Prp8 intein in fungal pathogens: target for potential antifungal drugs.
ABSTRACT Inteins are self-splicing intervening sequences in proteins, and inteins of pathogenic organisms can be attractive drug targets. Here, we report an intein in important fungal pathogens including Aspergillus fumigatus, Aspergillus nidulans, Histoplasma capsulatum, and different serotypes of Cryptococcus neoformans. This intein is inside the extremely conserved and functionally essential Prp8 protein, and it varies in size from 170 aa in C. neoformans to 819 aa in A. fumigatus, which is caused by the presence or absence of an endonuclease domain and a putative tongs subdomain in the intein. Prp8 inteins of these organisms were demonstrated to do protein splicing in a recombinant protein in Escherichia coli. These findings revealed Prp8 inteins as attractive targets for potential antifungal drugs to be identified using existing selection and screening methods.
- SourceAvailable from: Hideo Iwaï[Show abstract] [Hide abstract]
ABSTRACT: Protein sequences are diversified on the DNA level by recombination and mutation and can be further increased on the RNA level by alternative RNA splicing, involving introns that have important roles in many biological processes. The protein version of introns (inteins), which catalyze protein splicing, were first reported in the 1990s. The biological roles of protein splicing still remain elusive because inteins neither provide any clear benefits nor have an essential role in their host organisms. We now report protein alternative splicing, in which new protein sequences can be produced by protein recombination by intermolecular domain swapping of inteins, as elucidated by NMR spectroscopy and crystal structures. We demonstrate that intein-mediated protein alternative splicing could be a new strategy to increase protein diversity (that is, functions) without any modification in genetic backgrounds. We also exploited it as a post-translational protein conformation-driven switch of protein functions (for example, as highly specific protein interference).Nature Chemical Biology 08/2013; · 12.95 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Protein trans-splicing by split inteins has many uses in protein production and research. Splicing proteins with synthetic peptides, which employs atypical split inteins, is particularly useful for site-specific protein modifications and labeling, because the synthetic peptide can be made to contain a variety of unnatural amino acids and chemical modifications. For this purpose, atypical split inteins need to be engineered to have a small N-intein or C-intein fragment that can be more easily included in a synthetic peptide that also contains a small extein to be trans-spliced onto target proteins. Here we have successfully engineered multiple atypical split inteins capable of protein trans-splicing, by modifying and testing more than a dozen natural inteins. These included both S1 split inteins having a very small (11-12 aa) N-intein fragment and S11 split inteins having a very small (6 aa) C-intein fragment. Four of the new S1 and S11 split inteins showed high efficiencies (85-100%) of protein trans-splicing both in E. coli cells and in vitro. Under in vitro conditions, they exhibited reaction rate constants ranging from ∼1.7×10(-4) s(-1) to ∼3.8×10(-4) s(-1), which are comparable to or higher than those of previously reported atypical split inteins. These findings should facilitate a more general use of trans-splicing between proteins and synthetic peptides, by expanding the availability of different atypical split inteins. They also have implications on understanding the structure-function relationship of atypical split inteins, particularly in terms of intein fragment complementation.PLoS ONE 01/2013; 8(4):e59516. · 3.73 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: The PRP8 intein is the most widespread intein among the Kingdom Fungi. This genetic element occurs within the prp8 gene, and is transcribed and translated simultaneously with the gene. After translation, the intein excises itself from the Prp8 protein by an autocatalytic splicing reaction, subsequently joining the N and C terminals of the host protein, which retains its functional conformation. Besides the splicing domain, some PRP8 inteins also have a homing endonuclease (HE) domain which, if functional, makes the intein a mobile element capable of becoming fixed in a population. This work aimed to study 1) The occurrence of this intein in Histoplasma capsulatum isolates (n = 99) belonging to different cryptic species collected in diverse geographical locations, and 2) The functionality of the endonuclease domains of H. capsulatum PRP8 inteins