| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
Departments of Reproduction, Obstetrics, and Herd Health,2 and Virology, Parasitology, and Immunology,3 Faculty of Veterinary Medicine, Ghent University, 9820 Merelbeke, Belgium
ABSTRACT
In the present study, the in vitro interaction of embryos with pseudorabies virus (PRV) and porcine reproductive and respiratory syndrome virus (PRRSV) was investigated by viral antigen detection and by evaluating the expression of virus receptors, namely, poliovirus receptor-related 1 (PVRL1; formerly known as nectin 1) for PRV and sialoadhesin for PRRSV. Embryonic cells of zona pellucida intact embryos incubated with PRV remained negative for viral antigens. Also, no antigen-positive cells could be detected after PRV incubation of protease-treated embryos, since the protease disrupted the expression of PRVL1. However, starting from the five-cell-stage onwards, viral antigen-positive cells were detected after subzonal microinjection of PRV. At this stage, the first foci of PVRL1, also a known cell adhesion molecule, were expressed. At the expanded blastocyst stage, a lining pattern of PVRL1 in the apicolateral border of trophectoderm cells was present, whereas the expression in the inner cell mass was low. Furthermore, PVRL1-specific monoclonal antibody CK41 significantly blocked PRV infection of trophectoderm cells of hatched blastocysts, while the infection of the inner cell mass was only partly inhibited. Viral antigen-positive cells were never detected after PRRSV exposure of preimplantation embryos up to the hatched blastocyst stage. Also, expression of sialoadhesin in these embryonic stages was not detected. We conclude that the use of protease to investigate the virus embryo interaction can lead to misinterpretation of results. Results also show that blastomeres of five-cell embryos up to the hatched blastocysts can become infected with PRV, but there is no risk of a PRRSV infection.
embryo, poliovirus receptor-related 1, porcine reproductive and respiratory syndrome virus, pseudorabies virus, receptor, sialoadhesin
Due to the recent progress in nonsurgical embryo transfer (ET) in pigs [1] and cryopreservation of porcine embryos [2], most obstacles for the implementation of commercial porcine ET have disappeared. It is generally accepted that the risk of pathogen transmission by ET is lower than that by trading animals or semen. However, there are still some sanitary risks inherent to ETs that warrant more research on embryo-pathogen interaction. Until now, most studies on embryo-virus interactions have been performed with embryos that are surrounded by the zona pellucida (ZP) [3–5], and the ZP is considered to be a firm physical and chemical barrier against pathogens [6]. Whether its barrier function remains guaranteed under all circumstances is not clear. Some of the smallest viruses seem to be able to trespass the ZP in mouse [7] and perhaps pig embryos [5]. Embryo handling itself, or routinely used techniques, such as intracytoplasmic sperm injection, blastomere biopsy, and embryo cryopreservation, may induce ZP damage [8] and form ports for viral entry. Under these conditions, every virus may reach embryonic cells. For that reason, it is necessary to have better insights in to the susceptibility of these cells to infection with viruses with a tropism for the reproductive tract. In pigs, two such viruses are pseudorabies virus (PRV) and porcine reproductive and respiratory syndrome virus (PRRSV). These viruses, originating from blood or semen, may come in contact with embryos and, as such, may represent a sanitary risk for the international trade in porcine embryos.
PRV, a member of the 
herpesviridae, causes return to oestrus, abortion, or birth of dead or weak piglets in sows. In vitro inoculation of ZP-intact embryos with PRV [3, 4] and ET experiments [9, 10] showed that PRV can be associated with embryos. Using electron microscopy, Bolin et al. [3] observed that PRV particles remain limited to the outer surface of the ZP. In an in vitro PRV inoculation experiment with ZP-free, 2- to 16-cell-stage embryos, no antigen-positive blastomeres were detected 48 h after inoculation. Therefore, Bolin et al. [11] concluded that 2- to 16-cell-stage embryos were refractory to PRV.
PRRSV, a member of the Arteriviridae, causes reduced conception rates, late-term abortion, and an increase in dead/mummified fetuses. Incubation and microinjection of ZP-intact 4- to 16-cell-stage embryos with PRRSV had no effect on further embryonic development, and, using RT-PCR and indirect immunofluorescence, PRRSV was not detected in association with the embryos. Prieto et al. [12] concluded that 4- to 16-cell-stage embryos were not susceptible to PRRSV infection.
