01/02 Research Results - 5.Possum Biocontrol Category

5. Possum Biocontrol Category

5. Possum Biocontrol Category

5.1 PBC 220

Programme Title:	Population responses of wild possum to fertility control
Programme Leader:	Dr P E Cowan
Institution:	Landcare Research

Summary

Density reductions resulted in population sizes in October 2001 that were lower than the long-term mean October abundances for all sites, with the exception of the Turitea 50% sterility site. However, population reductions appear to be sustained only on the 50% sterility site in the Orongorongo Valley, with an 8% increase in population abundance in Feb/March 2002 compared with October 2001. This compares with a 27% and 121% increase on the control site and 80% sterility sits, respectively. For all sites at Turitea, population abundance in Feb/March 2002 has increased dramatically since October 2001 with a 137%, 80% and 132% increase in population abundance on the control, 50% and 80% sterility sites, respectively.

The contribution to the recruitment rate of yearlings over this period from breeding by resident females could only be determined for three sites. On the Orongorongo control site, breeding by resident females contributed 64% of the total recruitment of yearlings. In comparison, breeding by resident females on the Orongorongo 50% sterility site and Turitea 80% sterility site only contributed around 20% of the total recruitment of yearlings.

Background

The effect of fertility control on the population dynamics of possums is being investigated on six 14-ha study sites. Sterilisation treatments, consisting of either 0%, 50% or 80% of resident females sterilised at a particular site, were applied in early 1996, and adjustments to maintain the levels of sterility were made annually on recruits. Possums were live trapped for-4 consecutive nights, three times a year, and population parameters such as breeding, survival and recruitment were estimated and compared between sterilisation treatments. Population monitoring is ongoing.

During March/April and again in October 2001, we experimentally reduced populations on the existing trapping grids by approximately 80% to examine how fertility control acts to slow rates of recovery after conventional control. To date, limited monitoring data is available to fully assess this. However, from October 2001 to Feb/March 2002, we estimated changes to population size, breeding, survival and recruitment for each sterilising treatment to assess the effect of density reductions on rates of recovery so far.

Approach & Outcomes

Following density reductions, the majority of populations had wholly or partly recovered to near their long-term equilibrium density by February/March 2002, the exception being the Orongorongo Valley 50% sterility site and to a lesser extent, the 80% sterility site. Recruitment rates were generally higher overall on density reduction sites than on the control site (no density reduction). However, local recruitment rates of yearling possums appeared to make a much greater contribution to yearling recruitment on the control (no density reduction) site than on either the Orongorongo Valley 50% sterility site or the Turitea 80% sterility site. Hence, immigration of possums was largely responsible for driving the recovery rate of populations subject to density reductions, especially for sites at Turitea.

To date, we do not have sufficient data, post-density reduction, to enable any meaningful analysis of potential differences in population vital rates between sterility treatments. However, due to the recovery of population density to pre-control levels on 3 of the 5 density reduction sites, there will be little relevant data on these potential effects for the current annual cycle (October 2001-October2002). In addition, despite the intensive trapping, too few pouch young were tagged this year to enable precise estimation of juvenile survival rates. This is likely to be an ongoing problem, as the combination of high sterility levels and low population size mean sufficient numbers of pouch young are unlikely to be tagged in subsequent years, especially on 80% sterility sites.

Monitoring should continue to determine the effect of fertility control on the rate of population recovery and the possible characterisation of density-dependent processes.
Sterility levels on 80% treatment sites should be reduced to better enable precise estimation of both juvenile and adult breeding female survival at reduced density. This approach will maximise the potential to measure the likely variability in these vital rates, which will be a critical factor for the development of the stochastic population model.

5.2 PBC 221

Programme Title:	Blocking Embryonic Development in Brushtail Possums
Programme Leader:	Professor Lynne Selwood
Institution:	University of Melbourne

Summary

Unlike other aspects of reproduction, possum embryonic development has many unique developmental features that make suitable targets for controlling reproduction. Our previous projects have examined three of these:

The proteins associated with polarised vesicles (VAPs) that form during oogenesis and are responsible for the first lineage allocation into pluriblast (embryonic) and trophoblast (placental) tissues (Frankenberg and Selwood, 1998, 2001). We have identified 10 vesicle-associated molecules (VAMs), 9 VAPs and hyaluronate, and consider four of these are worth further testing because of possum specificity and/or embryo lethality effects (Vap5, Vap7, VASA protein, and alpha fetoprotein).
The outer egg coats (CPs) that are essential for normal embryonic development (Selwood, 2000) and for maintenance of blastocyst epithelia in the possum (Cui et al, 2001). We have isolated 7 shell coat proteins of which amino acid sequences have been obtained for 5 and three unique proteins (CP2, CP4 and CP5) are suitable for further studies.
Leukaemia inhibitory factor (LIF) that is present in the possum at the time of implantation (Cui and Selwood, 2000) and is essential for implantation. In each group, genes have been cloned and recombinant protein expressed and purified for testing in possums. We expect that these target molecules will be possum specific and not affect non-target species such as birds and other mammals.

Goal:

Three highly possum-specific targets already cloned, will be tested to determine their immune response and their effects on embryonic development in the possum. Three further identified, possum specific targets will also be cloned.

Background

The aim of our research was to identify proteins which are essential for embryonic development and use these for immunocontraception of the brushtail possum. In our previous projects we established that a number of molecules, many unique to marsupials, are essential for embryonic development in the brushtail possum. The substances examined span all of pre-implantation embryonic development. VAMs are essential for early development and cleavage, CPs are essential at blastocysts stages and LIF is essential at implantation and for proliferation of germ cells.

Approach & Outcomes

Objective 1

Assessment and Cloning of Vesicle-Associated Proteins (VAPs).

The gene vap5 has been already cloned and a recombinant protein expressed. It is a protein with extensive possum unique sequences that was isolated from vesicle-rich oocyte fractions. We are determining whether VAP5 recombinant protein initiates an immune response in the possum and have begun an assessment of its effect on development and oogenesis in in vitro assays.

VAP7 is another unique possum protein identified that we cloned and are expressing using recombinant protein technology for subsequent testing.

The Brushtail Possum oocyte is markedly polarised, with pronuclei and mitochondrial-rich cytoplasm at one pole and vesicle-rich cytoplasm at the other (Frankenberg and Selwood 2000). The polarisation of the conceptus is important in early development (Frankenberg and Selwood, 1998). We have identified 7 vesicle-associated molecules (VAMs) of which three (VAP5, 7 and a fetaprotein) and a protein associated with the germ cell lineage markers, VASA protein that is also polarised in location (Frankenberg and Selwood, 1998) have been selected for further study. VAP5 and 7 are examined here. These molecules are important for early development and for development of the germ cells, ie. sperm and eggs.

Protein expression Using the amino acid sequences previously determined from differential PAGE of oocytes we have identified two vesicle-associated proteins VAP5 and VAP7. We used the following as the trial method for protein expression for our test proteins. Protein production using the plasmid construct pGEX2T-VAP5 was commended for VAP5 using the E.coli bacterial strain, JM109 for transformation. Single colonies of the transformants were

used to inoculate LB culture medium and the cells incubated at 37^oC with agitation overnight. In the preparation scale, cultured cells were diluted by 1:50 in LB medium containing 100mg/ml ampicillin and grown at 37^oC till the density reached OD₆₀₀ 0.5. The inducer, isopropyl b-D-thiogalactoside (IPTG), was added to the culture medium and the incubation continued using the same condition for about 3 hours. The effectiveness of the induction was checked by PAGE analysis at this stage to make sure the protein was produced (Cui and Selwood 2000). The cells were harvested by centrifugation at 4^oC and resuspended in a mild solution. The cells were lysed by sonication and the cell debris was sedimented by centrifugation at 4^oC. Because the protein was fused with glutathione S-transferase, which can be used for further purification, the crude protein was mixed with Glutathione Sepharose 4B matrix and the protein-matrix binding allowed by incubation at room temperature. The fusion protein was finally purified by thorough washes and VAP5 released to the digestion/storage buffer by protease digestion. Where we had problems with this production system, we moved to develop alternative expression and purification systems as outlined below.

Once the protein was expressed we moved to testing the immune response of a limited number of possums. Animals were trapped, held in the colony for several months to stabilise before immune trials commenced. Animals received 300 mg protein in PBS per animal for the first injection in complete Freund’s adjuvant (CFA) and the subsequent injections will be incomplete Freund’s adjuvant (IFA). Controls contained only PBS or PBS/CFA or IFA.

Cloning and characterisation of VAP7. By using the short amino acid sequence obtained from differential PAGE analysis to design a degenerate primer to clone the gene for VAP7, now designated vap7, it was cloned from a PY ovarian cDNA sample. Sequence analysis was used to examine homologies in nucleotide sequences and in the deduced amino acid sequences. Sources such as GenBank, EMBL, etc were used to predict its secondary structural properties such as hydrophobicity, hydrophilicity and antigenicity and searching for domains or motifs that are important for biological functions.

In vitro assays. Assays for assessment of the effect of antibodies of polarised molecules on oocyte/follicle growth and development and embryonic development (cleavage stages) were developed by modification of the protocol developed by Ullman and Butcher (1996) and by further studies on the protocols originally developed by Selwood et al (1992) for other marsupials. For oocyte/follicle growth, follicle stimulating hormone (FSH) and homologous serum and fetal calf serum (FCS) were tested in DMEM.

VAP5 protein production. The plasmid construct pGEX2T-VAP5 with correct orientation was transformed into the bacterial cells, which expressed an inducible 25 kDa protein. The inducible 25 kDa protein expressed proved difficult to separate from the fused glutathione S-transferase (GST). Because of the difficulties encountered in VAP5 purification with the existing expression system, the GST-fusion system, several separation techniques have been attempted to purify the protein. Application of another expression system could be a solution to this problem, either by purifying the VAP5 polypeptide or by providing a fusion protein with a different carrier. The pMAL-c2 system was used for VAP5 purification. In this system, the VAP5 is fused with maltose-binding protein (MBP), which binds to maltose to provide a method of purification. The modified open reading frame of the vap5 cDNA was transferred to pMAL-c2 in the BamHI site, which was further used to transform the bacterial cells of JM109 strains. The fusion protein was also induced with isopropyl-b-D-galactosidase by culturing the cells at 37^oC and released from its cytoplasm by sonication. MBP-VAP5 bound to maltose while others were eliminated by washes. The fusion protein can be eluted or the VAP5 can be separated by the protease Factor Xa. This work is still in progress. VAP5 protein production works well when the protein is produced in small quantities, but when the system is upregulated the digestion system is sometimes unreliable. In our new proposal we have identified some alternative expression and purification systems to be tested.

VAP5 possum immune response. Three possums were trapped, housed in a captive colony and after acclimatisation a test bleed was done to determine basal levels of response. Possums received 300mg VAP5 recombinant protein in complete Freund’s adjuvant (CFA). Controls received saline and adjuvant. Subsequently, possums will receive VAP5 recombinant protein in incomplete Freund’s adjuvant (IFA) for up to three boosts. Test bleeds will be performed 1 week after boosts.

Development of in vitro assay to test the effect of VAPs on reproduction. Normal standards for oogenesis (Eckery et al., 1996; Frankenberg et al., 1996, Frankenberg and Selwood, 2001) and for early embryonic development (Frankenberg and Selwood, 1998, Frankenberg et al, 2001) have already been developed. The techniques for culture of early stages of embryonic development (Casey and Selwood, 2002 submitted) have been established recently. While this protocol was not specifically formulated for marsupials because of limitation of material, it appears adequate for apparently normal development and would allow assessment of VAP antibodies in vitro when possum embryos are available in the breeding season.

Follicle and oocyte growth were continuous in the possum, with significant (P<0.05) oocyte growth occurring after antral appearance, so growth in the possum as in other marsupials is not biphasic as it is in eutherian mammals. Presumably at least part of this is associated with the uptake of proteins into the vesicles shown in previous projects, especially during the antral phases.

Follicles were dissected from the ovary and cultured at the primordial, primary, secondary and tertiary stages. While follicles from pre-pubertal animals are easier to culture from stage to stage and particularly into antral stages, we decided to base our assays on adult ovaries, because they would always be available. Follicle performed better in culture if dissected out of the ovary rather than removed by enzymatic digestion. However, with sufficient time and enough samples it might be possible to develop a better enzymatic protocol. An assessment of follicle growth showed that Follicle Stimulating Hormone (FSH) at 1.0 and 1.5 IU. mL-1 (no significant difference) enhanced follicle growth. Follicles, except in the antral stages, grew in the absence of serum, but growth was significantly enhanced in the presence of either 3% homologous serum or 3% Fetal calf serum (FCS), with no significant difference between the two, except at the antral stage when FCS was more effective. Transition between stages appeared to occur but this requires histological confirmation. This assay is now ready to use for either uptake of labelled VAPs or to determine the effect of VAP antibodies on oocyte and follicle development.