and their phylogenetic relationship among the cryptic species. Our results suggest that the PRP8 intein is fixed in H. capsulatum populations and that an admixture or a probable ancestral polymorphism of the PRP8 intein sequences is responsible for the apparent paraphyletic pattern of the LAmA clade which, in the intein phylogeny, also encompasses sequences from LAmB isolates. The PRP8 intein sequences clearly separate the different cryptic species, and may serve as an additional molecular typing tool, as previously proposed for other fungi genus, such as Cryptococcus and Paracoccidioides.Infection, genetics and evolution: journal of molecular epidemiology and evolutionary genetics in infectious diseases 05/2013; · 3.22 Impact Factor
Prp8 intein in fungal pathogens: target for potential antifungal drugs
Xiang-Qin Liu*, Jing Yang
Department of Biochemistry and Molecular Biology, Dalhousie University, Halifax, NS, Canada B3H 4H7
Received 19 May 2004; revised 6 July 2004; accepted 9 July 2004
Available online 23 July 2004
Edited by Horst Feldmann
proteins, and inteins of pathogenic organisms can be attractive
drug targets. Here, we report an intein in important fungal
pathogens including Aspergillus fumigatus, Aspergillus nidulans,
Histoplasma capsulatum, and different serotypes of Cryptococ-
cus neoformans. This intein is inside the extremely conserved and
functionally essential Prp8 protein, and it varies in size from 170
aa in C. neoformans to 819 aa in A. fumigatus, which is caused
by the presence or absence of an endonuclease domain and a
putative tongs subdomain in the intein. Prp8 inteins of these
organisms were demonstrated to do protein splicing in a
recombinant protein in Escherichia coli. These findings revealed
Prp8 inteins as attractive targets for potential antifungal drugs
to be identified using existing selection and screening methods.
2004 Federationof European
Published by Elsevier B.V. All rights reserved.
Inteins are self-splicing intervening sequences in
Keywords: Intein; Protein splicing; Prp8 protein;
Fungal pathogen; Drug target
Inteins are protein intervening sequences that can self-excise
through protein splicing, which also joins the flanking se-
quences (N- and C-exteins) with a peptide bond to produce the
mature host protein (spliced protein) [1–4]. Most inteins have a
homing endonuclease domain that initiates intein homing [5,6],
and split inteins exist in two fragments and do protein trans-
splicing [7,8]. No biological role benefiting the host organism
has been determined for inteins, but intein’s protein splicing
function is required for and could potentially regulate pro-
duction of the mature host protein.
Inteins have been found in certain mycobacterial pathogens
and suggested as attractive targets for potential antimyco-
bacterial drugs (reviewed in ). Because no intein has been
found in animals including human, drugs targeting inteins are
less likely to have toxic side effects. Systems for in vivo selec-
tion and in vitro screening have been developed to find such
drugs. For example, coding sequence of a mycobacterial RecA
intein was inserted in a thymidylate synthase gene from
bacteriophage T4, and the resulting fusion protein could
complement a Escherichia coli thyA mutant if the intein could
self-excise through protein splicing. This allows positive
growth selection against the protein splicing, because the thyA
complementation leads to growth inhibition in the presence of
thymine and trimethoprim . Other genetic methods have
involved the CcdB cytotoxic protein  and a quinolone-
sensitive GyrA protein . An in vitro screening system has
been developed for identifying inhibitors of intein inserted in a
green fluorescence protein .
An intein was also found in the fungal pathogen Crypto-
coccus neoformans and suggested as potential drug target ,
although its protein splicing activity was not demonstrated. Its
host protein, Prp8, is an extremely conserved large protein and
a critical component of the catalytic core of spliceosome .
The spliceosome is a large ribonucleoprotein complex and
catalyzes the removal of introns from pre-mRNAs. The Prp8
protein is essential for RNA splicing, and evidences showed
that it stabilizes tertiary RNA interactions, facilitates forma-
tion of the catalytic core, and acts as a protein cofactor to the
RNA enzyme. The host organism C. neoformans has various
strains classified into different serotypes (A, B, C, D, and AD)
and varieties known as grubii (serotype A), neoformans (sero-
type D), and gattii (serotypes B and C). The Prp8 intein would
be a more attractive drug target, if it is active in protein
splicing and present also in other important fungal pathogens
and in different serotypes of C. neoformans.