The resistance of cells to a virus infection may be determined by several factors. Absence of one or more virus receptors in the plasma membrane prevents the virus from entering the cell and is, as such, one of the most important resistance factors. Poliovirus receptor (PVR)-related 1 (PVRL1, formerly known as nectin 1) has been shown to be a major receptor for PRV. An interaction between the envelope glycoprotein gD and PVRL1 enables the virus to enter mammalian cells [13, 14]. Physiologically, PVRL1 has been characterized as a Ca2+-independent immunoglobulin (Ig)-like cell-cell adhesion molecule in cadherin-based adherens junctions (AJ). It is expressed in a variety of cells, most notably in epithelial cells and neurons [13, 15]. Porcine sialoadhesin has recently been demonstrated to act as an internalization receptor for PRRSV [16]. Porcine sialoadhesin is a member of the sialic acid-binding Ig-like lectins expressed on specific subsets of macrophages [17].
The specific objectives of the present experiments were: (1) to determine the susceptibility of 2-cell-stage to hatched-blastocyst-stage embryos to PRV and PRRSV and (2) to relate the susceptibility with specific viral receptor expression.
Dulbecco PBS was obtained from GIBCO-BRL Life Technologies (Merelbeke, Belgium). Fetal bovine serum (FBS) was purchased from Biochrom AG (Berlin, Germany) and tested negative for the presence of neutralizing antibodies against PRV and PRRSV using a virus neutralization test described by Andries et al. [18] and Swenson et al. [19], respectively. Propidium iodide was purchased from Molecular Probes (Leiden, The Netherlands). The other components were bought from Sigma (Bornem, Belgium). All media were passed through a sterile 0.22-µm filter (Millipore Corporation, New Bedford, MA).
Embryo Collection, Culture, and Manipulation
All procedures used were in accordance with guidance principals for care and use of laboratory animals of the Laboratory Animal Ethical Commission of Ghent University. In 2 yr, a total of 103 multiparous culling sows (Rattlerow-Seghers, Buggenhout, Belgium) were used in the present study. The sows were free of PRV and PRRSV, and had anti-PRV and anti-PRRSV antibodies due to vaccination. They were superovulated with 1500 IU eCG i.m. (Folligon; Intervet, Boxmeer, The Netherlands) 3 days after weaning, followed by 1500 IU hCG i.m. (Chorulon; Intervet) 72 h later. The sows were fixed-time inseminated with boar semen of proven fertility 24 h after hCG administration. They were killed either 48 h after insemination to generate 2-cell-stage embryos (n = 1960), or 144 h after insemination to obtain hatched blastocysts (n = 245). The oviducts of the sows killed 48 h after insemination were flushed with 15 ml of prewarmed Dulbecco PBS supplemented with 0.5% FBS. After 10 washings [20] with NCSU-23 [21] under oil, embryos were cultured in NCSU-23 at 39°C in 5% CO2 in air. Two-cell (48 h after insemination; n = 235), four-cell (57 h after insemination; n = 235), five- to eight-cell (77–86 h after insemination; n = 470), morula (102 h after insemination, n = 510) and early blastocyst (135 h after insemination; n = 510) -stage embryos were identified using morphological criteria [22]. To get ZP-free embryos, they were incubated in 2.5% protease (Sigma P-6911) in NCSU-23 for approximately 2 min at 37°C [23]. In the group of sows that were killed 144 h after insemination, the proximal third of both uterine horns was clamped, removed, and subsequently flushed with 150 ml of prewarmed Dulbecco PBS supplemented with 0.5% FBS. After washing procedures, expanded blastocysts (n = 245) were cultured for 24 h in NCSU-23 with 2% FBS until reaching the hatched-blastocyst stages, determined based on morphological criteria.
Pseudorabies strain 89V87 and PRRSV strain Lelystad virus (LV) were used for inoculation. Pseudorabies strain (89V87) [24] was passaged twice on PK-15 cells, leading to a stock concentration of 109.0 50% tissue culture infective dose (TCID50)/ml. The LV strain was isolated from aborted fetuses [25] and passaged 13 times on porcine alveolar macrophages, resulting in a stock virus titer of 106.9 TCID50/ml. To carry out the subzonal microinjections, the PRRSV stock had to be concentrated. Therefore, the viral stock was ultracentrifuged at 100 000 x g at 4°C for 3 h. Then, the PRRSV titer was adjusted to 108.7 TCID50/ml, using the method of Reed and Muench [26].
