publications:
Wąsik, R., Wińska, P., Poznański, J., Shugar, D. Isomeric mono-, di-, and tri-bromobenzo-1H-
triazoles as inhibitors of human protein kinase CK2α. 2012. PLoS One 7(11), e48898

RUSSIA
Title: IsoAsp7-Aβ as potential trigger of Alzheimer’s disease
Principal Investigator: Alexander A. Makarov, Laboratory of Conformational Stability of Proteins
and Physical Methods of Analysis, Engelhardt Institute of Molecular Biology RAS, Vavilov str. 32,
Moscow 119991, GSP-1, Russia. Tel: +7-499-1354095, Fax: +7-499-1351405, Email:
aamakarov@eimb.ru
ICGEB Reference No.: CRP/RUS08-02
Abstract: Alzheimer's disease (AD) is characterised by the formation of amyloid plaques as a
result of the extracellular accumulation of the amyloid-β peptide (Aβ) in the brain of AD patients.
The Aβ aggregates contain high concentrations of zinc ions bound to Aβ, and most of Aβ aspartate-
7 amino-acid residues undergo a specific chemical modification resulting in the formation of
isoaspartate-7. The project has addressed the role of the isoAsp7-containting Aβ (isoAβ) and its
interactions with metal ions in AD pathogenesis. We have developed and tested an analytical MS-
based method for measuring the relative abundance of isoAβ in blood samples. In vitro
experiments have allowed us to establish the following Aβ structural determinants: (i) the minimal
zinc-binding site, (ii) the primary zinc-binding site, and (iii) the site mediate zinc-induced
dimerisation. We also have shown that isoAβ has significant differences from intact Aβ in terms of
interactions with copper ions. Additionally, 3D structures of the isoAβ and rat Aβ zinc-binding
domains are solved. Finally, results of in vivo experiments have shown that intravenously injected
isoAβ triggers cerebral amyloidosis in transgenic mice.
Objectives:
(i) To develop an appropriate analytical method for measuring the relative abundance of isoAβ in
blood samples derived from healthy individuals and AD patients in order to determine whether this
parameter could be used as a biomarker for diagnostic purposes.
(ii) To determine the 3D structures of the human isoAβ and rat Aβ zinc-binding domains and to
investigate the molecular mechanism of interactions between zinc/copper ions and various natural
variants of the Aβ metal-binding domain 1-16.
(iii) To investigate whether isoAβ is the active pathogenic agent of amyloid plaque extracts, which
induce cerebral amyloidogenesis in transgenic mice, so as to determine if isoAβ could be used as a
target molecule for AD therapy.
Results Obtained:
(i) To evaluate levels of isoAβ in human blood samples, a quantitative method for determining of
the relative abundance of the isomerised and the native forms of Aβ in their binary mixtures was
developed. This method is based on the differences in the MS fragmentation spectra of the two
isoforms of Aβ. IsoAβ demonstrates fragments c6+57 and z10-57 in the ECD spectra, and
fragment b6+H2O in the CID spectra, which are not present in the fragmentation spectra of the
normal peptide. Using synthetic domain 1-16 of Aβ (Aβ16) and its isoaspartate-7-containing
isoform (isoAβ16) ESI MS calibration curves of characteristic ECD and CID fragments of isoAβ were
obtained and then used for quantitative analysis of the relative abundance of the isomerised form
in the Aβ fraction extracted from plasma samples. It was determined that for the 4 patients with
Alzheimer’s disease in heavy stages the relative abundance of the isomerised fragment 1-16 of the
Aβ peptide is about 6% (using CID LTQ and CID FT methods), while in the samples from other
donors only the native Aβ fragments were detected.
(ii) The thermodynamics of the binding of Zn2+
to Aβ fragments and Aβ(1-16) mutants was
characterised by isothermal titration calorimetry (ITC) in order to determine the minimal Zn2+
-
binding site of Aβ under physiological conditions. The amino acid region 6-14 of Aβ was determined
as the minimal Zn2+
-binding site wherein the ion is coordinated by His6, Glu11, His13, and His14.
We also used the quantum mechanics/molecular mechanics (QM/MM) method to model Zn2+
recognition by Aβ. Both ITC and QM/MM showed that three of four residues from the Aβ region 11-
14 (EVHH) contribute to zinc binding, and that this tetrapeptide readily forms dimers linked
through a zinc ion similarly to a model proposed for Aβ aggregation in a recent molecular dynamics
study by Miller et al. (2010). These data allow to us consider Aβ(11-14) tetrapeptide as a primary
Zn2+
-recognition site of Aβ and an important drug target candidate to prevent Zn2+
-induced
aggregation of Aβ.
(iii) The interactions of the immobilised Aβ metal-binding domain with soluble peptides
corresponding to the Aβ(1-16) and its fragments were characterised using surface plasmon
resonance biosensing in order to localise the interface of metal-induced Aβ dimerisation. We found
that zinc induces dimerisation of the Aβ domain 1-16, while copper does not influence this process.
The amino acid region 11-14 has been determined as the critical site for Zn2+
-induced dimerisation
of Aβ(1-16). The stoichiometry of zinc-bound complexes mimicking the Aβ(1-16) dimerisation
interface was studied by electrospray ionisation mass spectrometry (ESI-MS). ESI-MS experiments
have demonstrated that only one zinc ion participates in binding of the interacting Aβ molecules.
So, our results have shown that the primary zinc recognition site of Aβ (residues 11-14) controls