Cloning and characterisation of VAP7

Cloning: VAP7 is present in the vesicle-rich section. A short amino acid sequence available for VAP7 (the seventh identified protein on the PAGE) was used to clone the gene, now designated as vap7. A single cDNA fragment was amplified with the primer pair vap7f and Universal Amplification Primer (UAP) by RT-PCR, and shown on an agarose gel to be about 0.6 kb. These fragments were recovered and sub-cloned and sequenced. The DNA sequence analysis confirmed that the 0.6 kb PCR fragment was a correct copy of the vap7 gene because it contained the vap7f gene specific primer binding region at its 5’ end, a short stretch of putative amino acid sequence matching the determined amino acid residues that are immediately after the N-terminal region where vap7f primer was degenerated and the cDNA fragment showing a poly(A) tail of the mRNA before the UAP. The whole length of the cloned cDNA was 323 bp and encoded 107 amino acid residues. However, a stop codon was present at position 86 of the sequence.

Characterisation: Sequence analysis revealed a short stretch of the sequence sharing homology with less defined genomic DNA sequence from various species, including human, mouse Drosophila cow and rat. Using deduced amino acid sequence for homology searching against the protein database, the preliminary results showed that only about one third of the sequence is homolgous to mouse hypothetical protein XP_140286. The biochemical significance of this homology is to be investigated.

The computer program PeptideStructure was used to predict helices, sheets, and turns. It was also used to predict secondary structure and biochemical properties of the molecule, including hydrophilicity, surface probability, glycosylation sites and antigenic index (AI). No potential glycosylation site was identified across the molecule. The hydropathicity plotting predicted the features of important VAP7 physico-chemical properties. The N-terminal part of VAP7 is more hydrophobic while its C-terminal part is hydrophilic. The C-terminal part possesses higher antigenicity to its N-terminus according to Parker plotting although the whole molecule showed high antigenicity to its N-terminus according to the Welling Method. Four major solvent accessibility sites were predicted at about positions 20, 50, 60 and 75 with the most accessible region lying between 50 and 85.

Protein expression: Because expression of the other proteins has been slower than expected when we upregulated expression there has been insufficient time to develop a protein expression system for VAP7.

Cloning and characterisation of a VASA homologue in the possum. VASA protein or RNA is associated with the germ plasm of a variety of vertebrates including amphibians and mammals (Toyooka et al, 2000) and targeted mutation of mouse VASA (Mvh) results in aspermia in homozygote mice and shows that VASA is essential for proliferation and differentiation of mouse male germ cells (Tanaka et al, 2000). Because we identified VASA-like products in oocytes (Frankenberg and Selwood, 2001) and early cleavage stages Frankenberg and Selwood, 1998) and in a polarised location, we thought VASA cloning was desirable. Based on published information, gene specific primers were developed by computer program and used in the RT-PCR reactions. Two cDNA fragments were amplified from testis tissue (4.5 kb and 4.2 kb) and these fragments were submitted to nucleotide sequence analysis, showed that the 4.2kb fragment contains the DNA sequence encoding the DEAD box, the most conserved region of the vasa gene from eutherian species.

Possum immune response to VAP5 Because of the low yield of VAP5, initial trials are being done with two possums and one control. The numbers will be built up when sufficient protein is available. Possums were immunised with 300mg/Kg possum + CFA. Subsequent injections will be with IFA. Bleeds will be taken after the 2^nd and 3^rd injection.

Objective 2

Assessment and Cloning of Coat Proteins

For the coat protein (CP4) that has already been cloned, characterised and expressed as a recombinant protein, we tested its ability to raise an immune response in possums and examined the secretion pattern and the effect of its antibody on uterine tissues in vitro. This coat protein has major unique sequences and is expected to be a novel protein. We attempted to clone and characterise one of the two other coat proteins, CP2 and CP5, which have been identified and also appear to have major unique sequences.

Because the shell coat is marsupial specific and we have demonstrated earlier that it is essential for normal blastocyst formation and for maintenance of blastocyst epithelia (Selwood, 2000), we expected an immune response to the shell coat proteins to terminate embryonic development well before implantation stages. Early analysis of the coat proteins showed them to be unique to marsupials at least, with no similar proteins located in the data base (Casey et al, 2002)

Protein production – CP4. Recombinant CP4 protein was over expressed using the Glutathione S-Transferase (GST) fusion system as described previously.

Testing possum immune response. Possum immune response was tested as described above for VAP5.

Begin assessment of effect of CP4 on reproduction using an in vitro assay system.

Normal standards have been established by Frankenberg and Selwood (1998) and Casey and Selwood (2002 submitted) for pre-implantation development and for coat protein secretion in the uterus by Casey et al (2002) and were used as a basis for comparison.

In vitro assays were established by using uteri obtained at the late follicular phase when uteri are about to enter the proliferative phase (Casey et al, 2002). These were minced and digested with collagenase prior to establishment of primary culture over matrigel. Protein production was then determined by immunocytochemistry of the sections of cell monolayers and secretion by examination of media by PAGE and Western blot analysis using the polyclonal antibody to CP4.

Up regulation of transcription was attempted using oestradiol and secretion by using progesterone at physiological levels (see Fletcher and Selwood, 2000 for review). The effect of CP4 antibody on monolayers was determined.

Cloning and characterisation of CP2 and/or CP5. PAGE analysis followed by amino acid sequencing of possum CPs identified 5 proteins (Casey et al, 2002). The second coat protein (CP2) and/or fifth (CP5) would be the molecules to target according to the preliminary analysis of the short stretch of amino acids, which showed unique sequences. The cloning and characterisation of genes coding for CP2/CP5 (designated cp2 and cp5 respectively) from the possum were attempted based on the amino acids sequence(s).

The techniques used for the cloning and characterisation of cp4 are suitable for cp2/cp5. Briefly, a degenerate oligonucleotide primer and a nesting primer were designed on the basis of the N-termini of CP2/CP5 and synthesised. These primers were used in cDNA cloning. The uterine mRNA were isolated and used as templates to synthesise cDNA. The technique 3’RACE would be then used for isolation and characterisation of the cp2/cp5 using the gene specific primers and an anchor primer which binds specifically to the 3’-end of the cDNA. Alternatively a uterine cDNA library may be established and screened in the laboratory for cp2/cp5 cloning. Sequence analysis would emphasise homology searching against the existing database, such as GenBank, EMBL etc and on predicting its secondary structural properties such as hydrophobicity, hydrophilicity and antigenicity and searching for domains or motifs that are important for biological functions.

Protein production – CP4. Our protein production procedures were based on the system we used for production of the cytokine LIF from the possum, tvLIF. The cDNA encoding for CP4 was cloned into a plasmid vector and used for production of recombinant protein with a bacterial expression system. The glutathione S-transferase (GST)-fusion protein was expressed and purified while it bound to specific matrix and a further protease cleavage procedure was used to release the CP4. It was found that the purification of CP4 polypeptide was satisfactory using the established conditions, but the protein did not survive long-term storage. For this reason we have decided that fusion proteins (GST fusion protein) would be used in further molecular characterisation and immune trials in conjunction with using GST protein as a control. Conditions established for mouse and human protein production are now undergoing modification. Large amounts of recombinant CP4 were produced and tested in the immune trial and antibody screening.

Testing possum immune response. An earlier trial was not continued because we found that the protein degraded on storage. Trials have been recommenced. Three possums were trapped. housed in a captive colony and after acclimatisation a test bleed was done to determine basal levels of response. Two possums received 300mg CP4 recombinant protein in complete Freund’s adjuvant (CFA). Controls received saline and adjuvant. Subsequently, possums will receive CP4 recombinant protein in Freund’s incomplete adjuvant (IFA) for up to three boosts. Test bleeds will be performed 1 week after boosts. Numbers will be built up when sufficient protein available.

Assessment of CP4 secretion and assessment effect of CP4 on reproduction using an in vitro assay system. Casey et al (2002) in an immunological and ultrastructural study established that the shell coat is secreted in the utero-tubal junction, where uterine glands are present and in the glandular epithelium of the uterus. Secretion occurs in late follicular stages, cleavage and early unilaminar blastocyst stages, declines in late unilaminar and bilaminar blastocyst stages and is elevated in trilaminar and embryonic stages and probably continues until implantation. Immunestaining using a polyclonal antibody raised in mice against S. macroura coats, showed positive staining in apoptotic cells sloughed off before the proliferative phase, uterine glandular cells of the more superficial glandular regions and at trilaminar and emryonic stages in milk producing cells of the basal regions of uterine glands. These findings have been replicated by our CP4 antibody and also by the red stain of Masson’s trichrome.

We have established an in vitro assay system using uterine monolayers grown over matrigel. This has been applied both to the possum and the dunnart and allows us to test our products at most times in the year, instead of being restricted to the possum breeding season. Epithelial cells attached within 72 hrs of plating and formed a pavement epithelium by day 7. Glandular structures developed from day 1 and formation could be regulated hormonally and was cell density dependent. Preliminary staining has confirmed the secretion of shell coat protein in these assays. Unfortunately, coat secretion, while present as shown by antibody staining, was not sufficient to be detected in the media by PAGE. Uterine epithelial cell proliferation increased significantly in the presence of 110ng mL-1 progesterone and CP4 antiserum. These effects need more analysis.

Cloning and characterisation of CP2 or CP5. Because of the difficulties encountered in the recombinant protein isolation and purification for tvLIF, CP4 and VAP5, on which we have invested more time so that enough proteins could be produced for functional assays, little has been done for cloning and characterisation of the cp2 and cp5 genes. A partial amino acid sequence has been obtained for CP2 and CP5 by electrophoresis and sequencing analysis. These partial sequences showed unique features and interesting biochemical characteristics, which makes them a good candidate target molecule for fertility control in possum. Degenerate primers were synthesised according to deduced nucleotide sequence. Several cDNA fragments have been successfully obtained by 3’RACE technique. These amplified products are to be clarified.

Objective 3

Assessment and cloning of implantation proteins

On the basis of what is known of LIF biology, interfering with normal LIF function either through LIF or its receptor, should affect normal embryonic development at implantation stages (flat embryo) and also primordial germ cell (PGC) proliferation between the primitive streak stages to soon after birth (Eckery et al, 1996, Frankenberg et al, 1996).

LIF protein is essential for implantation (Stewart et al, 1992) and is also one of the growth factors required for proliferation of the germ cells (Matsui et al., 1992). We have previously cloned and characterised the possum LIF gene (pLIF) and found expression patterns similar to eutherian mammals (Cui and Selwood, 2000).

LIF protein expression. Described as basic initial protocol under Objective 1.

Testing possum immune response. Possum immune response was tested as described above for VAP5.

Begin assessment of effect of LIF on reproduction using an in vitro assay system.

Assay systems used were based on fibroblast feeder layers, which are a standard source of LIF in the laboratory, primordial germ cells, which have been shown by Matsui et al (1992) and others to be dependent on LIF for proliferation, and uterine monolayers described above. Normal standards for late pre-implantation development established by Casey and Selwood (2002 submitted) and Richardson et al (1997), for uterine structure by Casey et al (2002) and for primordial germ cell proliferation (Eckery et al, 1996) and morphology (Frankenberg and Selwood, 2001) were used as a basis for comparison when experimental tissues were examined.

LIF Protein Expression

The tvLIF was produced and purified through genetic engineering using the Glutathione S-transferase (GST) fusion system, which uses pGEX2T plasmid vector for expression in bacterial cells. The fusion protein was purified through its GST tag that bound to matrix, which can be eluted as fusion protein or tvLIF can be released by protease digestion to the supplied buffer. tvLIF was successfully purified in the procedure of monoclonal antibody screening. It was found that the purification of tvLIF polypeptide was problematic because while the method was satisfactory for producing a pure protein, the yield is low and so production was time consuming. For this reason we have decided that fusion proteins (GST fusion protein) would be used in the molecular characterisation and immune trials in conjunction with using GST protein as a control.

Testing possum immune response. Three possums were trapped, housed in a captive colony and after acclimatisation, a test bleed was done to determine basal levels of response. Two possums received 300mg LIF recombinant protein in complete Freund’s adjuvant (CFA). Controls receive saline and adjuvant. Subsequently, possums will receive LIF recombinant protein in Freund’s incomplete adjuvant (IFA) for up to three boosts. Test bleeds will be performed 1 week after boosts. Numbers will be increased when sufficient protein available.

Begin assessment of effect of LIF on reproduction using an in vitro assay system LIF and primordial germ cell (PGC) assay system. The optimal day for collection of PGCs for assessment of the effect of LIF and LIF antibody has been determined. Using the number of germ cells present for each days of pouch life determined by Eckery et al (1996), PGCs were harvested on days 0-2, 7-8, 10-13 and 19-29. Numbers were highest on days 10-13, with significantly less occurring on days 7-8. These results suggest that by days 19-29 PGC were transforming to later stages in germ cell lineage. High numbers can be maintained for 12 days in culture. Possum PGCs collected between days 10-13, have significantly higher proliferation rates when cultured in 500 IU mL^-1 or 100 IU mL^-1 compared to PGCs grown in the absence of LIF. Proliferation increased by 30% by day 8 after initiation and was maintained at that level until day 12. These results suggest that an immune response to LIF may affect proliferation of primordial germ cells. The other LIF assays have been developed and results given previously.