ZP-intact (n = 280) and ZP-free (n = 320) preimplantation embryos were incubated with 105.0 TCID50 PRV (89V87) or PRRSV (LV) per ml for 1 h. Control embryos (n = 480) were incubated under the same conditions in NCSU-23, but without virus. Subzonal virus microinjection of pig embryos was carried out using an inverted Leica microscope (Leica Microsystems GmbH, Heidelberg, Germany), at a magnification of 400x, in microdrops of NCSU-23 medium covered with mineral oil in Petri dishes. Embryos (n = 240) were microinjected with 103.0 TCID50 PRV or PRRSV suspended in 1 nl PBS. Control embryos (n = 40) were microinjected under the same conditions with PBS. To validate the subzonal microinjection model, 103.0 TCID50 PRV or PRRSV suspended in 1 nl PBS was injected close to cells of permissive cell lines (PK-15 cells and macrophages, respectively).
After virus exposure, all embryos were washed 10 times in NCSU23 under oil. At 0, 24, and 48 h after virus incubation, embryos were evaluated for morphology and development. Embryonic development was determined by investigating the rate at which embryos attained the next developmental stage. Impaired in vitro development was defined as embryos not reaching the next developmental stage. Embryos were collected 48 h after incubation, and blastomeres were examined for viral antigen as a marker for viral infection.
The ratio of infected embryos and the ratio of viral antigen-positive cells per infected embryo were determined by immunofluorescence staining. The embryos were fixed in 4% paraformaldehyde at room temperature for 1 h. After fixation, the embryos were washed three times in polyvinyl pyrrolidone (PVP) solution (1 mg PVP/ml PBS). The embryos were treated with 0.5% Triton X-100 (in PVP solution) for 1 h. After three 5-min washes in PVP solution, treatment and control embryos were incubated for 1 min at 39°C in 10% goat serum. Next, the treatment and control embryos were incubated for 1 h at 39°C with a 1:100 dilution of monoclonal antibodies (mAbs) against PRV (1C11), or against PRRSV (A27). After washing thoroughly in PVP solution, embryos were transferred to a 1:100 fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG for 1 h at 39°C. Nuclear staining by propidium iodide was performed to assess the embryonic cell number. After mounting the embryos in glycerol with 1,4-diazabicyclo (2.2.2) octane (25 mg/ml), the number of fluorescent cells was counted in each embryo at a magnification of 400x with a Leica TCS SP2 laser scanning spectral confocal system linked to a Leica DM IRB inverted microscope (Leica Microsystems GmbH, Heidelberg, Germany). An Argon laser was used to excite FITC (488 nm) and PI (586 nm) fluorochromes. Total embryo height was evaluated, and sections were made at 3-µm intervals. Analysis of the images was performed with Leica confocal software.
Distribution of PVRL1 and Sialoadhesin
Staged embryos (n = 250) were incubated in Ca2+-free medium for 30 min before fixation in PBS containing 1% paraformaldehyde. After fixation, the embryos were thoroughly washed in PVP solution and treated with 0.5% Triton X-100 (in PVP solution) for 1 h. After three 5-min washes in PVP solution, embryos were stained for 1 h with either mAb CK41 (1:100) (provided by C. Krummenacher, G. Cohen, and R. Eisenberg, University of Pennsylvania), directed against PVRL1, or with mAb 41D3 (1:100), directed against porcine sialoadhesin [27]. The isotype-matched (IgG1) mAb, 13D12 (1:100), directed against PRV glycoprotein gD [28], was used as a control. After 3 wash steps in PVP solution, embryos were labeled with a 1:200 dilution of FITC-labeled goat anti-mouse secondary antibody. Nuclear staining by propidium iodide was performed to assess the embryonic cell number. After mounting in glycerol with 1,4-diazabicyclo (2.2.2) octane (25 mg/ml), embryos were analyzed using confocal laser scanning microscopy at a magnification of 400x or 630x.