Zn2+
-induced dimerisation of Aβ(1-16). Since this process may be the trigger for further formation
of pathogenic Aβ oligomers and aggregates, inhibition of this site to prevent its interactions with
zinc ions may be a potential strategy in prevention of Alzheimer’s disease onset and progression.
(iv) Effect of isomerisation of aspartate 7 in chelating of the copper ions by the metal-binding
domain of Aβ was examined by isothermal titration microcalorimetry. It was found that the N-
terminal amino group, which is the copper chelator in the native Aβ, does not participate in the
coordination of the copper ion by the Aβ fragment 1-16 containing isoaspartate instead of
aspartate 7. Thus, the isomerisation of aspartate 7 significantly changes the structure of the
complex of Aβ with copper and consequently might play a pathogenic role in Alzheimer’s disease.
(v) Solution structure of isoAβ(1-16) in the free-state was determined by NMR spectroscopy and
molecular modeling. The data show that in the absence of zinc ions the 3D structures of isoAβ(1-
16) and native Aβ(1-16) are quite similar, whereas zinc binding significantly changes zinc
coordination by these peptides.
(vi) In an attempt to reveal the mechanism of rats’ resistance to Alzheimer’s disease, the structure
of the metal-binding domain 1–16 of rat amyloid-β (rat Aβ(1–16)) in solution in the absence and
presence of zinc ions was determined. Zinc induced dimerisation of the domain was detected. The
zinc coordination site was found to involve residues His-6 and His-14 of both peptide chains. We
used experimental restraints obtained from analyses of NMR and isothermal titration calorimetry
data to perform structure calculations. The calculations employed an explicit water environment
and a simulated annealing molecular-dynamics protocol followed by quantum-
mechanical/molecular-mechanical optimisation. We found that the C-tails of the two polypeptide
chains of the rat Aβ(1–16) dimer are oriented in opposite directions to each other, which hinders
the assembly of rat Aβ dimers into oligomeric aggregates. Thus, the differences in the structure of
zinc-binding sites of human and rat Aβ(1–16), their ability to form regular cross-monomer bonds,
and the orientation of their hydrophobic C-tails could be responsible for the resistance of rats to
Alzheimer’s disease.
(vii) In order to test our hypothesis about potential pathogenicity of isoAβ, in vivo investigations of
the seeded aggregation of Aβ were carried out by using commercially available APP/PS1 mice of
strain B6C3-Tg(APPswe, PSEN1dE9)85Dbo/j received from the Jackson Laboratory (USA) (stock
number 004462). These double transgenic mice manifest characteristic cognitive features of AD-
like pathology and possess significant amounts of dense congophilic amyloid plaques starting from
6-7 month age. To test whether inoculation of synthetic peptide isoAβ might trigger Aβ misfolding
and deposition in the brain, we administered each month intravenous injections (100 µl each; 50
µg of isoAβ) to two cohorts of young (2-month-old) female APP/PS1 tg mice. After performing
injections (5 for one cohort and 9 for the other), the host mice were analysed. IsoAβ induced
robust cerebral β-amyloidosis in all intravenously inoculated mice compared with untreated
littermate controls, whereas intravenous inoculation with phosphate-buffered saline or synthetic
peptide Aβ was ineffective.
A method for structure simulation of small peptides in aqueous environment has been developed
using the Amber-03 force field and Gromacs program after modification of its code.
Publications:
Istrate, A.N., Tsvetkov, P.O., Mantsyzov, A.B., Kulikova, A.A., Kozin, S.A., Makarov, A.A.,
Polshakov, V.I. NMR solution structure of rat aβ(1-16): Toward understanding the mechanism of
rats' resistance to Alzheimer's disease. 2012. Biophys. J. 102(1), 136-143
Shelkovnikova, T.A., Kulikova, A.A., Tsvetkov, P.O., Peters, O., Bachurin, S.O., Buchman,
V.L., Ninkina, N.N. Proteinopathies - forms of neurodegenerative disorders with protein
aggregation-based pathology. 2012. Mol. Biol. 46(2), 172-183
Indeykina, M.I., Popov, I.A., Kozin, S.A., Kononikhin, A.S., Kharybin, O.N., Tsvetkov, P.O.,
Makarov, A.A., Nikolaev, E.N. Capabilities of MS for analytical quantitative determination of the
ratio of α- and βAsp7 isoforms of the amyloid-β peptide in binary mixtures. 2011. Anal. Chem.
83(8), 3205-3210
Kozin, S.A., Mezentsev, Y.V., Kulikova, A.A., Indeykina, M.I., Golovin, A.V., Ivanov, A.S.,
Tsvetkov, P.O., Makarov, A.A. Zinc-induced dimerisation of the amyloid-β metal-binding domain
1-16 is mediated by residues 11-14. 2011. Mol. Biosyst. 7(4), 1053-1055
Istrate, A.N., Mantsyzov, A.B., Kozin, S.A., Pol'shakov, V.I. Optimisation of the methods for
small peptide solution structure determination by NMR spectroscopy. 2010. Mol. Biol. 44(6), 1075-
1085
Tsvetkov, P.O., Kulikova, A.A., Golovin, A.V., Tkachev, Y.V., Archakov, A.I., Kozin, S.A.,
Makarov, A.A. Minimal Zn(2+
) binding site of amyloid-β. 2010. Biophys. J. 99(10), L84-86
Tsvetkov, F.O., Makarov, A.A., Archakov, A.I., Kozin, S.A. Effect of isomerisation of aspartate-
7 on the binding of copper (II) ion by the beta-amyloid peptide. 2009. Biofizika 54(2), 197-201

SERBIA
Title: Structural alterations of DNA induced by initiator protein
Principal Investigator: Jelena Kusic, Laboratory for Molecular Biology, Institute of Molecular
Genetics and Genetic Engineering, University of Belgrade, Vojvode Stepe 444a, P.O. Box 23, 11010
Belgrade, Serbia. Tel: +381-11-3976658, Fax: +381-11-3975808, E-mail: jkusic@imgge.bg.ac.rs
ICGEB Reference No.: CRP/YUG08-01
Abstract: To ensure faithful replication of the genome in a limited time period, replication of
chromosomal DNA in eukaryotes initiates at multiple DNA sites called replication origins. Initiation
of DNA replication is highly coordinated biological process involving numerous protein-DNA and
protein-protein interactions. Nevertheless, it is still poorly understood how particular locations in
the chromosomes are selected as origins of replication, especially in the metazoan where the mode
of interactions between initiator proteins and DNA are largely unclear.
In eukaryotes, DNA replication is initiated by the Origin Recognition Complex (ORC), which binds
DNA at replication origins and promotes the assembly of the pre-Replication Complex (pre-RC). In
all eukaryotes homologues of ORC have been identified and biochemical studies revealed their
requirement for the initiation of DNA replication, suggesting that the mechanism of initiation is
conserved.
Although there is striking conservation of ORC, sequence requirements for ORC binding are not
well understood. Even in budding yeast, in which well-defined sequences serve as origins, specific
sequences are not required for ORC binding in vitro. Low sequence specificity of metazoan ORC is
consistent with the lack of common sequence motifs within metazoan origins, but does not match
the selectivity of initiation observed in somatic cells. Origin transfer studies, in combination with
mutational analysis of native and ectopic initiation sites, demonstrate that even metazoan origins
function as replicators, and therefore must contain required genetic elements. The most common
property of metazoan replication origins is frequent occurrence of the clustered adenines and
thymines distributed in either alternating or homogenous fashion. Since different pairs of clustered
residues relate to each other as either mirror or glide reflection images, they could interact in
alternative manners and force interacting DNA to assume unorthodox structure. The capacity of
origin DNA to build specific structural features could thus define replicator function.
Structural alteration of origins could be induced or stimulated by initiators. All six ORC subunits
belong to the initiator subgroup of the AAA+ superfamily of proteins, which undergoes
conformational changes or induces changes in interacting partners after binding of ATP. As
suggested by the mechanism of prokaryotic origin remodelling and emerging analogy between
eukaryotes and prokaryotes in initiator structure and function, eukaryotic initiators are expected to
remodel DNA in nucleotide-dependent manner.
Previously, we have identified specific unorthodox structures in DNA fragments lboI and SphI-
EcoRV, isolated from the essential and functionally interchangeable regions of the human LMNB2
and the hamster DHFR replication origins and we have observed the ability of HsOrc4 to stimulate
formation of these structures. To better characterise HsOrc4`s action towards different DNA
substrates, we have used oligonucleotides that mimic origin regions involved in alternative
interactions and we have registered the action of the wild type HsOrc4 or its mutants, deficient in
ATP binding or hydrolysis. We have observed that HsOrc4 stimulated formation of previously
identified DNA structures such as TAT triplexes and created novel structures such as such as AA
duplexes or quadruplexes. For catalytic action of HsOrc towards oligoadenine substrates an intact
ATP binding site was strictly required.
In this project we have focused on the role of HsOrc4 and human ORC holocomplex in origin
restructuring. In addition to the above described assays, we used circular, origin containing
molecules as substrates for HsOrc4. In circular molecules DNA unwinding is promoted by negative
supercoiling, which is a far superior and more natural manner of strand separation than thermal
denaturation. Based on our preliminary analysis with different agents we have chosen the plasmid
pMCD as substrate for protein mediated origin restructuring. This plasmid makes a convenient
substrate because it contains the whole DHFR origin and one unorthodox element identified in the
SphI-EcoRV region.
In addition to assays on linear DNA fragments (different electrophoretic techniques, chemical
assays with agents sensitive or resistant to single stranded DNA, substitution of 7-Deaza-dATP for
dATP), we have established methods specific for circular, origin containing DNA molecules - mung
bean nuclease sensitivity assays of the chosen origin regions and topoisomerase footprints on
chloroquine gels. Mung Bean nuclease sensitivity assays enable identification of DNA sequences in
which protein mediated structural alteration occurred, whereas topoisomerase footprints register
topological changes of DNA induced by protein. Combined results points to the capacity of essential
ori elements to adopt altered DNA structures and to the ability of HsOrc4 to direct structural
alterations of DNA, as potentially important elements of origin activation.
Objectives:

The primary goal of our research project was better understanding of the sequence elements
involved in origin restructuring during initiation of DNA replication and the potential of initiator
proteins to direct this structural alterations. The DNA replication origins of metazoan genomes are
the sites of complex protein-DNA interactions. Previously, we have observed that one subunit of
the Origin Recognition Complex, Orc4, could mediate formation of unorthodox structures from
linear, single-stranded DNA fragments. Since (-) supercoiling is more natural manner of strand
separation we investigated how different DNA topology conformations and Orc4 binding affect
origin structure.
Results Obtained:
MB nuclease assay demonstrated that, in negatively supercoiled DNA and at neutral pH, AT-rich
regions of hamster DHFR origin of replication adopt unorthodox DNA structure. Following addition
of initiation protein HsOrc4, this structure partly disappears indicating either protein binding to
unorthodox DNA structure or structural DNA remodelling mediated by protein. Analysis of MB
sensitivity in entire plasmid revealed that in topologically constrained DNA structural alteration in
ori sequences could be transmitted across DNA after binding of initiator protein HsOrc4.
Topoisomerase relaxation assays has confirm our assumption that HsOrc4 mediates topology
changes in pMCD. DNA remodelling by protein depends on (-) supercoiled state of plasmid DNA
and are stabilised by topoisomerase I action. Taken together our results bring us a step closer to
understanding the mode of interactions between initiator proteins and ori sequences.
We assumed that HsOrc4 will induce topology changes in constrained DNA molecules, but direction
of this change was unknown. The topoisomers distribution toward lower mobility compared to
reactions without the protein indicated introduction of positive writhe by initiation protein HsOrc4.
Publications:
Divac, A., Tomic, B., Kusic, J. The role of ATP in human ORC4 protein function. 2010. J. Serb.
Chem. Soc. 75, 317-322
Kusic, J., Tomic, B., Divac, A., Kojic, S. Human initiation protein Orc4 prefers triple stranded
DNA. 2010. Mol. Biol. Rep. 37, 2317-2322

SLOVAKIA
Title: Misfolded tau protein and oxidative defense system in tauopathies
Principal Investigator: Peter Filipcik, Laboratory of Cellular and Molecular Neurobiology,
Institute of Neuroimmunology, Dubravska Cesta 9, Bratislava, Slovakia. Tel: +421-2-54788100,
Fax: +421-2-54774276, E-mail: peter.filipcik@savba.sk
ICGEB Reference No.: CRP/SVK08-01
Abstract: Oxidative stress has been implicated in the pathogenesis of many neurodegenerative
diseases including different forms of tauopathy and Alzheimer’s disease. We have investigated the
role of human tau protein in the process of neurodegeneration in context with oxidative stress.
Previously, we observed significant accumulation of ascorbyl free radicals in brains of transgenic
animals expressing pathological form of tau protein. We have, therefore, focused to more detailed
understanding of relationships between tau protein modifications and oxidative stress. Here we
discovered that Vitamin C was found to be selective in the free radicals scavenging activity in
neurons, suggesting that expression of truncated tau protein preferentially leads to increases in
aqueous phase oxidants and free radicals. Our transcriptional analysis of brain tissue using PCR
arrays revealed several disregulated genes coding for the proteins in redox signaling pathways. We
have also pointed to possible mechanisms that can lead to generation of pathological forms of tau
protein, which subsequently can induce the oxidative stress in neurons. We hypothesise that
environmental stress can lead to generation of pathological forms of tau protein, either hyper-
phosphorylated or truncated, which may precipitate neurodegeneration, induce oxidative stress
and inflammation in vulnerable populations of brain neurons.
Objectives:
(i) Analysis of the pathogenic events associated with tau protein modifications and oxidative stress,
the common risk factor of neuronal degeneration:
(i.a) at the level of recombinant proteins;
(i.b) “in vitro” model of cultured cells (modified neuroblastoma cells and rat primary neurons) and;
(i.c) “in vivo” system using transgenic animals expressing pathologically modified tau protein.
(ii) Determination of temporal order and sequence of pathogenic events at early stage of
pathogenesis in human tauopathies and Alzheimer’s disease.
Results Obtained:
Within the project we have achieved the following results:
(i) Oxidative stress strongly increases oligomerisation of recombinant truncated tau proteins, which
may accelerate accumulation of aberrant tau in neurons when exposed to oxidative conditions;
(ii) Expression of truncated tau protein leads to formation of ROS in the aqueous phase and
vitamin C is selective scavenger of ROS generated in truncated tau-protein expressing neurons;
(iii) The pathogenic modifications of tau protein may precede oxidative damage in tauopathies and
Alzheimer’s disease.
(i) Our data suggest that hypoxic stress alters 3R-4R ratio in cerebro-cortical neurons, which may
lead to precipitation of neuropathogenic changes typical for specific tauopathies.
(ii) Immobilisation stress leads to induction of pathologically phosphorylated forms of tau protein in
different brain areas of acutely stressed experimental animals, which may subsequently lead to
redox disbalance in neuronal tissue.
Publications:
Cente, M., Filipcik, P., Mandakova, S., Zilka, N., Krajciova, G., Novak, M. Expression of a
truncated human tau protein induces aqueous-phase free radicals in a rat model of tauopathy -
implications for targeted antioxidative therapy. 2009. J. Alzheimers Dis. 17, 913-920
Filipcik, P., Novak, P., Mravec, B., Ondicova, K., Krajciova, G., Novak, M., Kvetnansky, R.
Tau protein phosphorylation in diverse brain areas of normal and CRH deficient mice: Up-regulation
by stress. 2012. Cell. Mol. Neurobiol. (in press)