Here, we report the finding and characterization of Prp8
inteins from a wide range of important fungal pathogens,
including Aspergillus fumigatus, Aspergillus anidulans, Histo-
plasma capsulatum, and different serotypes of Cryptococcus
neoformans. We also demonstrate that Prp8 inteins from these
organisms were active in protein splicing when inserted in a
model host protein in E. coli. We discuss the usefulness of
Prp8 inteins as attractive targets for potential antifungal
drugs to be identified using existing selection and screening
2. Materials and methods
2.1. Gene cloning and sequence analysis
Fungal cells and/or their genomic DNAs were obtained from other
researchers (M. J. Bidochka, M. Momany, A. Sil, and J. Yu) and from
the University of Alberta Microfungus Collection & Herbarium
(UAMH). Coding sequences of Prp8 inteins and their flanking exteins
(partial) were amplified from total genomic DNA by doing polymerase
chain reaction (PCR) using degenerate and precise primers. Degenerate
oligonucleotide primers for Prp8 sequences of Aspergillus and Histo-
plasma included 50-ATGAAGAGCAAYCCNTTYTGGTGGAC-30
primers for Cryptococcus included 50-GCTCTCGGTGGTGTTGAG-
and G), H (A, C, and T), and N (A, C, G, and T). The amplified DNAs
*Corresponding author. Fax: +1-902-494-1355.
E-mail address: firstname.lastname@example.org (X.-Q. Liu).
0014-5793/$22.00 ? 2004 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
FEBS Letters 572 (2004) 46–50 FEBS 28651
were cloned in a pDrive plasmid vector (Qiagen) and subjected to au-
tomated DNA sequencing. GenBank searches and protein sequence
alignments were performed using the BLAST search program  and
the Clustal W program , respectively.
2.2. Protein splicing analysis in E. coli cells
To construct gene expression plasmids, Prp8 intein coding sequences
were inserted in the previously reported expression plasmid pMST 
between XhoI and AgeI sites, replacing the Ssp DnaB intein coding
sequence of pMST. Protein production in E. coli cells, gel electro-
phoresis and Western blot analysis were performed as before .
Briefly, cells containing the expression plasmid were grown in liquid
Lurie Broth (LB) medium at 37 ?C to late log phase (A600, 0.5). IPTG
was added to a final concentration of 0.8 mM to induce production of
the recombinant protein, and the induction was continued at 37 ?C for
3 h or at room temperature overnight. Cells were then harvested and
lysed in SDS- and DTT-containing gel loading buffer in a boiling water
bath before electrophoresis in SDS–polyacrylamide gel. Western blots
were carried out using anti-thioredoxin antibody (Invitrogen) and the
enhanced chemi-luminescence detection kit (ECL). Intensity of protein
band was estimated using a gel documentation system (Gel Doc 2000
with Quantity One software, Bio-Rad).
3.1. Prp8 intein in fungal pathogens
A suspected intein-containing fragment of Prp8 gene, which
corresponds to insertion site of the previously identified Prp8
intein of C. neoformans , was PCR-amplified from genomic
DNAs of a number of fungal species and strains listed in
Table 1. The resulting DNA fragments from some organisms
indicating the presence of intein-coding sequences. These or-
ganisms included A. fumigatus, A. nidulans (strains A28, 273,
C. neoformans (strains AmMs229 and YBC81). The DNA
fragments were cloned and their sequences determined, which
revealed Prp8 inteins of various sizes (Fig. 1).Some of the intein
sequences were also found in searches of databases and partial
genome sequences of corresponding organisms (Table 1). A
number of other Aspergillus species (flavus, niger, oryzae, para-
intein-less Prp8gene, indicatingabsence ofthePrp8intein (data
Prp8 inteins of the different pathogens have the same cor-
responding insertion site in Prp8 protein, but their sizes differ
greatly. In particular, the 819-aa Afu Prp8 intein is nearly 5
times as large as the 170-aa Cne Prp8 intein. These size dif-
ferences are mostly due to the presence or absence of a homing
endonuclease domain in the intein, based on sequence align-
ment and comparison of Prp8 inteins of different organisms
(Fig. 2). The structural domains of Prp8 inteins (splicing do-
main, endonuclease domain, and putative tongs subdomain)
were predicted through comparison with the Sce VMA intein
whose crystal structure is known [20,21]. Conserved sequence
blocks (A through G, N) were identified using consensus se-
quences of known inteins . A homing endonuclease domain
is present in Prp8 inteins of A. fumigatus, A. nidulans, and H.
capasilatus, but absent in Prp8 inteins of different C. neofor-
mans strains and varieties (Fig. 2A). The Afu Prp8 intein has a
222-aa sequence (Fig. 2B) that has no counterpart in Prp8
inteins of the other organisms, including the closely related A.
nidulans. This 222-aa sequence is located immediately after the
conserved sequence block B and corresponds in position to the
tongs subdomain in the crystal structure of Sce VMA intein
. However, its sequence shows no apparent similarity to
and is over three times longer than the 69-aa tongs subdomain
of Sce VMA intein.