PRV Blocking Assay Using mAb CK41
Various concentrations of CK41 were added to hatched blastocysts (n = 60) cultured in Dulbecco PBS with 2% FBS and held at 15°C for 1 h. The isotype-matched (IgG1) mAb, 41D3 (10 µg/ml), directed against sialoadhesin [27], was used as a control (n = 15). Subsequently, embryos were incubated in NCSU23 with various concentrations of CK41 and 105 TCID50 PRV (strain 89V87) per ml for 1 h at 39°C in 5% CO2 in air. Afterwards, embryos were washed 10 times in NCSU23 under oil and cultured in NCSU23 with 2% FBS for 12 h. After fixation in 4% paraformaldehyde, viral infection was assessed by investigating viral antigen expression using immunofluorescence staining and confocal laser scanning microscopy.
The number of viral antigen-positive embryos and differences in rates of development were analyzed using chi-square analysis. Fisher exact tests were applied when small numbers were involved. Logistic regression was used to compare the average ratio of antigen-positive cells in infected embryos for the developmental stages, and univariate analysis of variance with replicate as random variable was used to analyze PVRL1 expression and PRV infection in the PRV blocking assay. Variables were considered to be significant at a 0.05 level (two-sided). The statistical analyses were performed using SPSS (SPSS 12; SPSS Inc., Chicago, IL).
Interaction of Preimplantation Embryos with PRV
Virus exposure of ZP-intact and ZP-free embryos by incubation with PRV. For all ZP-intact embryos incubated with PRV, embryonic cells were negative for viral antigens after a 48 h embryo culture period, and further embryonic development was not different from negative control embryos. PRV-positive cells were also not detected after the incubation of PRV with preimplantation embryos that were made ZP free using 2.5% protease before virus incubation. Incubation of hatched blastocysts with PRV led to antigen-positive embryonic cells and had a detrimental effect on further embryonic development (see Table 1).
|
Virus exposure of ZP-intact embryos by subzonal microinjection of PRV. Embryos up to the 4-cell stage were negative for viral antigen 48 h after microinjection. Antigen-positive blastomeres were found in ZP-intact 5- to 8-cell embryos, morulae, and blastocysts 48 h after subzonal microinjection with PRV, and their further embryonic development was significantly lower compared with that of control embryos injected with PBS (see Table 1 and Fig. 1).
|
Interaction of Preimplantation Embryos with PRRSV
The incubation or microinjection of ZP-intact and ZP-free preimplantation embryos with PRRSV did not result in viral antigen-positive blastomeres, nor was there any effect on further embryonic development (see Table 2).
|
Presence and Distribution of PVRL1 and Sialoadhesin in Preimplantation Embryos
To investigate whether the detection of viral antigen-positive cells after virus exposure was consistent with the presence/absence of virus receptors, immunofluorescent stainings were performed to visualize receptors for PRV (PVRL1) and PRRSV (sialoadhesin) (see Figs. 2 and 3). In ZP-intact embryos, the ZP showed foci of intense PVRL1 staining (Fig. 2A, arrowhead). The first distinct PVRL1 staining of embryonic cells occurred in the precompact 5- to 8-cell stages of development. The PVRL1 expression was diffusely distributed throughout the cells, being more concentrated in a scattered pattern of punctuate foci on the cell membrane (Fig. 2, B''' and C''', arrowheads) up to the young blastocyst stages. However, in addition to the diffuse cytoplasmic signal, the expanded and hatched blastocyst stages displayed a clear lining pattern of PVRL1 staining (Fig. 2, D''' and E''', arrowheads) at the apicolateral border of some trophectoderm cells. This lining pattern, localized at presumed sites of tight junctions (TJ), was expressed in mural (adjacent to the blastocoelic cavity) trophectoderm cells in expanded blastocyst (Fig. 2D'''), and in mural and polar (adjacent to the inner cell mass [ICM]) trophectoderm cells in hatched blastocyst (Fig. 3, A'' and B''). Only weak signals were detected at contact sites within the ICM (Fig. 3A''). Having demonstrated that no viral antigen-positive cells were detected after PRV exposure of protease-treated embryos, the PVRL1 expression in these protease-treated embryos was also determined, and no positive signals could be detected (Fig. 2, F'' and 2F''').
|
|
Sialoadhesin staining was never observed in preimplantation embryos up to the hatched blastocyst stage (data not shown; n = 20 per stage of development, 3 replicates). Also, the isotype-matched 13D12 staining, which was used as a negative control, did not give a signal in the preimplantation-stage embryos.