SLOVENIA
Title: Molecular mechanism of the recognition of double stranded viral RNA mediated by TLR3
Principal Investigator: Mateja Manček-Keber, Department of Biotechnology, National Institute of
Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia. Tel: +386-1-4760393, Fax: +386-1-4760300,
E-mail: mateja.mancek@ki.si
ICGEB Reference No.: CRP/SLO08-01
Abstract: The innate immune response is an essential component of our defense against
pathogenic microorganisms. TLRs recognise molecules characteristic for pathogenic
microorganisms (PAMPS). Endosomal TLRs (TLR3, TLR7, TLR8, and TLR9) recognise different
classes of bacterial, viral and endogenous nucleic acids. TLR3 is activated by ds RNA, which is
characteristic for viruses.
In order to design potential inhibitory molecules of endosomal TLRs, a resolved mechanism of
ligand binding is of high importance. Although a crystal structure of TLR3 has been resolved this
dimer cannot explain why molecules smaller then 46bp such as siRNA can also activate TLR3. That
is why we proposed a second binding site on the N-terminal side of TLR3 ectodomain and a
different mechanism of TLR3 positioning after ligand binding. Moreover, we showed additional
important feature corresponding only to TLR3 that is cell surface expression when chaperon
UNC93B1 is coexpressed. This surface expressed TLR3 might enable cells to response to released
RNA and thus amplifying antiviral response.
Small interfering RNA molecules (siRNA) are short dsRNA molecules, which are widely used for
silencing and also in a variety of therapeutic applications. Their off-target activity can potentially
lead to unwanted toxicities. Our goal was also to study one of the side effects of siRNA-based
methods, namely activation of the TLR3, which limits the use of siRNA in therapeutic applications.
Objectives:
After the crystal structure of human TLR3 had been solved and site-directed mutagenesis of
individual amino acid residues within the ectodomain were done, the C-terminal binding site was
discovered. It includes H539 and N541. The resulting binding model showed dsRNA binding to the
C-terminal site on the nonglycosylated face of the ectodomain. According to this model dsRNA
shorter than 46 bp should not activate TLR3, but reports stated that the activation limit was 22 bp.
We proposed the existence of a second, N-terminal binding site. For this part we structured the
experimental plan where we selected, prepared and tested point mutations of TLR3. To evaluate
the mutations also binding studies of dsRNA to TLR3 mutants were performed. Additionally, we
proposed a binding model in which the ectodomains are shifted for the length of one dsRNA turn
closer together. In our model, TLR3 can be activated by ligands of a minimal length of 21 bp such
as siRNA.
Additional work as proposed was done in this project. TLR movement to endosomes is driven by
UNC93B1 chaperon. Surprisingly, we observed that UNC93B1 also mediated a translocation of
TLR3 to the cell surface, so we tried to determine, which segment of TLR3 governs this
translocation. Cell surface or endosomal localisation of TLR3 might have impacts on its signalling
properties, so we also tested how would surface expression influence poly (I:C) as well as siRNA
recognition and signaling. We also tried to explain what could be the physiological consequences of
cell surface expression in antiviral response.
Smaller part of the project was also extended to other endosomal TLR receptors (TLR7, TLR8 and
TLR9), which recognise nucleic acids other than dsRNA. The antimalarial drugs chloroquine,
hydroxychloroquine, and quinacrine have been used for a long time to treat immune-mediated
inflammatory disorders such as SLE, rheumatoid arthritis, and Sjogren’s syndrome. Chloroquine
and its analog quinacrine inhibit CpG DNA-driven cellular activation. With the aim to clarify the
mechanism of endosomal TLR inhibition, we therefore investigated direct binding of antimalarials to
nucleic acid TLR ligand.
Results Obtained & Unforeseen in the Original Project:
(i) A second binding site for double-stranded RNA in TLR3 and consequences for interferon
activation
We showed that mutations of H39 and H60 completely abrogated activation of TLR3. The presence
of two binding sites in the ectodomain probably determines the substrate specificity of TLR3. This
distance of 50 A° corresponds to two turns of a double-stranded RNA duplex, allowing
differentiation between nucleic acids in the A- or B-type conformation as we showed.
According to the results we proposed that dsRNA molecules of longer than 45 bp, which arise from
viral replication, form the most stable complex with TLR3, and consequently lead to robust
interferon production. Shorter RNA duplexes of between 21 bp and 30 bp can form less stable
complexes, resulting in weaker activation of the TLR3 signalling pathway.
(ii) UNC93B1 directs TLR3 towards the cell membrane
UNC93B1 is a chaperon, which was shown to deliver intracellular Toll-like receptors (TLRs) to
endolysosomes. We observed that overexpression of UNC93B1 elevated the abundance of TLR3,