The splicing domain (HINT domain) of Prp8 intein is highly
conserved among the different organisms (Fig. 2A), after ex-
cluding the homing endonuclease domain and the putative
tongs subdomain. Sequence identities of the HINT domain are
78% between the Aspergillus species, ?73% between Aspergil-
lus and Histoplasma, ?47% between Aspergillus and Crypto-
coccus, and over 85% among different Cryptococcus strains.
The homing endonuclease domain is 387–455 aa long and
Prp8 intein distribution in fungal organisms
Organism Intein size
Source of sequence information
flavus, niger, oryzae, parasiticus,
Serotype D, neoformans
Serotype A, grubii
This work: strain AmMs229, ATCC66031 Butler et al. : strain JEC21
This work: strain YBC81, ATCC76484 InBase (www.neb.com/inteins/): strain
PHLS 8104 Also GenBank AY422974
InBase (www.neb.com/inteins/): strain CBS132
InBase (www.neb.com/inteins/): C. bacillisporus Also GenBank AY422975
Serotype C/B gattii
Fig. 1. Schematic comparison of Prp8 inteins. Structural domains of
Prp8 inteins are illustrated, which include three parts of the protein
splicing or HINT domain (gray boxes), the endonuclease domain
(hatched boxes), the putative tongs subdomain (dotted box), and linker
sequence (open box).
X.-Q. Liu, J. Yang / FEBS Letters 572 (2004) 46–50
shows 32–41% sequence identity and 47–55% sequence simi-
larity among the different Aspergillus and Histoplasma species.
Presence of conserved sequence blocks, especially blocks C and
E, identified them as DOD (or LAGLIDADG) type endo-
nucleases. The Prp8 inteins are not similar to other known
inteins in host protein and in insertion site, and they showed
less than 20% sequence identity to other inteins.
3.2. Protein splicing activity of Prp8 intein in E. coli
To test the Prp8 inteins for protein splicing activity in E. coli,
each intein coding sequence was inserted in a fusion gene on a
plasmid. In the resulting fusion protein, the Prp8 intein (plus 5
aa of its native extein sequence on each side) was flanked by a
maltose binding protein at its N-terminus and a thioredoxin at
its C-terminus (Fig. 3). Similar fusion protein constructs had
Fig. 2. Protein sequence comparison of Prp8 inteins. (A) Sequence alignment of splicing (HINT) domain, with a 5-aa extein sequence on each side of
the intein shown in lower case letters. Positions of the endonuclease domain (ED) and the putative tongs subdomain (TSD) are indicated, with the
numbers of amino acids shown in parenthesis. Conserved intein sequence motifs (blocks A, B, F, and G) are underlined. (B) Sequence of the putative
tongs subdomain of Afu Prp8 intein. C. Sequence alignment of endonuclease domain. Conserved sequence motifs (blocks C, D, E, and H) are
underlined. The compared sequences are from A. fumigatus (Afu), A. nidulans (Ani), H. capsilatum (Hca), C. neoformans var. neoformans (Cne), C.
neoformans var. grubii (Cgr), and C. neoformans var. gattii (Cga). Symbols: -, represent gaps introduced to optimize the alignment; *, and. mark
positions of identical and similar amino acids, respectively.
X.-Q. Liu, J. Yang / FEBS Letters 572 (2004) 46–50
been used in previous studies of other inteins [18,19,22,23],
thus the protein splicing products could readily be identified in
SDS–PAGE and Western blotting. The precursor protein,
spliced protein, and excised intein were identified by their
predicted sizes. The precursor and spliced proteins were also
identified through antibody recognition of their thioredoxin
Efficiency of protein splicing was estimated by comparing
protein band intensities of the precursor and spliced proteins
on Western blot. As seen in Fig. 3, each of the Prp8 inteins
showed efficient protein splicing at room temperature, and the
precursor protein was converted completely into the spliced
protein. At 37 ?C, Prp8 inteins from A. fumigatus, A. nidulans,
and C. neoformans showed efficient protein splicing, and little
or no precursor protein remained. The Hca Prp8 intein from
H. capsulatum, however, showed significantly less protein
splicing at 37 ?C, because approximately 75% of the precursor
protein remained unspliced.