Conclusions

Each of the three targeted proteins has sufficient possum specificity and developmental significant to make them suitable for testing to determine their effect on possum immune response and fertility. Expression and other studies have indicated that their time of action will effectively cover the pre-implantation period and germ cell proliferation. Other targeted proteins also have these characteristics.

PUBLICATIONS

Cui, S., Griffith N., Nanayakkara, K., Cruz, Y. and Selwood, L. (2001). Developmental targets for fertility control in the brushtail possum. National Science Strategy Committee for Possum and Bovine Tb Control Report. Ministry of Agriculture and Forestry, N.Z.

Casey, N., Martinus, R. and Selwood, L. (2002). Outer egg coats of the marsupial conceptus: Secretion and protein composition. Molecular Reproduction Development. 61, 181-194.

Nicholas Casey and Lynne Selwood (2002 submitted). Removal of the shell coat affects maintenance or epithelia in blastocysts of the brushtail possum in vitro.

S.L. Ullman, A.J. Russell, J.I. Mason and L. Selwood (2002 submitted). Species differences in the ovarian distribution of 3b-hydroxysteriod dehydrogenase / isomerase (3bsd) in two marsupials: the Brushtail Possum, Trichosurus vulpecula and the Grey, Short-tailed opossum, Monodelphis domestica.

5.3 PBC 222

Programme Title:	To isolate, characterise and genetically manipulate possum viruses for the biological control of possums
Programme Leader:	Dr Tao Zheng
Institution:	AgResearch Wallaceville

Summary

Thirty-nine possums from Orongorongo, an area previously shown to contain possums sero-positive for anti-MaHV-1 cross-reactive antibodies were sero-negative.

Seventy-five possums from Akatarawa, an area that was not tested previously for anti-MaHV-1 cross-reactive antibodies, were sero-negative.

Culture viruses from faecal samples of 20 possums from the Akatarawa area were performed. No virus was isolated.

SNT was conducted on sera of 28 parma wallabies and 46 dama wallabies collected from Kawau Island. No sero-positive animal was identified.

Culture viruses of nasal/conjunctival swabs and tissue samples from immune compromised 10 parma wallabies and 10 dama wallabies were conducted. No virus was isolated.

One serological suspicious animal was identified from screening 42 dama wallabies from Te Puke for anti-MaHV-1 antibodies. A virus reactivation programme was conducted on this animal. No virus was isolated from this animal.

No sero-positive animal was identified by SNT from 30 Bennett’s wallabies collected from Timaru.

Background

The goal of this programme is to isolate viruses from possums and wallabies and to assess their usefulness for the biological control of possums. Herpesviruses have many desirable characteristics for use as a virus vector in biological control, and technologies have been developed over the past two years which will facilitate the isolation of herpesviruses from possums. In the 2001/02-year, possums will be trapped from areas with a high prevalence of antibodies against herpesvirus. Possums with antibodies to herpesvirus will be treated with corticosteroids to reactivate viral infections and various swabs and tissues from these animals will be cultured for viruses. In the second part of the objective, attempts will be made to isolate viruses from wallabies. Viruses of wallabies pay prove to be useful for the biological control of possums in New Zealand as wallabies are closely related to possums and there is a high chance that viruses isolated from wallabies will also infect possums. Wallabies from Kawau Island and the Canterbury region will be trapped or shot. Serum samples from these animals will be collected and tested for antibodies to herpesviruses. A proportion of the trapped animals will be treated with corticosteoids and subsequently, swab and tissue samples collected for virus isolation.

Approach & Outcomes

Isolation of viruses, especially herpesviruses from possums and wallabies was the primary aim for the programme. Herpesviruses are large viruses that can accommodate foreign genes and are highly transmissible. Many herpesviruses are sexually transmitted which could boost antibody titres at mating and they frequently establish latent infections. These viruses are generally easy to propagate in cell culture, which is important for development of a virus vector. Electron microscopic evidence has identified herpesvirus-like particles in faecal samples of possums from New Zealand. Serological evidence has shown that there were antibodies cross-reactive to antigens of a wallaby herpesvirus in possums in New Zealand, but there were regional differences in the prevalence of these antibodies. Screening for anti-herpesvirus sero-positive possums from different geographic areas would increase the chances of success in the isolation of a possum herpesvirus.

Viruses of wallabies may prove to be useful for the biological control of possums. It is highly likely that viruses isolated from wallabies would also infect possums. A laboratory strain of wallaby herpesvirus, macropodid herpesvirus 1(MaHV-1), was shown to be able to establish a primary infection in possums. A significant difference in DNA sequence of the DNA polymerase gene between this strain and the wild strain was identified, suggesting attenuation of the virus might have happened during the laboratory passage of the virus. Hence, it would be very useful if a fresh MaHV-1 or other types of herpesviruses (MaHV-2) could be isolated from wallabies present in New Zealand. The antigems of the isolated virus could also be used in the immunoassay for screening possum herpesviruses.

Virus culture method is the key methodology for virus identification and isolation from possums and wallabies to meet the essential requirement that any isolated virus would need to be culturable for future use in possum biological control.

Screening for anti-herpesvirus sero-positive possums would increase chances of success in the isolation of possum herpesviruses. An immunological assay for the detection of antibodies to herpesvirus using purified MaHV-1 as antigen was conducted on 39 possum sera collected from Orongorongo in 2001 to assess current antibody prevalence. A previous serological survey in the same area collected in 1994 had identified two antibody positive possums. However, antibody positive samples were not identified from these 39 sera collected in the latest survey. The survey was extended to a new area and 75 possums were trapped from the Akatarawa area. Blood samples were collected and sera were assayed, but no sero-positive possums were identified.

A herpesvirus has been identified by electron microscopy in two of 100 samples of intestinal contents from New Zealand possums. To circumvent the possibility that the immonoassay for herpesvirus antibody using a wallaby herpesvirus antigen may not detect antibody to all possum herpesvirus, culturing faecal samples from sero-negative possums was attempted. Faecal samples were collected from 20 of the Akatarawa possums, two of them had diarrhoea. Faecal extracts were cultured onto monolayers of opossum kidney (OK), potoroo kidney (PTK), possum kidney (PK), and trigeminal (PTG) and dorsal root ganglia (PDRG) for virus isolation. In the second passage of two faecal extracts, one of the three wells of PTG cells (on a 24-well tissue culture plate) showed a couple of cytopathic effect (CPE) foci. However, the cytopathic foci did not appear in the following passages. Some herpesviruses have haemagglutinin and confer the infected cells with haemadsorption effect. Hence, haemadsorption tests were performed with bovine and possum erythrocytes on the different passage levels of the PTG cells, which exhibited suspicious CPE. Haemadsorption was not observed from these cell cultures.

Serological screening for anti-MaHV-l sero-positive wallabies

Wallabies from Kawau Island

Wallabies are other Australian marsupials in New Zealand. Viruses isolated from wallabies may infect possums and play a role in possum biological control. At least 10 strains of MaHV have been isolated from Australian marsupials. Webber & Whalley (1978) reported that 12 of 23 sera collected from parma wallaby (Macropus parma)from Kawau Island, New Zealand had neutralising antibodies against MaHV-1. It is important to determine the current prevalence of anti-MaHV-1 antibodies in wallabies from Kawau Island and other areas in New Zealand. Twenty-eight parma wallaby and 46 dama wallaby (Macropus eugenii) serum samples were collected from Kawau Island, and herpesvirus cross-reactive antibodies were tested by serum neutralising test (SNT) against MaHV-1. Serum was diluted from 1:4 to 1:256 and virus neutralising antibodies were detected by their ability to inhibit cytopathic effect resulting from the addition of 100 TCID₅₀MaHV-1 viruses to OK cells, using a known anti-MaHV-1 positive possum serum as a positive control. No positive serum was identified from these samples. In order to circumvent the possibility that such SNT may not detect all wallaby herpesviruses, 10 parma and 10 dama wallabies from Kawau Island were treated with Dexafort (1 mg/ml dexamethasone sodium phosphate and 2 mg/ml dexamethasone phenylpropionate, Intervet, Castle Hill, Australia) to reactivate possible latent virus infection. The wallabies received Dexafort at 0.2 mg/kg body weight intramuscularly at Days 0 and 7 and were swabbed at Days, 7, 9, 11, 13 and 14. Trigeminal/dorsal root ganglia, lung, liver, spleen, mesenteric lymph nodes, small intestine and its contents were collected for virus isolation when animals were sacrificed. After centrifugation, the supernatant of swabs of nasal/conjunctiva, intestinal content, and 10% tissue preparation were applied in triplicate onto monolayers of OK cells, PTK cells and PK cells in 24-well dishes (100 ml/well). After absorption at 37^oC, 5% CO₂ for 2 hours, the media were removed and topped up with culture medium. Cell cultures were observed daily for any CPE. The cellular monolayers were passaged three times before being deemed negative. No sustained CPE was observed from these cultures. SNT against MaHV-1 was also conducted on sera, prior to and post the administration of the Dexafort. While positive and negative sera worked fine, no positive serum was identified from these wallaby sera. Therefore, the study was expanded to screen wallabies from different geographic areas for anti-MaHV-1 sero-positive animals.

Wallabies from Te Puke

Forty-two dama wallabies (Macropus eugenii) were trapped from Te Puke. Blood samples were obtained from these animals and tested for herpesvirus cross-reactive antibodies by SNT against a wallaby herpesvirus, MaHV-1. Although none of the serum samples completely neutralised the MaHV-1 infection at 1:4 serum dilution, one serum sample did delay the onset of viral CPE at dilutions of 1:4 and 1:8. Therefore, a more sensitive antibody detection method, virus plaque reduction assay was employed for further clarification. The result showed that when serum was diluted at 1:3, it could reduce viral plaques on the monolayers of OK cells by more than 50%, which suggested that a low titre of cross-reactive antibody to MaHV-1 was present in this animal. A virus reactivation programme was conducted on this dama wallaby. When the reactivation programme started, the animal exhibited diarrhoea and lost 1 kg body weight 9 weeks after the first bleeding. Dexafort was administered at 0.2 mg/kg body weight intramuscularly at Days 0 and 8. Nasal and conjunctival swabs and faeces were collected at Days 0, 7, 8, 9 and 10. Blood samples were collected at Days 0 and 10. The animal was euthanized at day 10. Tissues from lung, liver, spleen, kidney, and small intestine, trigeminal and dorsal root ganglia were collected for virus detection. Extracts from swabs, tissues and faeces were subjected to cell cultures for virus isolation. Cell lines of OK, PTK, PK, PTG and PDRG were used for virus isolation. After three passages, no viruses were isolated. A similar virus reactivation programme has been applied in the reactivation of a marsupial herpesvirus from two sero-positive kangaroos in Australia. Herpesvirus was isolated from one of the two sero-positive animals. Searching for more sero-positive animals, especially with high antibody titres, is crucial for the success of virus isolation.

Wallabies from Timaru

Thirty Bennett’s wallabies (Macropus rufogriseus rufogriseus) from Timaru, Canterbury were shot and blood samples were collected for the detection of anti-MaHV-1 antibodies using SNT. No positive serum was identified.

Conclusions

Although no viruses were isolated from possums and wallabies in the past year, the methods used in this study have been successful in the isolation of viruses from marsupials in Australia. It is apparent that to isolate virus from New Zealand possums, a larger population of possums will need to be screened initially, using molecular biology techniques and serology before focussing on small groups of possums for virus isolation.

Publications

Zheng, T., Napier, A.M., O’Keefe, J.S., Buddle, B.M., 2002: Experimental infection of possums with macropodid herpesvirus 1: New Zealand Veterinary Journal (submitted)

REFERENCES

Guliani, S., Smith, G.A., Young, P.L., Mattick, J.S., Mahony, T.J. 1000: Reactivation of a macropodid herpesvirus from eastern grey kangaroo (Macropus giganteus) following corticosteroid treatment. Veterinary Microbiology 68, 59-69.

Rice, M.; Wilks, C.R. 1996; Virus and virus-like particles observed in the intestinal contents of the possum, Trichosurus vulpecula. Archives of Virology 141:945-950

Warburton, B. 1986: Wallabies in New Zealand: history, current status, research, and management need. FRI Bulletin No. 114, Forest Research Institute, Christchurch, New Zealand.

Webber, C.E.; Whalley, J.M. 1978: Widespread occurrence in Australian marsupials of neutralising antibodies to a herpesvirus from a parma wallaby. Australian Journal of Experimental Biological and Medical Science 56: 351-357.