but not TLR9 at the surface of the plasma membrane as well. In cells overexpressing UNC93B1 a
hyperglycosylated form of TLR3 was found, which was located exclusively on the cell surface and
was sensitive to deglycosylation. UNC93B1 also led to upregulation of intracellular TLR3 and TLR9,
but not surface TLR4 protein expression by affecting retention time of associated proteins in cells
as prevention of de novo synthesis with cycloheximide had no influence.
With knowledge that intracellular TLR receptors interact with UNC93B1 via their transmembrane
segments and cytosolic domains, we set the study of these segments for their importance for
localisation of TLR3 and TLR9. We prepared chimeric constructs, where the transmembrane
segments or cytosolic domains of TLR3 and TLR9 had been exchanged. The results showed that the
luminal domain of TLR3 governs UNC93B1-dependent localisation of TLR3 receptor to the plasma
membrane. Additionally, physiological relevance of UNC93B1 governed TLR3 translocation to the
cell surface was investigated. Stimulation of human primary cell line HUVEC with TLR3 ligand
poly(I:C) elevated transcription of UNC93B1 and TLR3 and increased the amount of TLR3 on the
cell surface. So, our results show that surface expressed TLR3 might enable cells to response to
released RNA thus amplifying antiviral response.
(iii) Differential activation of TLR3 by short and long dsRNA
Small interfering RNA molecules (siRNA) are short dsRNA molecules, which are widely used for
silencing and also in a variety of therapeutic applications. Their off-target activity can potentially
lead to unwanted toxicities. Our goal was to study one of the side effects of siRNA-based methods,
namely activation of the innate immune system, which limits the use of siRNA in therapeutic
applications. We showed that siRNA activated a TLR3-dependent inflammatory response on the
surface of primary endothelial cells. The NF-kB response was completely inhibited by antibodies,
reactive to the N-terminal binding site of the receptor's ectodomain. On the other hand, the
interferon type I response started predominantly by activating TLR3 in endosomal compartments,
as in the case of poly(I:C).
Additionally, we showed that poly(I:C) activated a stable IFN-β and IL-8 response in human
lymphatic endothelium cells, which was inhibited by bafilomycin A1 but not by anti-TLR3
antibodies. On the contrary, bafilomycin A1 could not inhibit activation with siRNA. According to
our results we believe that the outcome of TLR3 activation depends on receptor localisation and the
length of dsRNA. Activation of TRL3 on the cell surface by short dsRNA leads to a predominantly
proinflammatory response, while TLR3 activation in endosomal compartments by long dsRNA
mainly tends to activate type I IFN.
(iv) Mechanism of endosomal TLR inhibition by antimalarial drugs and imidazoquinolines
Endosomal Toll-like receptors (TLRs) play an important role in innate immune response as well as
in autoimmune processes. In the therapy of systemic lupus erythematusus, some antimalarial
drugs, such as chloroquine, hydroxychloroquine and quinacrine, have been used for a long time.
Their suppression of endosomal TLR activation has been attributed to the inhibition of endosomal
acidification, which is a prerequisite for the activation of these receptors. We discovered that
chloroquine inhibits only activation of endosomal TLRs by nucleic acids, while it augments
activation of TLR8 by a small synthetic compound R848. We detected direct binding of antimalarials
to nucleic acids by spectroscopic experiments and determined their cellular colocalisation. Further
analysis revealed that other nucleic acid binding compounds, such as propidium iodide also
inhibited activation of endosomal TLRs and colocalised with nucleic acids to endosomes. We found
that imidazoquinolines, which are TLR7/8 agonists, inhibit TLR9 and TLR3 even in the absence of
TLR7 or TLR8 and their mechanism of inhibition is similar to the antimalarials. In contrast to
bafilomycin, none of the nucleic acid binding compounds inhibited endosomal proteolysis,
confirming that inhibition of pH acidification in the presence of antimalarials and imidazoquinolines
is not the underlying cause of inhibition. We conclude that the direct binding of inhibitors to nucleic
acids changes their TLR binding epitope and may explain the efficiency of those compounds in the
treatment of autoimmune diseases.
Publications:
Pirher, N., Ivicak, K., Pohar, J., Bencina, M., Jerala, R. A second binding site for double-
stranded RNA in TLR3 and consequences for interferon activation. 2008. Nat. Struc. Mol. Biol. 7,
761-763
Kuznik, A., Bencina, M., Svajger, U., Jeras, M., Rozman, B., Jerala, R. Mechanism of
endosomal TLR inhibition by antimalarial drugs and imidazoquinolines. 2011. J. Immunol. 186,
4794-4804
Pirher, N., Pohar, J., Mancek-Keber, M., Jerala, R. Differential activation of TLR3 by short and
long dsRNA. (in preparation)
Pohar, J., Pirher, N., Bencina, M., Mancek-Keber, M., Jerala, R. Determinants of trafficking
TLR3 towards the cell membrane. (ready for submission)

SOUTH AFRICA
Title: Structural studies of enzymes in the context of anti-malarial drug development
Principal Investigator: Bryan Trevor Sewell, Electron Microscope Unit, University of Cape Town,
Private Bag, Rondebosch, Cape Town, South Africa. Tel: +27-21-650-2817, Fax: +27-21-689-
1528, E-mail: Trevor.Sewell@uct.ac.za
ICGEB Reference No.: CRP/ZAF08-04
Abstract: It is well established that patients with acute malarial infection suffer substantial
depletion of blood glutamine. This glutamine is utilised by the parasite that also has an endogenous
glutamine synthetase. There are, in addition, several unusual features of the metabolic pathways
concerned with nitrogen metabolism in Plasmodium falciparum. This applies particularly to
enzymes involved in maintaining the NAD+/NADH redox pair where three enzymes: NAD+ kinase,
NAD phosphorylase and nicotinamide phosphoribosyl transferase, which are part of the normal
pathway are absent from the parasite genome. The amidase domain of NAD+ synthetase (PfNADS)
was found to have an insertion and unusual sequence surrounding the active site cysteine which
could be evidence of specialisation required by the parasite. The expression of full-length NAD+
synthetase was found to be toxic to E. coli cells and thus far the protein has not been expressed
recombinantly. A method for preparing Plasmodium falciparum glutamine synthetase (PfGS)and
maintaining the enzyme stably in solution at concentrations high enough for crystallisation trials
was developed. The enzyme was visualised and shown to be homogeneous by both negative stain
and cryo-electron microscopy. A preliminary three-dimensional reconstruction has been made from
images of negatively stained protein.
Objectives:
(i) Clone, express, purify and crystallise NAD+ synthetase and glutamine synthetase from
Plasmodium falciparum;
(ii) Determine their structures by x-ray crystallography and determine whether they interact by
electron microscopy.
Results Obtained:
A method for preparing Plasmodium falciparum glutamine synthetase and maintaining the enzyme
stably in solution at concentrations high enough for crystallisation trials was developed. The
enzyme was visualised and shown to be homogeneous by both negative stain and cryo-electron
microscopy. A preliminary three-dimensional reconstruction has been made from images of
negatively stained protein.
The sequence of PfGS that was deposited in the PlasmoDB was incorrect. We experienced
considerable difficulty with the recombinant expression of both PfGS and PfNADS in E. coli. These
difficulties were overcome in the case of PfGS by using pCold expression plasmids and a good
purification procedure was developed for this enzyme. Further difficulties which have not yet been
overcome were encountered in crystallising the protein.
Networking:
The project enabled Prof. Trevor Sewell and Dr. Brandon Weber to visit the laboratory of Dr. Amit
Sharma of the ICGEB New Delhi Component for the crystallisation trials and to present the
outcome of the project at the MALSIG meeting. Dr. Jason van Rooyen, who was involved with this
project, is continuing the collaboration with Dr Sharma.
Publications:
Patel, S.I., Weber, B., Sewell, B.T. Recombinant expression and reconstruction of Plasmodium
falciparum glutamine synthetase. 2012. Protein Expr. Purif. (in preparation)