We have found Prp8 inteins in a number of fungal species
but not in many others (see Table 1), which indicates a wide
but sporadic distribution. Prp8 inteins of the different fungal
species are clearly homologous, based on their identical in-
sertion site in Prp8 protein and high degrees of sequence
conservation. The Prp8 intein sequence of A. fumigatus (also
A. nidulans) is more similar to that of H. capsulatum than
either is to that of C. neoformans, which is consistent with
the estimated evolutionary distances of these organisms. For
example, the filamentous fungus A. fumigatus appears dis-
tantly related to the basidiomycete fungus C. neoformans
and more closely related to the ascomycete fungus H. cap-
sulatum (also known as Ajellomyces capsulatus), based on
nucleotide differences in the ITS1-5.8S-ITS2 region of nu-
clear rRNA genes . The absence of Prp8 intein in many
other Aspergillus species, together with the finding of a
homing endonuclease domain in some Prp8 inteins, suggests
that this intein originated in the different fungal species
through lateral gene transfers, which were perhaps catalyzed
by the intein-contained homing endonuclease. Therefore, the
Cne Prp8 intein most likely had the homing endonuclease
domain and later lost it.
The Hca Prp8 intein is unusual, because it has an extra se-
quence predicted to be a putative tongs subdomain. Although
this prediction was based only on its same location as the tongs
subdomain in the crystal structure of Sce VMA intein, a se-
quence similarity is not necessary for this prediction to be
correct. The tongs subdomain of Sce VMA intein was believed
to participate in the binding of substrate DNA by the site-
specific homing endonuclease of that intein . If the putative
tongs subdomain of Hca Prp8 intein has a similar function, it
may not necessarily have a similar sequence, because it is ex-
pected to bind a very different substrate DNA at a very dif-
ferent intein insertion (homing) site. If our prediction is
correct, the putative tongs subdomain of Hca Prp8 intein is
three times longer than the tongs subdomain of Sce VMA
intein, which has implications on the endonuclease function
and intein evolution.
Fig. 3. Protein splicing activity of Prp8 inteins. (Top) Schematic illustration of the fusion-protein construct consisting of maltose binding protein
sequence (M), intein sequence (gray box), and thioredoxin sequence (T). (Middle) Predicted sizes of protein products from fusion-protein constructs
containing the specified inteins. The Ssp DnaB mini-intein was included as a known standard for identifying the spliced protein . (Bottom)
Observation of protein splicing. The specified fusion-proteins were produced in E. coli either at 37 ?C or at 25 ?C (room temperature) as indicated.
Total cellular proteins were resolved by SDS–PAGE and visualized by Coomassie Blue staining (left panel) or Western blotting (right panel) using
anti-thioredoxin antibody. Positions of precursor protein, spliced protein, and excised intein were indicated with letters P, S, and I, respectively.
Letter C marks a putative product resulting from cleavage at the C-terminus of intein.
X.-Q. Liu, J. Yang / FEBS Letters 572 (2004) 46–50
We have demonstrated for the first time that Prp8 inteins are
active for protein splicing, at least in E. coli. This makes it
likely that Prp8 inteins can actively self-excise in vivo, which
implies that these inteins are not inactive sequences tolerated in
mature Prp8 protein. The Hca Prp8 intein from H. capsulatum
showed significantly lower splicing activity at 37 ?C in E. coli,
which could be potentially interesting if this happens also in its
native Prp8 protein in vivo. A shift from 25 to 37 ?C is known
to stimulate the dimorphic transition of this fungus from the
multicellular mycelial form to the unicellular yeast form
[25,26]. The demonstration of protein splicing in a non-native
protein in E. coli suggests that Prp8 inteins may be used in
some of the existing selection or screening systems for identi-
fying intein inhibitors , which may produce or lead to po-
tential antifungal drugs. Because Prp8 protein has an essential
cellular function, inhibiting Prp8 inteins should effectively kill
The finding of Prp8 inteins in several important fungal
pathogens further increases the attractiveness of Prp8 inteins
as drug targets. A. fumigatus, H. capsulatum, and C. neofor-
mans are among the most significant opportunistic fungal
pathogens [26,27], and A. nidulans has been associated with
infections in humans and animals . These environmental
fungal pathogens exist widely in soil and/or on decaying or-
ganic matters, often become airborne in and out doors, and
when inhaled into the lung, can cause invasive or non-invasive
diseases. Because these opportunistic fungal pathogens pri-
marily infect immunocompromised individuals, controlling
them has become more urgent due to the pandemic of AIDS
and the widespread use of immunosuppressive therapy.