Zheng, T., O’Keefe, J.S., Dickie, A., Tempero, J., Wickstrom, M., Buddle, B.M., 1999: Viruses for use in biocontrol of possums. In: Advances in the biological control of possums. Miscellaneous series 56. The Royal Society of New Zealand, Wellington, Pp 1-3.

5.4 PBC 223

Programme Title:	Control of reproduction in possums by targeting the pituitary gland through the use of GNRH-toxin conjugates
Programme Leader:	Dr Doug Eckery
Institution:	AgResearch

Summary

Results over the last year have strengthened the feasibility of using GnRH-toxin conjugates to control reproduction in possums. The establishment of immortal cell lines expressing GnRH_Rhas enhanced this programme in allowing for more efficient screening of reagents and has significantly reduced the need for animals in this research. A sufficient amount of GnRH-PAP has been obtained for testing in possums and is currently underway. Significant progress has been made in the production of a recombinant GnRH-PAP fusion protein and in the identification of alternative low molecular weight toxins to use in this technology.

Background

A major goal of this programme is to investigate ways in which hormone-toxin conjugates might be used to disrupt reproductive hormone synthesis or function and thereby cause permanent sterility in possums. The advantage of using hormone-toxins as sterilants is that the hormone delivers the toxin directly to specific target cells without relying on the immune system. The strategy is to use very low doses of toxin and to concentrate them only on cells which control reproductive processes. One method being investigated utilises gonadotrophin-releasing hormone (GnRH) as a vehicle to deliver toxins (GnRH-toxins) to gonadotrophe cells in the pituitary gland. Initially, GnRH binds to specific receptors on the surface of gonadotrophe cells and then becomes internalised to initiate the synthesis and release of luteinising hormone (LH) and follicle-stimulating hormone (FSH). IN turn LH and FSH act on the ovary and testis to regulate ovulation and spermatogenesis. Abolition of LH and FSH leads to sterility. The premise for the use of GnRH-toxin conjugates is that after GnRH binds to specific receptors on gonadotrophe cells, the toxin as well as the hormone becomes internalised to cause cell death. The consequence of this treatment is permanent sterility to both male and female possums without compromising the health and well-being of the animals.

We have previously reported that GnRH is indeed a primary regulatory of FSH and LH secretion in possums and that the actions of GnRH are exclusive to the pituitary gland via specific high affinity receptors. Moreover we have shown that GnRH is internalised rapidly into cells after binding to its receptor and that it is possible to introduce toxins into a cells using GnRH to cause cell death. Work in this proposal is aimed at testing the effects of GnRH toxins on reproductive function in possums, producing a recombinant GnRH-toxin and identifying alternative toxins for use in this technology.

Objective 1 – Production And Testing Of Gnrh-Pap

The aims of this objective are to:

produce a large amount of GnRH conjugated to pokeweed antiviral protein (PAP; GnRH-PAP) for testing in possums,
establish three immortal cell lines expressing GnRH receptor (GnRH_R),
establish protocols for screening GnRH-toxins conjugates for the ability to bind to GnRH_R and specifically kill cells, \determine the in vivo ED₅₀for a GnRH analogue in possums, and
begin treating possums with GnRH-PAP to determine the effects on gonadotrophin concentrations in blood and reproductive activity.

One of our technicians, Ms. Evelyn Bauer, visited the lab of Professor Terry Nett, Colorado State University, with the purpose of making a sufficient amount of GnRH-PAP to use in our programme and to learn several new laboratory techniques. These techniques included the purification of PAP from pokeweed leaves, conjugation of PAP to GnRH, separation of GnRH-PAP from free GnRH and PAP, assessing the bindability and cytotoxicity of GnRH-PAP, and measurement of intact GnRH-PAP by ELISA. Batches of GnRH-PAP (75mg) and PAP (87mg) were prepared and brought back to our lab and are currently being used in our programme. The techniques listed above have been used to purify PAP from the NZ variety of pokeweed, commonly known as inkweed. Two immortal cell lines expressing GnRH_Rhave been established in our lab, namely aT3 and CHO-GnRH_R-GFP. We hope to have a third cell line, LaT2 cells, established in the near future. Each of these cell lines has unique features and different requirements for their maintenance. Using these cells, protocols have been established for testing the binding and cytotoxicity of GnRH-toxins. For GnRH-PAP, several conditions, including dose, exposure time and time to recover following treatment have been tested (Table 1). An important result from these experiments has shown that cells

must be exposed for at least 8 hours to GnRH-PAP for it to be effective in killing cells. In addition, we have shown that pre-treatment of GnRH-PAP with possum serum does not influence its ability to kill cells. The immortal cell lines established in our lab have become valuable tools for the testing of GnRH-toxins and have dramatically reduced the number of possums required for this research.

Exposure Time (h)	Recovery Time (h)	Dose ng/ml)
1	72	1, 10, 100
2	72	1, 10, 100
4	24,48,72	1, 10, 100
8	24,48	1, 10, 100
24	24,48,72,96	1, 10, 100, 1000
48	24,48,72	1, 10, 100, 1000
72	24,48,72	1, 10, 100, 1000

Table 1. Treatment conditions of immortal cell lines with GnRH-PAP

The GnRH analogue used in the GnRH-PAP conjugate is D-Lys⁶GnRH. To help determine the appropriate dose of GnRH-PAP to use to treat possums, the ED₅₀for D-Lys⁶GnRH was determined in male and female possums. Overall, the calculated ED₅₀was 0.02ug, however, there was a difference between males and females being 0.007ug and 0.09ug, respectively. Using this information, combined with in vitro results (e.g. minimum 8 hour exposure time) and advice from Prof. Nett an experiment was designed to determine the effects of GnRH-PAP on intact male and female possums during the breeding season. That experiment is currently underway.

Objective 2 – Recombinant GnRH-PAP

In a recent paper, a group in France reported the production of GnRH-PAP as a recombinant fusion protein that was able to specifically kill cells expressing GnRH_R. These results present the possibility of having an unlimited source of pure GnRH-PAP and this use of recombinant technology may lead to new ways of delivery hormone-toxin conjugates (e.g. through engineering of possum-specific nematodes). The aim of this objective is to produce biologically active recombinant GnRH-PAP.

Pokeweed antiviral protein is a member of a family of proteins called ribosome inhibiting proteins. Members of this protein family, in particular PAP, have been the focus of many studies investigating their use as therapies for cancer, the treatment of AIDS and the control of fertility. Both PAP and its isoform PAP II can inhibit ribosomal function in eukaryotic and prokaryotic cells and cause cell death. The PAP used most commonly is purified from the pokeweed plant, Phytolocca americana, which is found in many part of North America. A related species of the plant, Phytolocca octandra (commonly known as inkweed) is found in NZ. We have successfully cloned the genes for PAP and PAP II from the NZ species, as confirmed by sequence analysis. A comparison of the predicted amino acid sequences for PAP and PAP II between the two species of pokeweed plants showed 94% and 91% identity, respectively. The high percentage of identity suggests that PAP and PAP II from the NZ pokeweed species will have similar biological activity to the North American species. The gene for PAP (NZ species) has been transferred into E.coli and used to produce recombinant PAP. This recombinant product has been shown to be the correct size and crossreact with antibodies generated against native PAP by Western blot analysis. An assay to test the biological activity of the various forms of PAP is currently being established in our lab. Moreover, expression plasmids have been constructed for GnRH-PAP fusion proteins and are currently being used to transform cells for the production of recombinant GnRH-PAP.

Objective 3 – Alternative toxins

The toxin currently being used, PAP, is highly efficient, but also relatively large protein that may be difficult to deliver in a bait. Our aim in this objective is to find new toxins that are small molecular weight and suitable for conjugation to GnRH.

In collaboration with colleagues from AgResearch, Ruakura, we have begun testing four different quinone derivatives as potential toxins. An initial screening of the general toxicity of the quinones on both possum cells (pituitary, kidney, liver, lung) in primary culture and immortal cell lines (CHO, aT3, CHO-GnRH_R-GFP) showed that the toxicity varied between the different cell types. The cells least affected by the quinones were possum kidney cells, whereas those most sensitive were aT3 cells. Possum pituitary cells were similar in sensitivity to kidney cells. In general, the quinones showed less toxicity than PAP. In order to simplify the conjugation of the quinones to GnRH, a unique GnRH analogue was synthesized, namely D-Cys⁶ GnRH. This analogue is not routinely available and thus little information was available on its biological activity and therefore needed to be tested. Fortunately, this analogue showed good biological activity (binding and ability to stimulate LH secretion) both in vitro and in vivo. Of the four quinones tested, one was chosen for conjugation to D-Cys⁶ GnRH. Because of the costs involved in using D-Cys⁶GnRH for the initial synthesis, glutathione was used as a model compound. Following development of an efficient process, successful conjugation of D-Cys⁶GnRH to a quinone derivative was achieved. Using our immortal cell lines, we have begun testing the ability of this new GnRH-toxin conjugate to specifically kill cells expressing GnRH_R. In CHO-GnRH_R-GFP cells, this conjugate showed little effect after a 24 hour exposure with doses up to 1000ng/ml. In contrast, doses of 100ng/ml and 1000ng/ml were able to kill about 50% of aT3 cells after 24 hours exposure. The same doses of quinone or GnRH alone showed no effects on these cells. Moreover, this new conjugate showed no detrimental effects on a control cell line not expressing GnRH_RThese favourable results provide evidence that quinone derivatives may be suitable as low molecular weight toxins for use in this technology.

Approach & Outcomes

Milestone Achieved

Objective 2 –Recombinant GnRH-PAP

Produce recombinant GnRH-PAP

Expression plasmids have been constructed for GnRH-PAP fusion proteins and are currently being used to transform cells for the production of recombinant GnRH-PAP, however no recombinant protein has been produced thus far.

Objective 3 – Alternative toxins

Cytotoxicity of quinone derivatives in possum cells and immortal cell lines

Following on from the work done on the toxicity of 4 different quinone derivatives on CHO cells, similar data was obtained for different types of possum cells in primary culture (pituitary, lung, kidney and liver) and two additional immortal cell lines each expressing GnRH receptor (aT3 and CHO-GnRH_R-GFP). A range of concentrations from 1000ng/ml to 2x10⁶ng/ml was tested. Results are shown in Table. The cells that were least affected by the quinones were possum kidney cells, whereas the most sensitive were the aT3 cells. Possum pituitary cells were similar in sensitivity to kidney cells.

Table 1 Concentration at which quinones become non-toxic. Average of 3 trials

Cell type	5-hydroxy-1,4-naphthoquinone	5-hydroxy-2-methyl-1,4-naphthoquinone	2-chloro-1,4-naphthoquinone	2-chloromethyl-1,4-naphthoquinone	PAP
Lung	6700ng/ml	2600ng/ml	6700ng/ml	1500ng/ml	300ng/ml
Liver	600ng/ml	600ng/ml	5000ng/ml	2500ng/ml	5000ng/ml
Kidney	10,000ng/ml	1000ng/ml	10,000ng/ml	5000ng/ml	3700ng/ml
Pituitary	10,000ng/ml	3700ng/ml	10,000ng/ml	2600ng/ml	1300ng/ml
CHO-GnRH_R-GFP	1000ng/ml	4000ng/ml	<1000ng/ml	2000ng/ml	<1000ng/ml
aT3	Still toxic	<1000ng/ml	Still toxic	Still toxic	<1000ng/ml

Cytotoxicity of a GnRH-quinone conjugate in immortal cell lines

One of the four quinones tested (5-hydroxy-2-methyl-1, 4-naphthoquinone) was chosen for conjugation to D-Cys⁶GnRH. Using our immortal cell lines expressing GnRH receptor, we have begun to test the ability of this new conjugate to specifically kill these cells. In CHO-GnRHR-GFP cells, the new conjugate showed little effect after 24 hour exposure with doses up to 1000ng/ml. In contract, doses of 100ng/ml and 1000ng/ml were able to kill approximately 50% of aT3 cells after a 24 hour exposure. The same doses of quinone or GnRH alone showed no effects on these cells. Moreover, this new conjugate showed no detrimental effects on a control cell line not expressing GnRH receptor. These favourable results provide evidence that quinone derivatives may be suitable as low molecular weight toxins for use in this technology.

Publications

Eckery, D.C., Whale, L.J., Lawrence, S.B., Wylde, K.A., McNatty, K.P., Juengel, J.L. 2002 Expression of mRNA encoding growth differentiation factor 9 and bone morphogenetic protein 15 during follicular formation and growth in a marsupial, the brushtail possum (Trichosurus vulpecula). Molecular and Cellular Endocrinology 192: 155-126.

Eckery, D.C. The brushtail possum: fertility control of an introduced species. 2001. Invited seminar. Smithsonian Institute, Conservation Research Center, Front Royal, USA

Whale, L.J., Eckery, D.C., Juengel, J.L. 2002. Localisation of mRNA encoding proteins important for steroidogenesis in the brushtail possum ovary. Annual Conference of the New Zealand Society of Endocrinology, Christchurch.