SRI LANKA
Title: Performance of recombinant proteins as diagnostic intermediates for Chikungunya, Dengue
and Leptospirosis
Principal Investigator: Menaka Hapugoda, Molecular Medicine Unit, Faculty of Medicine,
University of Kelaniya, Ragama, Sri Lanka. Tel: +94-112-960483, E-mail: menakaha@yahoo.com
ICGEB Reference No.: CRP/SRI08-02
Abstract: Chikungunya (CHIK), dengue and leptospirosis are important diseases with explosive
outbreaks occurring in Sri Lanka. Many clinicians as well as general practitioners find it difficult to
confirm diagnosis of these three infections only on clinical grounds, particularly when they exist as
co-infections. Confirmation of disease outbreaks is important for clinicians for proper management
of patients. Laboratory diagnosis of these three diseases in Sri Lanka is hindered by the high cost
of commercial diagnostic kits and inaccessibility of reagents. Because the burden of CHIK, dengue
and leptospirosis is greatest in Sri Lanka, there is a need to develop tests which can be
discriminated these three diseases and produced at low cost and also easily standardised for use in
field settings. Further, differential rapid, optimised diagnostic assay with high sensitivity and
specificity of each disease is essential for clinical management and epidemiological studies.
Laboratory diagnosis depends on Ensyme-Linked Immunosorbent Assay (ELISA) based on whole
viral/bacterial antigens which cause biohasard risk, high production cost and cross reactivity with
other organisms of the same genus/family. A diagnostic intermediate using a single recombinant
protein antigen to detect both IgM and IgG antibodies of each disease is important to overcome
problems associated with whole viral/bacterial antigen/lysate is important.
Objectives:
The overall objective of this project was to assist confirmation of CHIK, dengue and leptospirosis
outbreaks through developing rapid laboratory diagnostic assays.
Methodology:
Novel recombinant protein antigens for all three diseases were prepared. Here, synthetic genes of
CHIK were designed and expressed in both bacteria and yeast vector systems. Synthetic genes for
dengue and leptospirosis were designed and expressed in bacteria vector systems. Resulted
proteins for each disease were purified using a single affinity chromatographic step. Then, potential
use of each protein as a diagnostic tool for the detection of both IgM and IgG antibodies produced
against the particular disease organism was evaluated using a large panel of well characterised
serum samples. Indirect ELISAs using each novel recombinant protein as a capture antigen for
detection of both anti-IgM and IgG antibodies of each disease were developed using reference
serum samples obtained from a reference laboratory. Specificity, and sensitivity of IgM and IgG
ELISAs developed for each disease were analysed with currently available diagnostic assays
including the Gold standard assay using a large panels of well characterised serum samples.
Results Obtained:
The resulted proteins which show the highest sensitivity and specificity with currently available
diagnostic assay were selected. In order to synergise activities, the work was carried out jointly by
the Molecular Medicine Unit, Faculty of Medicine, University of Kelaniya, Sri Lanka (i.e., main
counterpart Institution) and by the Mammalian Biology Recombinant Gene Products (RGP) Group,
International Centre for Genetic Engineering and Biotechnology (ICGEB), New Delhi Component.
These two Institutions worked with each other through exchanging knowledge and resources (both
human and technical). In particular, the main counterpart Institution worked together with
clinicians, general practitioners and epidemiologist in notifying and locating cases at the regional
level when they collected clinical samples from suspected patients of each disease. The Sri Lankan
Institute collaborated with national and international reference centres to characterise three panels
of serum samples by the Gold standard assay for each disease. At national level networking was
established with: (i) clinicians from the Ministry of Health; (ii) Medical Research Institute; (iii)
National Science Foundation (RG/2009/BT/01). At international level collaboration was established
with: (i) Department of Virology (WHO Reference Centre for viral reference and research), Institute
of Tropical Medicine, Nagasaki University, Japan; (ii) Department of Biomedical Research,
WHO/FAO/OIE and National Collaborating Centre for Reference and Research on Leptospirosis,
Royal Tropical Institute, Amsterdam.
During the operation period, suspected patients for each disease warded in hospital/s in outbreak
area/s in Sri Lanka were benefited by having results of laboratory diagnostic assays (molecular and
currently available serological assays) within 48 hours of collection of the sample.
Funds received by the main counterpart institution through the project were used to meet the cost
of equipment, consumables, training and travel. The main counterpart institution has set up
facilities for molecular diagnosis and preparation of recombinant proteins for each disease.
One under graduate and three post graduate students were directly benefited by the project.
Project staff were benefited by obtaining local and international training through the project.
Results of the project were disseminated through a number of presentations made at national and

international conferences and were part of a post graduate thesis. In the near future,
dissemination of project results will be made through journal publications and other two post
graduate thesis.
Resulted proteins are available at two counterpart Institutions for field use and further studies. At
the end of the project target group beneficiaries will be the infected people and healthy persons in
outbreak area/s of these three diseases by confirming each disease using ELISAs developed using
these recombinant protein antigens. This type of study on development of diagnostic intermediates
has a significant effect for rapid confirmation of outbreaks and limit spread of such outbreaks from
one geographical area to another. Confirmation of disease outbreaks will avoid loss of working
hours and will also have socio-economical impact on individuals and government.
Publications:
In preparation.