Acknowledgements: We thank M.J. Bidochka, M. Momany, L. Sigler,
A. Sil, and J. Yu for providing the fungal cells and DNAs used in this
study. This work was supported by a research grant from the Canadian
Institutes of Health Research.
 Perler, F.B. et al. (1994) Nucleic Acids Res. 22, 1125–1127.
 Paulus, H. (2000) Annu. Rev. Biochem. 69, 447–496.
 Liu, X.Q. (2000) Annu. Rev. Genet. 34, 61–76.
 Gogarten, J.P., Senejani, A.G., Zhaxybayeva, O., Olendzenski, L.
and Hilario, E. (2002) Annu. Rev. Microbiol. 56, 263–287.
 Cooper, A.A. and Stevens, T.H. (1995) Trends Biochem. Sci. 20,
 Perler, F.B. (2002) Nucleic Acids Res. 30, 383–384.
 Wu, H., Hu, Z. and Liu, X.Q. (1998) Proc. Natl. Acad. Sci. USA
 Caspi, J., Amitai, G., Belenkiy, O. and Pietrokovski, S. (2003)
Mol. Microbiol. 50, 1569–1577.
 Paulus, H. (2003) Front. Biosci. 8, S1157–S1165.
 Derbyshire, V., Wood, D.W., Wu, W., Dansereau, J.T., Dalgaard,
J.Z. and Belfort, M. (1997) Proc. Natl. Acad. Sci. USA 94, 11466–
 Lew, B.M. and Paulus, H. (2002) Gene 282, 169–177.
 Adam, E. and Perler, F.B. (2002) J. Mol. Microbiol. Biotechnol. 4,
 Gangopadhyay, J.P., Jiang, S.Q. and Paulus, H. (2003) Anal.
Chem. 75, 2456–2462.
 Butler, M.I., Goodwin, T.J. and Poulter, R.T. (2001) Yeast 18,
 Collins, C.A. and Guthrie, C. (2000) Nat. Struct. Biol. 7, 850–
 Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang,
Z., Miller, W. and Lipman, D.J. (1997) Nucleic Acids Res. 25,
 Higgins, D.G., Thompson, J.D. and Gibson, T.J. (1996) Methods
Enzymol. 266, 383–402.
 Wu, H., Xu, M.Q. and Liu, X.Q. (1998) Biochim. Biophys. Acta
 Liu, X.Q., Yang, J. and Meng, Q. (2003) J. Biol. Chem. 278,
 Duan, X., Gimble, F.S. and Quiocho, F.A. (1997) Cell 89, 555–
 Werner, E., Wende, W., Pingoud, A. and Heinemann, U. (2002)
Nucleic Acids Res. 30, 3962–3971.
 Liu, X.Q. and Yang, J. (2003) J. Biol. Chem. 278, 26315–
 Yang, J., Meng, Q. and Liu, X.Q. (2004) Mol. Microbiol. 51,
 Henry, T., Iwen, P.C. and Hinrichs, S.H. (2000) J. Clin.
Microbiol. 38, 1510–1515.
 Magrini, V. and Goldman, W.E. (2001) Trends Microbiol. 9, 541–
 Woods, J.P. (2002) Fungal Genet. Biol. 35, 81–97.
 Latge, J.P. (2001) Trends Microbiol. 9, 382–389.
 Dotis, J., Panagopoulou, P., Filioti, J., Winn, R., Toptsis, C.,
Panteliadis, C. and Roilides, E. (2003) Infection 31, 121–124.
X.-Q. Liu, J. Yang / FEBS Letters 572 (2004) 46–50