PRV Blocking Assay Using mAb CK41
To examine if PRV uses PVRL1 to enter embryonic cells, a blocking assay was performed in hatched blastocysts using mAb CK41 in different concentrations. Monoclonal Ab CK 41 reduced PRV infection in the trophectoderm cells in a clearly dose-dependent manner—up to 98% at a concentration of 10 µg/ml. Also, CK41 addition resulted in a trend of reduced PRV infection in the ICM cells, but the differences between the different concentrations were not significant. Addition of the isotype-matched control mAb, 41D3, at a concentration of 10 µg/ml had no effect on PRV infection (see Figs. 4 and 5).
|
|
In the present study, the susceptibility of embryos for a viral infection was determined by investigating viral antigen expression after virus exposure, indicating that the virus successfully entered, replicated its genome, and produced viral antigens in the embryonic cells. For PRV, the first embryonic stages of development were not susceptible for an infection. Starting from the five-cell stage, viral antigen-positive cells were detected after subzonal microinjection of PRV. This coincided with the cellular expression of a known PRV entry receptor, PVRL1. All preimplantation stages up to the hatched blastocyst remained resistant to a PRRSV infection. The absence of sialoadhesin, the internalization receptor for PRRSV, is likely to be of major importance for this refractory state of embryos.
In analogy with the results of Bolin et al. [11], our data indicate that the porcine ZP forms a protective barrier against PRV infection. PRV-positive embryonic cells were never detected in ZP-intact embryos incubated with PRV, whereas subzonal microinjection in embryos starting from the 5-cell stage led to a PRV infection. Important to mention here were the differences in the results of the microinjection experiments compared with the results obtained by PRV incubation of ZP-free embryos after protease treatment. In one way or another, protease treatment made blastomeres refractory to a PRV infection. This likely explains why Bolin et al. [11], who also used protease to remove the ZP from pig embryos, found that porcine embryonic cells up to the 16-cell stage were refractory to a PRV infection. In the present study, an explanation was found for this contradiction. It was shown that protease treatment caused a disruption of the dimensional structure of PVRL1, since the conformation-dependent mAb, CK41 [29], was not able to bind to the V-like ectodomain of PVRL1. The finding that enzymes such as proteases have an effect on viral receptors has implications for previous data regarding virus susceptibility of embryonic cells. The present results of the subzonal microinjection experiments indicate that this technique is the method of choice to circumvent the use of protease for investigating the interaction between pathogens and embryonic cells.
Based on the results of the in vitro embryo-PRRSV experiments, we can conclude that preimplantation stages up to the hatched blastocyst stage are not susceptible to a PRRSV infection. Because it has been shown that a transplacental infection of embryos can be established in a 20-day pregnancy [30], further research is needed to investigate at which developmental stage pig embryos become susceptible to a PRRSV infection.
PVRL1 is a known entry receptor for herpes viruses, including herpes simplex virus types 1 and 2, bovine herpes virus type 1, and PRV [13], and it is also a molecule that plays an important role in cell adhesion and in a variety of cell-cell junctions, such as AJ and TJ [31]. These AJs and TJs only become fully functional at embryo compaction [32, 33]. In a mouse embryo model, the expression of nectin-2 was undetectable before compaction [34]. However, in the present study, the appearance of PVRL1 expression coincided with the development of precompact five- to eight-cell-stage embryos. This latter finding is in agreement with data from murine embryos, where other AJ components, such as cadherin 1, and ßcatenins, and vezatin, are already present in early precompaction stages of development [35]. The PVRL1 distribution in blastomeres is also similar to that described for nectin-2 and these other AJ components; namely, diffusely distributed in the blastomeres, being enriched at cell-cell contact sites. At the expanded blastocyst stage, the first continuous staining of PVRL1 in the cell membrane was detected in the apicolateral border of mural trophectoderm cells (Fig. 2D'''). This PVRL1 expression may contribute to the formation of the TJ seal surrounding the blastocoel cavity that controls paracellular movement of water, by which it also controls the expansion of the blastocyst [36]. In the hatched blastocyst stage, the PVRL1 