UNITED ARAB EMIRATES
Title: Studying chromatin-modifying proteins in transcription regulation
Principal Investigator: Ahmed H. Hassan, Department of Biochemistry, Faculty of Medicine and
Health Sciences, United Arab Emirates University, Al-Ain, United Arab Emirates. Tel: ++971-3 713-
7478, Fax: +971-3-767-2033, E-mail: ahmedh@uaue.ac.ae
ICGEB Reference No.: CRP/ARE08-01
Abstract: It is clear that the targets for signalling pathways regulating gene expression include
both the DNA and protein components of chromatin. In the last decade, a number of transcriptional
regulators that function by modifying the structure of chromatin have been discovered. The
Saccharomyces cerevisiae SWI/SNF is the founding member of the chromatin-remodelling family.
Phylogenetic analysis shows that chromatin-remodelling proteins share several common features
including the presence of a distinct ATPase domain. Based on sequence homology to this domain of
the Swi2/Snf2 subunit of the SW/SNF, we have identified some new candidate remodelling proteins
including the Fun30 (Function unknown now 30) protein. With purified complexes in hand, a
growing panel of in vitro and in vivo assays, a new wealth of knowledge regarding the mechanisms
of actions of many chromatin-remodelling complexes, and the ease of genetic manipulations in
yeast, we have carried out mechanistic biochemical and some genetic investigations of their
functions and their inter-relationships. Our overall objective was to better understand the
mechanisms by which the Fun30 protein is involved in chromatin remodelling and transcription
regulation. Toward this end, we have carried out the biochemical characterisation of the yeast
Fun30 protein and have performed a series of in vitro assays as well as deletion studies. We are
now working on understanding the in vivo roles of Fun30 in gene regulation.
Objectives:
The overall objectives of this proposal were to better understand the mechanisms of action of
Fun30, a newly identified chromatin remodeller in Saccharomyces cerevisiae, in regulating gene
expression.
Results Obtained:
We have been able to purify Fun30 from the yeast and have shown that it is present as a
homodimer in our purifications. We had also performed a variety of in vitro experiments to
elucidate the mechanism of action of this chromatin remodeller. We observed that Fun30 binds to
DNA and nucleosomes, and has a variety of ATP-dependent remodelling activities. Since we have
shown that Fun30 can be targeted to promoters via direct interactions with acidic activation
domains, a new line of research on identifying the subunits or domains within fun30 that can bind
the acidic activators has now opened up and we will be conducting studies to identify the
domain(s) within Fun30 that interact with transcription activation domains. In addition to these
experiments, we have extended our dimer exchange studies and have shown that Fun30 exhibits
specificity towards H2AZ. We have also compared the ability of Fun30 to transfer H2B and H2AZ
dimers from HeLa donor chromatin to a tetrasome acceptor. We observed that the transfer of H2AZ
was slightly less efficient than H2B transfer. It would be interesting to investigate the effects of the
CUE motif or ATPase domain deletions on the activity of Fun30.
To explore the physiological functions of Fun30, we have also investigated the phenotypes of the
fun30 null mutant (fun30Δ strain). These phenotypic studies should help identify genes whose
promoters are targets of the Fun30 complex, which is an important part of our proposal. First we
investigated the involvement of Fun30 in the functions of metabolic pathways. The strain was
propagated on a variety of different carbon sources (rafinose, lactose and galactose). We found
that the fun30Δ strain grows on rich media with a growth rate comparable to wt strains. It has
been suggested that Fun30 is involved in chromosome stability and potentially DNA repair based
on the pattern of genomic instability. Thus, we investigated growth under conditions that could
highlight the role of Fun30 in the maintenance of DNA structure or DNA repair. To do this, the wild-
type (wt), snf2Δ, and fun30Δ strains in the same genetic background were spotted onto YPD plates
after their OD600 reached 1.5, and then different treatments to induce DNA damage. We first
investigated resistance/sensitivity to UV irradiation. UV causes cross-linking between adjacent
cytosine and thymine bases resulting in pyrimidine dimers which consequently inhibit polymerases
and arrest replication. While the wild-type and snf2Δ strains are very sensitive to even low doses of
UV irradiation, the fun30Δ yeast cells confer resistance and grew well even at the highest UV
exposure time interval. This supports a previous report of UV resistance in the absence of Fun30.
We have also found that the cells of the fun30 Δ strain were more resistant to increasing amounts
of radiation.
We have recently found that fun30 null mutant cells were slightly sensitive to caffeine. This finding
indicated that Fun30 may be important for the maintenance of cell viability during S phase arrest.
To explore any links between Fun30 and genes involved in progression through the cell cycle and
DNA replication, we have investigated the ability of fun30 deletion to progress through the cell
cycle. For this, we have used flow cytometry following propedium iodide staining to monitor the

DNA content of cultures synchronised using α-factor. We have observed a delay in the progression
through the G1/S phase of the cell cycle in the fun30Δ strain. We have now examined the
expression of several checkpoints marker proteins.
In addition, more recently, we have been concentrating our efforts on performing ChIP-Seq
experiments to identify the targets (localisation) of the Fun30 complex at the whole genome level,
which should help us understand the in vivo role(s) of this remodeller. These experiments will help
in identifying the involvement of the Fun30 protein in gene activation as well as repression.
We have faced some technical challenges over the years on this project that we have overcome
and the results obtained thus far are very promising and would lead to increased understanding of
the role of this novel chromatin remodeller.