Recent publications from work partially funded by MAF Policy

Eckery, D.C., Lawrence, S.B., Juengel, J.L., Greenwood, P., McNatty, K.P., Fidler, A.E. 12001 Gene expression of the tyrosine kinase receptor c-kit during ovarian development in the brushtail possum (Trichosurus vulpecula). Biology of Reproduction 66:346-353.

Eckery, D.C., Lun, S., Thomson, B., Juengel, J.L., 2002 Ovarian expression of mRNA encoding the receptors for LH and FSH in a marsupial, the brushtail possum (Trichosurus vulpecula). Biology of Reproduction Biology of Reproduction 66:1310-1317.

Eckery, D.C., Juengel, J.L., Whale, L.J., Thomson, B.P., Lun, S., McNatty, K.P. 2002. The corpus luteum and interstitial tissue in a marsupial, the brushtail possum (Trichosurus vulpecula). Molecular and Cellular Endocrinology 191:81-87.

5.5 PBC 224

Programme Title:	Control of reproduction in possums by targeting the oocyte-specific growth factors, GDF-9 and GDF-9B
Programme Leader:	Dr Doug Eckery
Institution:	AgResearch

Summary

Until recently, the oocyte (egg) has been viewed as a passive cell that is nurtured and directed by neighbouring cells. Recent discoveries have shown that the oocyte is instead more like the ‘control centre’ of the ovarian follicle; being very active in directing the growth and development of the follicle. Function of the oocyte is dependent, in large part, on local growth factors that have paracrine and/or autocrine actions. Growth and differentiation factor 9(GDF-9) and GDF-9B are two such growth factors that have been identified. The gene for GDF-9B was recently found to be responsible for mutant phenotypes found in sheep. This discovery has been of great interest to the international science community and has the potential to lead to the development of new treatments for enhancing and blocking fertility. To date, we have cloned and sequenced cDNAs for both of these growth factors in possums and have determined the ontogeny of their expression during development of the ovary in pouch young and at the various stages of follicular growth. As in other species, both GDF-9 and –9B are oocyte-specific, but of particular interest was the observations that GDF-9B is expressed at an earlier stage of follicular development than in all other species reported to date. This earlier expression represents the possibility of targeting the pool of non-growing follicles within the ovaries, or in other words, all the eggs within the ovaries. Thus, it may be possible to develop a permanent form of fertility control that is species-specific to possums. The aim of this work is to determine the effects of immunising possums against GDF-9 and GDF-9B on ovarian function and reproductive activity.

Approach & Outcomes

OBJECTIVE 1 – IMMUNISATION OF POSSUMS AGAINST GDF-9 AND GDF-9B

Work in our group has identified regions of the GDF-9 and –9B proteins that are non-conserved between species. Immunisation of sheep against these peptides had significant consequences on ovarian function. Peptides representing these non-conserved regions in possum GDF-9 and 9B were synthesised and conjugated to KLH in preparation for immunising possums.

Three breeding groups of possums, each consisting of seven females and 2 males, were established. Animals in each breeding group were immunised against GDF-9, GDF-9B or KLH alone. Initial immunisations were given with Freund’s complete adjuvant and an additional five immunisations (using an incomplete adjuvant) were given at three week intervals. Immunisations were started so that animals would have received two boosters by the start of the breeding season (mid-March). Before each immunisation was given, a blood sample was taken to determine antibody titres. Reproductive activity of the female possums was monitored by vaginal cytology. In addition, pouches were checked at regular intervals for the presence of pouch young.

Antibody titres were elevated in all groups against the respective peptides and KLH. However, as shown in Table 1, there was no apparent effect of immunisation on reproductive activity. Because no effects were observed on fertility, ovarian morphology was assumed to be normal and was not further examined. We concluded that the non-conserved regions of GDF-9 and -–9B are not important to the function of these proteins in possums. Future work will involve immunising possums against the entire mature regions of GDF-9 and GDF-9B.

Table 1. Effects of immunisation of possums against GDF-9 and GDF-9B on reproductive activity.

	Control (KLH)	GDF-9	GDF-9B
Cycles per possum	1.8	2.1	2.1
Offspring per possum	0.5	0.8	1.0
Offspring per cycle	0.3	0.4	0.5
% possums cycled	100	100	100
% possum produced offspring	50	70	86

PUBLICATIONS

Eckery, D.C., Lawrence, S.B., Juengel, J.L., Greenwood, P., McNatty, K.P., Fidler, A.E. 2002 Gene Expression of the tyrosine kinase receptor c-kit during ovarian development in the brushtail possum (Trichosurus vulpecula). Biology of Reproduction 66: 346-353.

Eckery, D.C., Lun, S., Thompson, B., Juengel, J.L., 2002. Ovarian expression of mRNA encoding the receptors for LH and FSH in a marsupial, the brushtail possum (Trichosurus vulpecula). Biology of Reproduction Biology of Reproduction 66: 1310-1317.

Eckery, D.C., Juengel, J.L., Whale, L.J., Thomson, B.P., Lun, S., McNatty, K.P., 2002. The corpus luteum and interstitial tissue in a marsupial, the brushtail possum (Trichosurus vulpecula). Molecular and Cellular Endocrinology 191: 81-87.

Conferences/Seminars

Eckery, D.C., Juengel, J.L., Whale, L.J., Thomson, B.P., Lun, S., McNatty, K.P. 2001. The ovarian interstitium and luteal function in a marsupial. Regulation of Ovarian Function Workshop, Upper Hutt.

Eckery, D.C. The brushtail possum: fertility control of an introduced species 2001. Invited seminar. Smithsonian Institute, Conservation Research Center, Front Royal, USA.

Whale, L.J., Eckery, D.C., Juengel, J.L. 2002. Localisation of mRNA encoding proteins important for steroidogenesis in the brushtail possum ovary. Annual Conference of the New Zealand Society of Endocrinology, Christchurch.

5.6 PBC 225

Programme Title:	Controlled oestrus and synchronised breeding in possums to facilitate the testing of the new biocontrol agents
Programme Leader:	Dr Bernie McLeod
Institution:	AgResearch

Summary

Over several years of investigation, we have shown that a variety of proven methods of synchronising oestrus and ovulation in farm animals, fail to work in possums. The oestrogen/progesterone regimen that we use to control preovulatory follicle development, consistently results in the synchronised emergence of a single large follicle in 75-80% of treated possums. This rate of success would be acceptable in an oestrus/ovulation synchronisation programme, except that we are unable to synchronise the ovulation of these preovulatory follicles. The fact that LH, GnRH, oestradiol and hCG all fail to induce ovulation of large follicles in possums, indicates that there are significant differences in ovarian function in this species. This programme is identifying major differences in gene expression in developing follicles in possums. We continue to seek methods of inducing ovulation of preovulatory follicles in possums.

Objective 1: Hormonal control of follicle development

We have shown unequivocally that expression of mRNA for prolactin receptors is extensive throughout the possum ovary. The expression pattern indicates that prolactin has a role in steroidogenesis in possum ovaries and perhaps also in the selection and development of the preovulatory follicle. Prolactin also appears to regulate luteal function and may in part, explain the extraordinary high levels of progesterone secreted by corpora lutea in possums. We have been unable to demonstrate a direct role of prolactin in controlling ovarian cyclicity.

Objective 2: Synchronised ovulation and mating

We have confirmed the LH and hCG fail to induce ovulation of preovulatory follicles in possums, a feature that may be unique to marsupials. Despite its obviously important role in preovulatory follicle development, acute treatment with prolactin does not promote ovulation of preovulatory follicles in possums..

Background

The development of potential biocontrol agents will require the ability to synchronise oestrus and mating in possums. Standard methods used for oestrus synchronisation in farm animals simply do not work in possums. In part, this is due to the long follicular phase compared with that in farm animals, but it is becoming increasingly evident that there are fundamental differences in the hormonal control of follicle development and ovulation in possums. For example, it has been shown that high numbers of prolactin receptors are present in the ovaries of possums, a feature not seen in other animals.

Recently, we have taken the approach that there need to be two parts to a hormonal strategy for synchronising ovulation in possums. The first is to synchronise the emergence of preovulatory follicles within a group of animals and the second, to synchronously induce the rupture (ovulation) of those follicles. Our most successful hormonal treatment to date results in the presence of a presumptive preovulatory follicle in 75-80% of treated animals on a given day. The objective of this programme is to assess the viability of these induced preovulatory follicles and to develop a protocol that will consistently induce ovulation of the induced follicles.

Objective 1 investigates aspects of follicle development and compares spontaneously developing antral follicles with those in possums in which follicle development has been induced.

Objective 2 assess methods of synchronising the ovulation of these induced follicles.

One milestone for this programme (To assess the competency of corpora lutea resulting from induced preovulatory follicles) was replaced midway through the year. This milestone was dependent on the success of the hormonal treatments being used, to induce ovulation. As all of these treatments failed to induce ovulation, it was not possible to proceed. The replacement milestone investigated a possible role of prolactin on reproductive activity in possums.

Approach & Outcomes

Objectives 1

Hormonal control of follicle development

Description: We have developed a steroid-hormone regimen that will synchronise the emergence of a preovulatory in a high percentage of possums. In this objective the viability of these follicles (their ability to ovulate and be fertilised) will be assessed. The expression of mRNA for pituitary hormone receptors in both induced and spontaneously developing preovulatory follicles will be determined, to gain insight into ovulatory mechanisms in the possum.

Investigations of antral follicle development in possums in an allied FRST-funded programme (Genetic and Hormonal Control of Ovulation, AgResearch Wallaceville), had earlier identified some notable differences in possum follicles, especially in the expression of mRNA for receptors of the pituitary hormones LH and prolactin. Of particular interest, was that there was strong expression for prolactin receptors in the ovary, suggesting a possible unique role for this hormone in antral follicle development in the possum.

A series of experiments was undertaken to determine mRNA expression for receptors of the gonadotrophic hormones in growing follicles, in preovulatory follicles and in recently formed corpora lutea. This included spontaneously developing follicles recovered from untreated possums selected at random during the breeding season. Gene expression in these follicles was compared with that in preovulatory follicles collected following removal of pouch young and in follicles induced to grow following steroid hormone treatment.

The effect that prolactin may have on reproduction in possums was also assessed by treating possums with exogenous prolactin in Spring, a time when approximate 50% of female possums spontaneously show oestrous cycles and 50% are anoestrous.

Achievement of Milestones

Does Prolactin have either stimulatory or inhibitory actions on oestrus cyclicity?

This milestone replaces the original milestone (To assess the competency of corpora lutea resulting from induced preovulatory follicles) which could not proceed as treatments given to induce ovulation of preovulatory follicles have failed to do so.

The presence of high numbers of PRL receptors in ovarian tissues in possums raises the question of what action this hormone has in ovarian function in this species. For example, does PRL have a pivotal role in seasonality? To address this question we treated possums with prolactin in spring. In coastal Otago, we have shown that about 50% of animals show oestrous cyclicity at this time of year and about 50% are anoestrous. The aim was to determine whether prolactin treatment would increase (or decrease) the percentage of animals showing oestrous cyclicity.

A total of 40 adult female possums underwent the standard oestradiol/progesterone treatment (single injection of oestradiol following by progesterone implants in situ for 12 days). After removal of the progesterone implants, half of the animals were given prolactin and half were given saline, delivered by osmotic minipumps for a period of 6 days. All animals were then euthanased and their reproductive tracts removed.

There were no significant differences in reproductive tract weights (cul-de-sac, uteri or ovaries) or in preovulatory follicle development between groups. Fewer than expected –saline-treated animals (20% rather than 50%) had large presumptive preovulatory follicles present at the time of tissue collection. Nevertheless, prolactin treatment did not increase the number of animals showing oestrous cyclicity. The dose of source (ovine) of prolactin may have not been appropriate or the duration of treatment may have been too short. The mRNA investigations (see next milestone) clearly indicate a role for prolactin in ovarian function in possums, but how this influences reproduction in possums remains to be resolved.

Compare mRNA for LH and prolactin receptors between induced and spontaneously developing preovulatory follicles.

The objective of this study is to monitor mRNA expression for LH and prolactin receptors in spontaneously growing follicles in possum, to try to identify the role that these hormones play in preovulatory follicle development. Comparison was made with follicles in which development had been synchronised by treatment with our oestradiol/progesterone regimen and in possums that had been treated with exogenous prolactin. Ovarian tissue was collected from the following three treatment groups (n=8-10/group).

Non-pregnant, non-lactating females were randomly selected from the colony during the breeding season. Ovaries from these animals would contain spontaneously developing follicles at different stages of growth, including preovulatory follicles, and corpora lutea.
Lactating animals in which follicle development had been ‘synchronised’ by removal of their pouch young. These animals underwent repeated laparoscopy and ovaries were collected when a large (6mm) presumptive preovulatory follicle was present.
Non-lactating animals in which follicle development had been ‘synchronised’ by our steroid hormone treatment regimen. These animals underwent repeated laparoscopy and ovaries were collected when a large (6mm) presumptive preovulatory follicle was present.
Non lactating possums treated exactly as the animals in Group 3, except that they were given injections of exogenous prolactin (30mg ovine prolactin/injection) twice daily for three days immediately prior to recovery of the ovaries.