expression was continuous in all (mural and polar) trophectoderm cells (Fig. 2E'''). This might be related to a greater need of cell adhesion in hatched embryos to keep embryonic cells together without ZP in processes such as elongation, spacing, and migration. PVRL1 expression in the undifferentiated cells of the ICM of blastocysts was low. The first tissue to differentiate during mammalian development is the trophectoderm epithelium, and since PVRL1 is predominantly expressed in tissues of epithelial origin, it makes sense that expression is situated in the trophectoderm rather than in the undifferentiated cells of the ICM. Moreover, PRV entry in trophectoderm cells could be significantly blocked (up to 98%) in a dose-dependent way by mAb CK41, whereas the PRV infection of ICM cells could only be partially prevented (block up to 47%) by adding mAb CK41. The finding that the ICM cells that are enclosed by trophectoderm cells could be infected with PRV also provides an indirect proof of the existence of substantial gaps in the polar trophectoderm cells in eight-day-old embryos described by Barends et al. [37]. PRV blocking assays with CK41 in other cell types also showed variable results. In a CHO K1-derived cell line that stably expresses porcine PVRL1, PRV entry was blocked, whereas entry of PRV in PK15 cells was unaffected by CK41 [15]. A possible explanation for this difference in blocking capability of CK41 may be that PK15 and ICM cells express other PRV entry receptors, whereas trophectoderm and CHO K1-derived cells do not express those receptors. Five herpesvirus receptors have been identified to date: tumor necrosis factor receptor superfamily, member 14 (TNFRSF14), which has no entry activity for PRV; PVRL1; nectin-2; PVR; and 3-O-sulfated heparan sulfate [38–41]. Of these entry receptors, the expression of nectin-2 in the ICM has already been described in a murine embryo model [34]. Further research will show whether nectin-2, or possibly another receptor, may be responsible for the PVRL1-independent manner of PRV entry in porcine ICM cells.
In the present study, an innovative way of investigating virus-embryo interaction was applied, linking the results of in vitro virus exposure with virus receptor expression. In the model used, data for some embryonic stages were obtained after in vivo development, whereas other embryos were assayed after an in vitro culture. As prolonged in vitro culture of embryos can induce changes in early gene expression [42], the distribution of the virus receptors might have been influenced. However, the expression of PVRL1 in successive stages of development was logical in the sense that there was an evolution to a more extensive distribution during further embryonic development, even for embryos with a prolonged in vitro culture.
Based on our results, we can conclude that the use of proteases to remove the ZP and establish contact between blastomeres and viruses can affect virus receptor expression, and may lead to wrong conclusions. The findings of the virus incubation experiments confirm the hypothesis that an intact ZP forms a sufficient barrier against most pathogens. Exposure of embryonic cells to PRV revealed that blastomeres from the five-cell stage onwards are susceptible to PRV infection. Furthermore, the expression patterns of PVRL1 in blastomeres of preimplantation embryos were in accordance with this finding. Based on the results of the PRRSV exposure and sialoadhesin expression of embryos, we can conclude that blastomeres of preimplantation embryos up to the hatched blastocyst stage are not susceptible to PRRSV infection. By looking into the distribution of PVRL1 in expanded and hatched blastocyst, a difference in the expression of PVRL1 in the trophectoderm and ICM cells was detected. Also, the results of the blocking assay using mAb CK41 suggest that ICM cells can use a PRV entry receptor other than PVRL1. Further research will focus on the effects this difference can have on virus replication in embryonic cells.
ACKNOWLEDGMENTS
The authors thank B. Buysse and N. Buys of Rattlerow-Seghers for providing the donor sows; J. Mestach, I. Lemahieu, G. Spaepen, and C. Boone for excellent technical assistance; and G. Van Minnenbruggen, P. Delputte, and N. De Regge for helpful comments and discussion. They are also very grateful to C. Krummenacher for supplying CK41.
Correspondence: 1B. Mateusen, Department of Reproduction, Obstetrics, and Herd Health, Faculty of Veterinary Medicine, Salisburylaan 133, 9820 Merelbeke, Ghent University, Belgium. FAX: 32 9 264 77 97; e-mail: bart.mateusen{at}ugent.be
Received: 22 August 2006.
First decision: 13 September 2006.
Accepted: 13 November 2006.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