URUGUAY
Title: Design of anti-tumour vaccine by glycoprotein engineering
Principal Investigator: Eduardo Osinaga, Departamento de Inmunobiología, Facultad de
Medicina, Universidad de la República, Av. General Flores 2125, Montevideo, Uruguay. Tel: +598-
29249562, Fax: +598-29249563, E-mail: eosinaga@fmed.edu.uy
ICGEB Reference No.: CRP/URU08-03
Abstract: The project aimed at developing a new strategy for the production of glycosylated hemi-
synthetic vaccines for anti-tumour immunotherapy. Our strategy was based on the directed
enzymatic glycosylation of mucin-like proteins from parasitic origin, in order to produce
glycoproteins containing the tumour-associated Tn antigen. Tn-based glycopeptides or
glycoporteins were prepared by in vitro glycosylation of mucin-like peptides from parasite origin,
either by using several recombinant ppGalNAc-Ts, or by solid-phase synthesis. The obtained data
during the performance of this project suggest that extense glycosylation of peptides modifies the
immunological properties, such as the type and intensity of T cell responses, as well as the antigen
capture by dendritic cells and their maturation. Also, we found that parasite-derived peptides could
induce a cytotoxicity capacity by in vivo-primed splenocytes that might be mediated by dendritic
cells. This work has provided new insights on the rationale of the design of Tn-based vaccine
candidates against cancer. Moreover, considering that the Tn antigen has also been identified in
numerous helminth parasites, a Tn vaccine produced from a parasite protein might be useful in the
induction of protective immune responses against these microorganisms.
Objectives:
(i) To synthesise different O-glycosylated isoforms of the C317 parasitary protein bearing the Tn
antigen using recombinant ppGalNAc-Ts.
(i.a) C317 recombinant protein production in Escherichia coli;
(i.b) In vitro GalNAc transfer to C317 proteins and physico-chemical characterisation of the
resulting glycoconjugates;
(i.c) Antigenicity of C317-Tn glycoproteins.
(ii) To evaluate the anti-tumour properties of the immune responses induced by the C317-Tn
glycoproteins in animal models (TA3/Ha transplantable tumour in mice and NMU-induced breast
cancer in rats).
Results Obtained:
(i) Production of parasite-derived glycopeptides carrying the Tn antigen for tumour immunotherapy
This project is based on an innovative approach for the production of glycosylated hemi-synthetic
vaccines, combining recent discovered mucin-like proteins from helminth parasites and an
enzymatic glycosylation strategy. Three different peptides from helminth-derived mucin-like
proteins were subjected to glycosylation: Egmuc-I and Egmuc-II (derived from Echinococcus
granulosus) and Fhmuc (derived from Fasciola hepatica). In order to obtain peptides carrying the
Tn antigen, we used the glycosyltransferases (ppGalNAc-T) that transfer GalNAc residues in vivo to
a Thr or Ser in a protein. ppGalNAc-T2, -T3, and –T6 enzymes have been produced in insect cells
infected with recombinant baculovirus. Fhmuc and Egmuc-II were efficiently glycosylated by
ppGalNAc-Ts. The presence of the Tn antigen on these glycopeptides was confirmed by Western
blotting. We next carried out a semi-preparative scale glycosylation of Fhmuc and EgmucII (500
µg). Additionally, we carried out the identification of lymphocyte T CD4 immunodominant
sequences by epitope mapping by immunising mice with the Egmuc peptides. The peptide designed
EM5 was the only one capable of stimulating T cell proliferation, indicating that this sequence
contains a T CD4 epitope.
(ii) Evaluation of the immunological properties of parasite-derived glycopeptides carrying the Tn
antigen
Tn-EgmucII and Tn-Fhmuc were highly recognised by the anti-Tn mAb 83D4. In contrast, the 15-
mer di-glycosylated Egmuc33-47-2Tn peptide was not recognised by 83D4. The immunogenicity of
peptides and glycopeptides was evaluated by their ability to produce antibodies in mice and to
induce specific T cell proliferation of in vivo-primed lymph node cells. We were not able to detect
by ELISA, IgG or IgM antibodies in sera from mice immunised with the glycosylated peptide
Egmuc33-47-2Tn against none of the evaluated (glyco)peptides. Additionally, Egmuc in vivo primed-
splenocytes cross-react with Egmuc33-47, while recognition is abolished when incorporating GalNAc
residues, suggesting that glycosylation alters T cell epitope recognition. In contrast, Fhmuc-Tn
induced much higher levels of cell proliferation and IFN γ than non glycosylated Fhmuc, suggesting
that, for this peptide backbone, the incorporation of GalNAc residues confers more cellular
immunogenicity to the peptide. These results indicate that glycosylation of T or B epitopes may
alter their immunogenicity, clearly depending on the peptide backbone, as well as the number of
GalNAc incorporated.
We analysed the capacity of bone marrow derived dendritic cells (BMDC) to internalise the
Egmuc33-47 peptide and glycopeptide. A strong capture and endocytosis by BMDC was observed for

both the non-glycosylated Egmuc and Egmuc33-47 peptides and the Egmuc33-47-2Tn glycopeptide,
indicating that the presence of GalNAc in the peptide did not modify internalisation by BMDC. The
internalisation was not modified in the presence of EDTA, suggesting that MGL (Macrophage
Gal/GalNAc lectin) does not intervene in the capture or internalisation of the Egmuc33-47-2Tn
glycopeptide, which needs several carbohydrate residues to allow cross-linking. In order to localise
the internalised peptides, the cells were analysed by confocal microscopy. In contrast with results
obtained with Egmuc33-47-2Tn, that was internalised at the same rate as Egmuc33-47, glycosylated
Fhmuc-Tn presented an enhanced uptake by DCs, as compared to Fhmuc, suggesting that, for
Fhmuc peptide backbone, the incorporation of GalNAc residues potentiates antigen internalisation.
We also analysed the presence of IL6, IL10 and IL12 in dendritic cell culture supernatants. The
studied peptides modify the expression of co-stimulatory molecules on dendritic cells and induce
the production of pro-inflammatory cytokines. However, this effect on dendritic cell maturation
would not be dependent on the Tn antigen on the Egmuc33-47 peptide backbone.
(iii) Anti-tumour immune response induced by parasite-derived glycopeptides carrying the Tn
antigen
Antibodies induced by any of Egmuc peptides and glycopeptides did not recognise antigens derived
from lysates of TA3/Ha or Panc02 tumour cell lines. Egmuc33-47 and Egmuc33-47-2Tn primed
splenocytes proliferated when stimulated with TA3/Ha or Panc02 protein lysates and produced
modest levels of INFγ and IL-5. These in vivo primed-splenocytes were also incubated with tumour
cells, in order to evaluate their cytotoxic ability. Splenocytes from mice immunised with Egmuc33-47
or Egmuc-2Tn peptides were capable of killing both tumour cells, while spleen cells from mice
immunised with Egmuc were not. The fact that both spleen cells from mice immunised with
Egmuc33-47 or Egmuc33-47-2Tn were capable of killing tumour cells, suggests that the recognition
is not Tn-dependent. One possibility could be that the Egmuc33-47 peptide and glycopeptide are able
to activate dendritic cells, inducing a release of certain activating cytokines (such as IL12 that
activates natural killer cells) or inducing the over-expression of certain co-stimulatory molecules
that favors the development of a cytotoxic response.
(iv) In vivo studies in experimental cancer models
Taking into account the in vitro capacity of Egmuc33-47 primed-splenocytes to mediate tumour cell
killing, we next intended to carry out some in vivo experiments in order to evaluate whether
Egmuc33-47 peptide or glycopeptide were able to protect mice from tumour growth. In a preliminary
evaluation, we were not able to protect from tumour growth Egmuc-immunisated mice. We are
currently working on different mice experimental cancer models.
We characterised the Fhmuc as a good peptide substrate in order to produce glycoproteins
containing the tumour-associated Tn antigen, using several recombinant ppGalNAc-Ts. We confirm
that the incorporation of several residues of GalNAc in long peptides of Fhmuc considerably
changes its immunological properties. We found that Fhmuc-Tn was much immunogenic than non-
glycosylated Fhmuc, being capable of inducing in vitro proliferation of splenocytes that produced
more than 10 times more of IFNγ than non-glycosylated Fhmuc. Additionally, we observed that
Fhmuc-Tn is more efficiently internalised by dendritic cells. We are currently working on this
aspect.
Publications:
We are in the process of preparing a manuscript that will be submitted soon. This paper will
contain the data regarding the anti-tumour immune response induced by Egmuc peptides. On the
other hand, we will very soon start the preparation of another paper, which still requires some
experimental data. This paper will include the glycosylation and study of the immunogenicity of
Fhmuc peptide and glycopeptides, stressing the fact that cellular immunogenicity is enhanced when
attaching GalNAc residues on this peptide backbone.

Projects_Completed_2012

More Related Content

What's hot (20)

Viewers also liked (17)

Similar to Projects_Completed_2012 (20)

Projects_Completed_2012