To date, mRNA expression has been completed for possums in the first two groups in which follicles were developing spontaneously. Several cell types were observed to express prolactin receptors (Prl-R) in ovaries of brushtail possums. These included; granulosa cells, thecal cells, interstitial tissue and luteal cells. Expression was variable in granulosa cells and based on morphological criteria, it could not be predicted which follicles would or would not express Prl-R. In contrast, expression was very consistent in thecal cells of growing antral follicles. Both granulosa and thecal cells of preovulatory-sized follicles expressed Prl-R, suggesting that once the dominant follicle is selected, prolactin is involved in its regulation. Expression was also observed just after ovulation (1-3 days) in the forming corpus luteum. The corpus luteum continued to express Prl-R mRNA during the mid-luteal phase when maximum amounts of progesterone are secreted. In addition, regressing corpora lutea expressed Prl-R mRNA. The interstitial tissue expressed Prl-R during all stages of the oestrous cycle.

Overall, the expression pattern of Prl-R mRNA supports a role for prolactin in steroid production by the possum ovary. The expression of Prl-R in preovulatory follicles supports a role for prolactin in the selection and ovulation of a single follicle at each cycle. In addition, prolactin may be regulating both luteal and interstitial cell function.

The in situ hybridisation protocol for determining the cellular localisation of mRNA’s involves an extended period of exposure (of several weeks) after tissue sections have been incubated with the RNA probes. The in situ hybridisation protocol for LH receptors in spontaneously growing follicles is in progress. These tissues require an even longer period of exposure before they can be developed and stained, but the results will be available in the next few weeks.

Objective 2

Synchronised ovulation and mating

Description: Hormonal treatments given to mimic the preovulatory LH surge fail to induce ovulation of large presumptive preovulatory follicles in possums. In this objective we will determine whether there are hormonal components other than the preovulatory LH surge, that are required for ovulation to occur. Further, being able to successfully synchronise the time of ovulation is not the endpoint to this objective. The synchronised oestrus must be accompanied by expression of oestrus and fertile mating.

In spontaneously ovulating mammals, the rupture (ovulation) of preovulatory follicles is triggered by a surge release of luteinising hormone (LH). In most species, a single injection of LH alone can artificially induce ovulation of the preovulatory follicle. Alternative treatments to the use of LH, is to administer either the ovarian steroid oestradiol or the hypothalamic releasing hormone GnRH, both which will evoke an endogenous LH surge. In addition, human chorionic gonadotrophin (hCG), which has potent LH-like properties, is frequently used to induce ovulation. To our knowledge, all of these treatments will successfully induce ovulation of the preovulatory follicle in all of the species studied in which they have been used.

We have shown that, with few exceptions, LH, oestradiol, GnRH and hCG all fail to induce ovulation of large follicles in possums. This was a surprising finding as such hormonal treatments consistently induce ovulation in eutherian species. The exceptions are that massive (supra-pharmacological) doses of porcine LH (4mg/injection) or a series of injections of GnRH given over a 2-day period will induce ovulation in superovulated possums. However, these treatments are not practicable in terms of costs and time and they do not relate to normal physiological conditions. The first milestone in the objective was to confirm that LH (or hCG) fails to induce ovulation of spontaneously developing preovulatory follicles in possums, by restricting the treatment to only those possums that had a large macroscopically normal follicle present at the time of treatment. Again, these treatments failed to induce ovulation. This strongly suggests that hormonal mechanisms other than those known to be involved in eutherian mammals, are important in maturation of preovulatory follicles and ovulation in possums. As outlined earlier, we have shown that steroidogenesis within the preovulatory follicle in possums coincides precisely with the expression of prolactin receptors in the follicle. There is increasing evidence that prolactin is critical to follicle development in possums and its role is being investigated in Objective 1. Another unknown yet to be answered is whether prolactin is directly involved in inducing ovulation in possums.

Achievement of Milestones

Determine the incidence of ovulation of induced preovulatory follicles in possums with or without LH treatment

On two separate occasions, groups of 40 and 32 female possums were subjected to the oestradiol/progesterone regimen during the breeding season, to control antral follicle development. All animals underwent laparoscopy 8 days after the removal of progesterone implants to determine whether a large presumptive preovulatory follicle was present. A total of 32/40 (80% and 24/32 (75%) of the possums had a large (approximately 6mm diameter) follicle present at that time. These animals were randomly assigned to two treatment groups and were given a single intramuscular injection of either saline (0.5 ml) or of ovine LH (50mg in 0.5 ml saline, Experiment 1) or of hCG (500IU in 0.5 ml saline). All animals were euthanased 48h later to determine the incidence of ovulation.

Neither ovine LH nor hCG increased the incidence of ovulation with 4/12 saline-treated and 6/12 LH-treated (Experiment 1) and 4/16 saline-treated and 5/16 hCG-treated (Experiment 2) possums ovulating. These treatments confirmed that LH alone fails to induce ovulation of large antral follicles in possums. It should be noted that, on a bodyweight basis, the dose of hCG administered to possums was approximately 25 times greater than the dose routinely used for sheep.

Determine whether treatment with exogenous prolactin influences ovulation in possums

To determine whether prolactin is directly involved in the induction of ovulation in possums, we treated a total of 30 animals with our oestradiol/progesterone regimen to induce preovulatory development and examined their ovaries by laparoscopy 7 days after progesterone implant removal, to confirm that a single large (5-6mm_ follicle was present. At this time, that the ‘ovulatory’ hormonal treatments were administered. These involved either a single injection of saline (controls), of hCG (500IU) or of hCG (500IU) plus prolactin (60mg). The incidence of ovulation in response to these treatments was determined when the animals were euthanased 48h later.

Neither hCG alone nor hCG plus prolactin significantly increased the incidence of ovulation with 2/10, 3/10 and 5/10 animals ovulating in response to treatment with saline, hCG and hCG plus prolactin, respectively. Although acute treatment with prolactin did not increase the incidence of ovulation in possums, it does not exclude a possible direct action for this hormone in the ovulatory process. For example, the timing, dose of source of prolactin (we used ovine, not possum prolactin) may have been inappropriate.

5.7 PBC 226

Programme Title:	Oral delivery of Bioactives to Possums
Programme Leader:	Dr Bernie McLeod
Institution:	AgResearch

Summary

There was a significant delay in getting this programme underway and Milestones set for the first 6 months were not met on time. The major hold-up was that sub-contract agreements were delayed by several months, and staff recruitment to this project was not possible until this had been resolved. However, a junior research fellow was appointed specifically for the formulation component of the programme several months ago. As a consequence, the programme is now on track and the development of formulation strategies for oral delivery to possums is rapidly gaining momentum. The initial formulations produced have pinpointed some critical areas when designing formulations for possums and a review of currently available polymers and their interactions with formulation excipients that will meet these criteria is currently underway. In addition, the process for developing and testing these formulations is now much more efficient and streamlined.

Overall, as we gather more information on transit through the possum gastrointestinal tract and on the degradation of polymers within different regions of the intestine, real advances are being made in developing formulations capable of surviving transit through the stomach and small intestine. Further development to ensure that the formulations will discharge their bioactive payload in the hindgut is underway. Once this is achieved, the incorporation ofmetabolic inhibitors to protect the bioactive in the hindgut and permeation enhancers to promote its uptake (based on our earlier studies) will also be incorporated in formulations.

These formulations will then be tested for their ability to deliver model peptides to the possum hindgut.

Objective 1 Formulation for delivery to hindgut

We have identified a number of polymers that are extremely effective in protecting tableted formulations from breakdown in the stomach and small intestine of the possum, regions where a biocontrol agent would be very susceptible to rapid degradation. We are now concentrating on selecting different polymers or formulation strategies that will promote rapid breakdown within the possum hindgut. One concern is that the plasticizer we used may have anti-microbal actions that would have prevented degradation of the polymer, so alternatives are being investigated. A second approach is to use cellulose as the polymer as this is degraded by hindgut microflora. The film-forming properties of cellulose are currently being determined. We are also investigating the use of pectin as the polymer as this has been shown to be degraded by microflora in the colon of rats.

The in vitro disintegration tests we have established, appear to have good correlation with degradation within the possum intestine.

Objective 2 Targeting specific regions of gut

We have established a radio-labelling method for determining transit and residence times of solutions and particulates in the possum gastrointestinal tract. This has already identified differences in transit times between particles and solutions and investigation of transit and residence times of a range of particle sizes is underway. The CT scanning of tablets after oral delivery is providing additional information of intestinal transit. It also provides accurate information on the time and location of tablet disintegration.

Background

Previously we identified the luminal and mucosal enzymes that metabolise peptides and proteins in the possum gastrointestinal tract, and determined the relative activity of these enzymes in different regions of the gut. This demonstrated that the colon and caecum have low levels of both luminal and mucosal enzyme activities (up to 1000 times lower than in the small intestine), identifying it as the most appropriate region to target for oral delivery of peptides and proteins. In addition, we identified a number of inhibitors that protect peptides and proteins from enzymatic degradation in the gut. Further, we have studied the permeation of a hydrophilic marker through the intestinal wall in various regions of the intestine to show that permeation rates can be enhanced through the judicious use of permeation enhancers. We have investigated the relationship between permeation enhancement and damage to the mucosa, in order to identify useful permeation enhancers and to gauge what concentrations might be used.

The object of the current programme was to develop formulation strategies to target delivery to the hindgut. This entails protecting the bioactive during its passage through the stomach and small intestine and promoting its release in the caecum and colon. To this end tableted formulations have been produced, using selected polymer coatings that will protect from degradation in the stomach and small intestine. These coatings are designed to disintegrate in the hindgut. Degradation of the tableted formulations in the hindgut is being assessed both in vitro and in vivo.

The successful delivery of these formulations is dependent on an understanding of gastrointestinal transit times in possums and on how particle size influences the rate of passage through, and selection retention in various regions, of the possum intestine. We have conducted some preliminary research to investigate gastric residence times in the possum, but more detailed information is required to design delivery systems that target specific regions of the gut.

Approach & Outcomes

Objective 1

Formulation for delivery to hindgut

We have reviewed the literature on current polymer technology to identify products suitable for coating pelleted forumations for delivery to the hindgut. The aim of this objective is to use selected polymers to coat pellets and determine the stability of these polymer-coated pellets in luminal contents of the stomach, duodenum, jejunum and ileum of the possum and their disintegration in lumen contents of the caecum and colon. A second purpose is to screen putative biocontrol agents, using technologies we have developed for evaluating metabolism and absorption in the possum gastrointestinal tract.

The literature on current polymer technology was reviewed to identify products suitable for coating pelleted formulations for possums and polymers known to selectively release in the colon (pectin, chitosan and ethyl cellulose) were chosen. Tablets containing barium sulphate and a water-soluble dye were prepared, either as matrix tablets in which the polymer is incorporated throughout the tablet or as polymer-coated tablets. The range of formulations made, included some with or without inclusion of a disintegrant, which would induce rapid breakdown of the tablet once the integrity of the polymer coating was disrupted. Coating of tablets was undertaken by pan-coating to a range of coat thicknesses. Coat integrity was then assessed by standard dissolution testing in pH 7 buffer, in acid or in alkali, by incubation in luminal contents from possum intestine or by animal studies.

Achievement of Milestones

Prepare a range of polymer-coated pellets containing barium sulphate and a water-soluble dye, including different polymers and different coating thicknesses.

Matrix tablets (average weight 120-150mg) were prepared with barium sulphate and ethyl cellulose (polymer) by dry mixing with methanol.

Tablets (6 mm diameter) containing barium sulphate and a water-soluble fluorescent dye, as well as typical tableting excipients (primogel, magnesium stearate, aerosol) were prepared by dry granulation and compression and their average weight and hardness determined. A plasticizer was included within these formulations to enhance the formation of a uniform polymer layer during coating. These tablets were then coated to different thicknesses (single, double or triple-coating) with ethyl cellulose by pan-coating. Single-coated tablets were processed for 30 cycles, with double and triple-coated tablets undergoing a further 15 and 27 cycles, respectively. The polymer coatings increased the average weight of the tablets by 2.7, 6.7 and 11% for single, double and triple coated tablets, respectively.

Batches of each of the tablet formulations (matrix, uncoated tablet, tablets with single, double and triple polymer coatings) were tested for their disintegration rates by incubation for 2h at 37^oC in phosphate buffer at pH 6.8, and in 0.1M hydrochloric acid. Tablet disintegration is summarised in Table 1. None of the matrix tablets disintegrated, even after 48h incubation at 37^oC. These tablets were then assessed in vivo and there was no evidence of tablet disintegration over a 24h period of monitoring by CT scan. Consequently, the use of matrix tablets was discontinued in favour of polymer-coated tablets. As a proportion of single-coated tablets had disintegrated during incubation in buffer, these formulations were further tested in luminal contents and in animal studies (see below). All uncoated tablets disintegrated completely within 1 min when incubated in distilled water.

Table 1. Disintegration of matrix, uncoated and coated tablets.

	Buffer	0.1M HC1
Matrix tablets	No disintegration*	No Disintegration*

Uncoated	Complete disintegration	Complete disintegration
Single coat	Partial disintegration (2/6 tablets)	No disintegration Slight discolouration
Double coat	No disintegration No discolouration	No disintegration Slight discolouration
Triple coat	No disintegration No discolouration	No disintegration No discolouration

No disintegration evident after 48h of incubation

Determine transit and disintegration of selected polymer-coated pellets in vivo by CT scan

A series of experiments was carried out to monitor disintegration of tableted formulations within the gastrointestinal tract of possums by CT scan. In these experiments, possums were lightly anaesthetised by halothane inhalation for the administration of the tablets, and for each of the CT observations. Each period of anaesthesia lasted for less than 5 min and at other times throughout the experiment, the animals were fully conscious and were maintained under normal husbandry conditions. Although these experiments are primarily designed to determine the disintegration behaviour of the pellets, they also provide some data on gastrointestinal transit times of the tablets. For example, Figure 1 shows the location of barium-filled tablets within the stomach immediately after dosing and within the small intestine 4h later.

The matrix tablets were exceedingly robust when incubated in vitro and none had disintegrated after 48h of immersion in buffer or acid. However, as the polymer used in that formulation (ethyl cellulose) may be degraded by the micro-organisms of the hindgut, it is critical that hindgut microflora activity was not impaired when disintegration was being tested. As we cannot be sure that microflora activity was not affected in our in vitro tests, it was agreed that degradation of matrix tablets must also be assessed in vivo where they would be exposed to intestinal enzymes in situ. In these studies, a total of 4 matrix tablets were administered to each possum while under light anaesthesia. The animals were monitored immediately to confirm that all tablets were in the stomach and again at 4, 8, 12 and 24h after dosing. By 4h the tablets had entered the small intestine and at 8 and 12h were observed to have moved progressively further through the small intestine. By 24h all tablets were located in the hindgut or in the faeces. However, there was no evidence of disintegration or of polymer breakdown at any of the timed observations.

Pelleted formulations, selected on the basis of their in vitro disintegration characteristics, have also been assessed in vivo. The coated tablets were administered to possums by incubation, and their location and physical appearance (intact or disintegrated) was monitored by CT scan, immediately after administration and at 4, 8, and 24h after dosing. As observed with the matrix tablets, the coated tablets had entered the small intestine by 4h and were further down the small intestine at 8h. All tablets were located within the hindgut by 24h, but none showed any signs of disintegration.

Objective 2

Targeting specific regions of gut

To successfully design formulations that will target delivery to a specific region of the possum gut where they will release their payload of biocontrol agent, requires detailed information of gastrointestinal transit in this species, We have collected preliminary data on gastric retention times, but need information on transit through, and residence times in, each region of the gut and on how this is influenced by food intake and the particle size and density of the pellets. In this objective, gastrointestinal transit of solutions and of different sized particles through the possum gut will be determined using radioisotopes.

There are two specific aims of this objective for developing formulations that will target the possum hindgut. Firstly, to assess those formulations (especially the polymer component) developed in Objective 1 for characteristics that make them appropriate for targeting the hindgut and secondly, to gain an understanding of the factors that influence transit through the possum intestine and residence times within specific regions of the intestine.

The ideal formulation must be sufficiently robust to remain intact as it passes through the stomach and small intestine of the possum, but also must have characteristics that ensure that breakdown occurs rapidly within the caecum and colon. These features will be determined by incubating each formulation in vitro, initially in media that mimic the intestinal environment and finally in luminal contents from these specific regions of the possum intestine.

Gastrointestinal transit of solutions and of particles of various sizes through, and their selective distribution within the possum intestine, is primarily being measured by gamma-scintigraphy. Solutions or ion-exchange resins (to represent particles of different size) are radiolabeled, administered to possums and their transit through and distribution within the possum intestine is determined at various time intervals after administration.

Achievement of Milestones

Determine stability of polymer-coated pellets in lumen contents from stomach and small intestine, and their disintegration in contents from caecum and colon.

The formulations developed to date have all proved to be robust in acid media and in buffer, suggesting that they would also be stable within the stomach. Therefore, these in vitro studies have centred on disintegration within luminal fluid from the hindgut. Degradation of each of the polymer-coated tablet formulations has been assessed in luminal contents from the possum colon and caecum. In order to minimise any adverse effects that the procedure may have on microflora on enzyme activity, the luminal contents were collected under anaerobic conditions using an anerobic glove box flushed with medical-grade carbon dioxide. Luminal contents from each region were placed in sealed vials containing the tablets under investigation, and flushed with carbon dioxide. Replicates, each containing a single tablet of each of the formulations, were incubated in a 1:1 dilution (with phosphate buffer) of luminal contents. In addition, coated tablets that had previously been incubated in acid (0.1 M HC1) for 2h (to mimic the action of residence in the stomach) were then subjected to incubation in luminal contents. The vials were maintained at 37^oC in a shaking water bath for 24h. The integrity of the tablets (intact or disintegrating) was recorded at time 0, 1, 2, 3, 4, 5, 6, 7, 8, 12 and 24h after the start of incubation.

Few of the coated tablets had disintegrated by 8h of incubation in luminal contents. At 4h one single-coated tablet in colonic contents had begun to break up, but even by 24h the other single-coated tablets remained intact. When preincubated in acid, 1 tablet in colonic (single-coated, 6h) and 1 tablet in caecal contents (single-coated, 1h) had disintegrated, but all other single-coated tablets remained intact. None of the double-coated or triple-coated tablets showed any evidence of disintegration over the 24h incubation period.

Establish transit and residence time of radio-labelled solutioned and particulates in the possum gastrointestinal tract.

Transit studies using the radiosotope technetium (^99mTc) are underway for both solutions and particulates, with a non-absorbable form of technetium (Te-DTPA) being used for the solution preparations. Transit studies of particulates involve the use particles of a synthetic, indigestible and non-toxic anionic exchange resin (Amberlite IRA 400) of different size classes, again labelled with technetium. After preparation of the labelled particles, the resin is washed and filtered to remove unbound technetium.

Prior to carrying out in vivo transit studies, the stability of the technetium-resin complex was tested in the presence of gut contents from different regions in the possum gastrointestinal tract. It was found that, with a single exception, the binding of technetium-resin was greater than 85% in lumen contents from all regions of the possum gastrointestinal tract. The exception was that when incubated with contents from the possum proximal colon, technetium-resin binding was approximately 70%. To assess the effect of acid on technetium-resin binding, the resin particles were incubated in solutions of 0.1M HC1 and phosphate buffer(pH 7). Following incubation, approximately 99% of the radioactivity detected was bound to resin. It was also shown that binding appears to be independent of the ratio of gut content to resin, and also to the duration of the incubation period.

A protocol for delivering radiolabeled particles to the brushtail possum for in vivo studies has been developed. The animals are anaesthetised by halothane inhalation and radiolabeled particles are administered orally, by passing a tube to the back of the throat and expelling the labelled particles (suspended in water) using a pipette. For the fluid treatments, the solution of technetium-pertechnetate is delivered directly into the incubation tube. Following administration, possums are kept in individual cages and all faeces and urine collected. At pre-determined time intervals after administration of the radiolabeled dose (3, 6, 12, 24 and 36h) animals are euthanased and the distribution of the labelled particles or solution is determined by gamma scintigraphy. The gastrointestinal tract is removed, placed on the detector of the gamma camera and scintigraphic images taken. The total radioactivity is counted and the percentage activity in each region of the gut is determined. With radiolabeled fluid, 3h after dosing most of the radioactivity was located in the stomach, with some throughout the small intestine and even in the caecum. In contract, with particles in the size range of 500-700mm, 3h after dosing almost all of the labelled particles remained in the stomach, with a small amount having just entered the small intestine being present in the duodenum. By 6h after dosing, particles have progressed to the jejunum of the small intestine with some activity remaining in the stomach. After 24h, the particles were detected throughout small intestine, in the caecum and proximal colon and within the faecal pellets of the proximal colon.

PUBLICATIONS

McLeod, B.J., Thompson, E.G. (2002) Predation on house sparrows (Passer domesticus) and hedge sparrows (Prunella modularis) by brushtail possums (Trichosurus vulpecula) in captivity. Notornis 49, 97-101.

Butt, A.G., Mathieson, S.E., McLeod, B.J. (2002) Electrogenic ion transport in the intestine of the common Australian brushtail possum, Trichosurus vulpecula: indications of novel transport patterns in a marsupial. Journal of Comparative Physiology B 172 (in press)

Butt, A.G., Mathieson, S.E., McLeod, B.J. (2002) Aldosterone does not regulate amiloride-sensitive Na⁺ transport in the colon of the common Australian brushtail possum Trichosurus vulpecula. Journal of Comparative Physiology B 172 (in press)

Wen, J.Y., Ledger, R., Davies, N.M., Butt, A.G., McLeod, B.J., Tucker, I.G. (2002) Protein peptide degradation in the gut of the brushtail possum (Trichosurus vulpecula) Journal of Comparative Physiology B 172 (in press)

Wen, J.Y., Davies, N.M., Ledger, R., Butt, A.G., McLeod, B.J., Tucker, I.G. (2002) An isocratic HPLC assay for monitoring the degradation of Luteinising Hormone Releasing Hormone by extracts from the gastrointestinal tract of possums Journal of Chromatography B: Biomedical Applications 780 (in press)

Mahoney, P.M., Hurst, P.R., McLeod, B.J., McConnell, M.A. (2002) Quantification of mast cells and microflora in the cul-de-sac of the brushtail possum (Trichosurus vulpecula). Reproduction 124, (in press).

Wen, J.Y., Ledger, R., Butt, A.G., McLeod, B.J., Davies, N.M., Tucker, I.G. (2002) Inhibition of proteolysis in luminal extracts from the intestine of the brushtail possum (Trichosurus vulpecula ). Journal of Pharmacy and Pharmacology 5, (in press)

Conference Presentations

Wen, J.Y., Davies, N.M., Ledger, R., Butt, A.G., McLeod, B.J.., Tucker, I.G. (2001) Enzymatic barriers in the possum to intestinal absorption of proteins and peptides. 28^th International Symposium of the Controlled Release Society San Diego.

Mahoney, P.M., Hurst, P.R., McLeod, B.J., McConnell, M.A. (2001) Are mast cells important for reproductive function in the brushtail possum? Annual Conference of the Society for the Study of Fertility. Cambridge.

Wen, J.Y., Tucker, I.G., Davies, N.M., Ledger, R., McLeod, B.J., Butt, A.G., (2001) Effect of absorption enhancers on permeability of the hindgut mucosa in the brushtail possum (Trichosurus vulpecula). XXXIV International Congress of Physiological Sciences Christchurch.

Mahoney, P.M., Hurst, P.R., McLeod, B.J. (2001) Morphology of mast cells in the vaginal cul-de-sac of the brushtail possum. Annual Conference of the Society of Reproductive Biology, Gold Coast.

Wen, J.Y., Tucker, I.G., Davies, N.M., Ledger, R., McLeod, B.J., Butt, A.G. (2001) The effect of absorption enhancers on the permeability of the large intestine of the Australian brushtail possum, Trichosurus vulpecula. 5^th Annual Conference on Formulation and Delivery of Bioactives Dunedin.

Fink, J., McLeod, B.J., Nicholson, H. (2002) Identification of an oxytocin-like receptor in the prostate of the brushtail possum (Trichosurus vulpecula). Annual Conference of the New Zealand Society of Endocrinology. Christchurch.

Nicholson, H., Fink, J., Assinder, S., McLeod, B.J., Parry, L. (2002). Identification and localisation of a mesotocin receptor in the prostate of the brushtail possum. Annual Conference of the Society of Reproductive Biology. Adelaide

Mahoney, P.M., Hurst, P.R., McLeod, B.J., McConnell, M.A. (2002) Exogenous oestradiol influences mast cell populations in the vaginal cul-de-sac of the Brushtail possum Annual Conference of the Society of Reproductive Biology. Adelaide

5.8 PBC 227

Programme Title:	Transgenic worms for the control of possums
Programme Leader:	Dr Warwick Grant
Institution:	AgResearch

Summary

We have successfully demonstrated that transformation frequency with microinjection is high and that the transgene is most likely inherited as a genetically unstable extrachromosomal array in the free-living worms. We have also demonstrated that the transgene is present and inherited at much higher frequency than can be detected by assay of the transgenic reporter gene’s activity. These two data indicate that development of a protocol for genetically stable transformation and defining promoters for greater transgene expression are the next two hurdles to be overcome in the path to a biological control agent based on transgenic nematode parasites.

Background

Milestone #1.

Establishment of sib-selection culture methods for quantification of transformation efficiency and measurement of microinjection efficiency.

Milestone #2

Completion of the quantitative comparison of alternative transformation methods (electroporation, biolistic bombardment, possibly pantropic retrovirus) and determination of the optimum method.

Milestone #3

Construction and testing of the collagen and other gene promoter/secretory leader reporter constructs in Caenorhabolitis. elegans and Parastrongyloides. trichosuri.

Approach & Outcomes

Milestone 1

The aim of this milestone was to develop methods to permit measurement of the frequency of transformation in the progeny of the microinjected adult worm (the F1 generation) and then to measure the frequency with which the transgene is inherited and its stability in the second (F2) and subsequent generations.

Three approaches were used. First, F1 progeny were collected before they reached maturity and the presence of the transgene was determined by vector specific PCR on individual worms. This permitted measurement of the frequency with which F1 worms inherited the transgene from their microinjected mother. Approximately 50-70% of F1 worms appear to carry the transgene.

Second, the F1 worms were cultured until they had reached sexual maturity and F2 eggs or newly hatched larvae were present in the cultures. At this point the F1 worms were removed (for single worm PCR) and the F2’s allowed to grow to maturity, at which time they too were collected for single worms PCR. Data from two such experiments indicate that approximately 5-10% of the F2’s carry the transgene. In some experiments we split the F2 population in single wells into two subpopulations. One of these was subjected to PCR to determine the extent of transgene inheritance and the other subpopulation was stained for b-galactosidase activity to determine the expression of the transgene. The results from these analyses showed that there was good correlation between the PCR data and the expression: no cultures showed b-galactosidase activity if there was no transgene detected by PCR. The activity was, as expected, restricted to the gut (see Figure, the GATA promoter is gut specific in C. elegans). The converse was not observed, so that there were cultures that showed the presence on the transgene by PCR but where no b-galactosidase activity was detected.

The third approach aims to measure more accurately the frequency with which an individual F1 worm is likely to transit the transgene to its progeny and to differentiate between the possibilities that (a) many/all F1’s transmit the transgene to a small proportion of their F2 progency or (b) a small number of F1’s transmit the transgene to the majority of their progeny.

In order to do this, a method for small-scale culture of single male/female pairs was established, either in 5ml of liquid medium or on small agar plugs in the wells of microtitre plates. These small cultures require the addition of an organic extract of possum faeces as a supplement and are successful in approximately 50% of cases. We aim to improve the efficiency of these cultures. Single male and female fourth stage or early adults (i.e. apparently virgin) F1 worms are placed in each well and monitored until they produce eggs. The adults are then removed for single worms PCR and the progeny permitted to develop until they are late larvae/early adults. The efficiency of the mating and the yield of progeny per well are variable at this time.

In summary, we have established in vitro post-injection culture and PCR methods for free-living P. trichosuri to accurately measure the frequency of transformation in the Fl and F2 generations. These data suggest that the transformation frequency and inheritance of the transgene are similar to the free-living nematode C. elegans. We have not performed experiments to investigate the long term transmission of the transgene, but the data reported here suggest that the transgene is inherited as a genetically unstable extrachromosomal array. Thus it is important to establish either (a) conditions for the integration of these arrays into the genome in a genetically stable form (this can be accomplished by g-irradiation in C.elegans) or (b) alternative conditions for transformation using a method that produces low copy number stable integration at the transformation step (e.g. biolistic particle bombardment in C. elegans).

Milestone 2:

The aim of this milestone was quantitative comparison of alternative transformation methods (electroporation, biolistic bombardment, possibly pantropic retrovirus) and determination of the optimum method. This relied to a large extent on appointing a replacement for Dr Steve Skinner, who left the project in mid-2001. A suitable candidate was identified and an offer made. Unfortunately, and only after 2-3 months’ wait, the offer was finally refused. No other appointable candidates were identified at the time and the position remains unfilled. It will be re-advertised in June, 2002. Our failure to find a suitable replacement for Dr Skinner has severely hampered our ability to meet milestones 2 and 3.

Several electroporation experiments were carried out, varying the charge and voltage of the treatment. Conditions which gave good survival of eggs and worms (both of larvae and adults) were determined for C. elegans and P. trichosuri free-living worms. Trial transformation using the rol-6 plasmid pRF4 and the gut specific reporter gene constructs GATA-98 and GATA-212 were carried out for both species, against eggs, larvae and adults. No rollers or b-galactosidase staining worms were detected for any combination of plasmids, worms and electroporation conditions. Thus it appears unlikely that electroporation is a viable alternative to microinjection and no further experiments using this technique are planned.

At the time these experiments were nearly completed it became clear that a revised ERMA approval may be required for the use of alternative methods of transformation. No biolistic experiments are planned until the appropriate ERMA approvals are in place. The application process for these approvals is in progress.

Milestone 3

The data from milestone 1 suggested that the apparently low transformation frequency previously observed was more likely due to the poor expression of the transgene rather than from poor transformation frequency. This could be due to either an inherently poor expression of transgenes (for example, transgene expression in the germline of C.elegans is very poor compared to the somatic tissues), or it could be due to the poor efficiency of the promoters used. One of these (the hsp-70 promoter) is from P. trichosuri but nothing is known concerning tits usual expression level or tissue specificity and the second (the C.elegans GATA promoter) is from another species. Thus there is a need to test a greater range of promoters from better characterised P. trichosuri genes. The EST database is a good source of such candidate genes. We had previously taken two EST’s from the first few hundred that were available: glyceraldehyde-3-phosphate dehydrogenase (G3PDH) and a cDNA that is highly conserved and expressed in other parasitic nematodes. It encodes an abundant secretory antigen of unknown function called either “24kDa antigen” or “p22 upper” in filarial nematodes. We have been unable to recover genomic sequences for either of these by PCR and have determined that at least the G3PDH clone from which we were working is likely to be chimeric. We have also attempted to define the promoter region from a collagen clone isolated from a P.trichosuri genomic library by Dr Allen Gruenberg in 1998-99. The 5’ region of this clone has proven to be very difficult to sequence and we have not been able to unambiguously define the translation initiation codon (due to poor sequence homology in this region with other nematode collagens). Thus our initial attempts to isolate new promoters for transgene expression have been unsuccessful.

The EST database has increased greatly in size in the past year and now contains 7000-8000 P. trichosuri EST’s from several life cycle stages. These have been analysed for frequency of occurrence (a rough guide to expression levels) and likely homology to other nematodes (as a guide to likely tissue specificity and function) and a list of candidates has been drawn up. We have also attempted to revive the original genomic library constructed for Dr Allen Gruenberg in 1998. So far the titres of the library aliquots stored by Dr Gruenberg have been too low to permit any screening of this library and alternative PCR-based strategies for the isolation of promoter regions of the genes defined by the EST’s will probably be necessary (or construction of a new library).

5.9 PBC 228

Programme Title:	Bovine TB
Programme Leader:	Dr Bryce Buddle
Institution:	AgResearch

Summary

A DNA prime/BCG boost vaccination strategy in cattle induced a higher level of protection against challenge with Mycobacterium. bovis than BCG alone.
Vaccination of calves at birth or at 6 weeks of age induced a particularly high level of protection against development of tuberculous lesions following a M. bovis challenge, while vaccination with two doses of BCG was contra-indicated.
The numbers of interferon-g producing cells from peripheral blood detected following stimulation with bovine PPD did not correlate with protection, although the frequency of these cells was elevated in mycobacterial DNA-vaccinated calves.
Cannulation of the afferent lymph duct of calves revealed that T cells from lymph draining a BCG vaccination site become activated within 2 days after vaccination.
Preliminary studies using DNA micro-array technology have shown that it is feasible to use this approach to investigate gene expression following BCG vaccination or M. bovis infection of calves.
A trial is in progress to determine if a combination of two M. bovis specific antigens (ESAT-6 and CFP10) can enhance the specificity of the interferon-g test for the diagnosis of tuberculosis in cattle from the West Coast of the South Island.
A cytokine TNF-a has a role in protection against mycobacterial infections and a recombinant BCG strain has now been produced which expresses possum TNF-a. In possums, this potential vaccine strain appears to be less pathogenic than the parent BCG strain, but produces a similar level of protection against tuberculosis.
An effective means of delivery BCG to possums as an oral bait has been achieved by encapsulating the BCG in lipid. Three different types of lipid have been successfully used to encapsulate BCG, while a fourth type was not effective.
A cytokine, transforming growth factor-b (TGF-b), has potential adjuvant properties and can stimulate antibody responses as well as promote lgA switching at mucosal surfaces. Possum TGF-b has now been cloned by our group and is currently being expressed in yeast.
An immunohistochemical technique has been developed to identify T lymphocytes in gut associated lymphoid tissues of possums.
A recombinant BCG strain has been produced which expresses a sheep louse antigen (model antigen) to test the feasibility of producing a combined vaccine for protecting against tuberculosis and inducing an antibody response to achieve immunosterilisation of possums.
Fourteen refereed scientific papers were published, accepted or submitted for publication in the past year as well as numerous papers presented at conferences.

Background

a) Improved diagnostic tests and vaccines for control of bovine tuberculosis

Develop and evaluate diagnostic tests and vaccines for control of bovine tuberculosis in cattle and possums.

Determine whether vaccination with a DNA vaccine and revaccination with BCG vaccine induces a greater level of protection in cattle against bovine tuberculosis than BCG alone.
Determine whether age at which calves are vaccinated with BCG is a critical factor in protection against bovine tuberculosis, particularly when they have been sensitised to environmental mycobacteria.
Determine whether the frequency of IFNg producing cells as detected by ELISpot correlates with protection against bovine tuberculosis in cattle.
Cannulate the afferent lymphatic duct of the prescapular lymph node of calves to directly measure the immune responses to BCG vaccination.
Determine whether a combination of ESAT-6 and CFP10 used in the interferony test can be useful in differentiating cattle infected with M. bovis from those infected with environmental mycobacteria.
Establish whether vaccination of possums with a recombinant BCG expressing a possum cytokine, TNF-a, is more effective in protecting possums against bovine tuberculosis than BCG.
Compare the efficacy of four different oral bait BCG vaccines prepared by F. Aldwell (University of Otago) by challenging vaccinated possums with an aerosol of M. Bovis.
Determine the optimal dose of BCG in an encapsulated oral bait vaccine protection against tuberculosis by vaccinating groups of possums with one of three doses of BCG and later challenging them with M. Bovis.
Submit three papers to refereed journals.

b) Possum Immunology

Develop methods for measuring mucosal immune response in possums and use them to determine how responses are regulated by possum cytokines. Develop vaccine delivery systems suitable for combined delivery of tuberculosis and biocontrol vaccines.

Produce one new recombinant possum cytokine and determine whether it enhances mucosal immune responses.
Use immunohistochemistry to locate T cells in Peyer’s patches and mesenteric lymph nodes after oral immunisation with BCG.
Determine the feasibility of using a recombinant BCG vaccine for protecting possums against tuberculosis and inducing an appropriate antibody response to achieve immunosterilisation of possums.
Submit one paper to a refereed journal.

Approach & Outcomes

DNA prime/BCG boost for protection of cattle against M. bovis infection (funded predominantly from Department of Environment, Food and Rural Affairs (DEFRA), UK

In collaboration with Glyn Hewison and Martin Vordermeier, VLA Weybridge, a study has been undertaken to investigate the potential of a new vaccination strategy for bovine tuberculosis. A prime-boost strategy, using DNA vaccines and attenuated fowl pox virus vectors each expressing similar antigens, can stimulate protective cell-mediated immune responses against viral and bacterial infections. A similar strategy using DNA vaccines followed by BCG vaccine was investigated to determine whether it might be suitable for stimulating better protective immunity against bovine tuberculosis in cattle than DNA or BCG alone. In the current protocol a combination of three DNA vaccines, hsp65 DNA, hsp70 DNA and apa DNA was supplied by VLA and used in the current study. Forty-eight calves were divided into 4 groups. Prior to vaccination, the animals had strong interferon-g (INF-g) responses to purified protein derivative from M. avium (avian PPD) indicating exposure to environmental mycobacteria. Previous work has shown that BCG vaccine is less effective in protecting calves against bovine tuberculosis in this situation. Animals in Group 1 were immunised with two ‘prime’ vaccinations of DNA vaccine at 0 and 3 weeks given intramuscularly and intradermally and boosted with BCG at 6 weeks. Animals in Group 2 were given only the two DNA ‘prime’ immunisations. Animals in Group 3 were given only BCG at week 6 and the fourth group of animals was not vaccinated. They were all challenged with M.bovis at week 13. Immune responses were followed throughout and skin tests were carried out before challenge and after challenge. The level of protection was assessed at post-mortem by a reduction in macroscopic tuberculous lesions.

Only animals receiving BCG gave a positive skin test of > 4mm after vaccination. Strong IFN-g and interleukin 2 (IL-2) responses to bovine PPD were observed at 8-10 weeks in animals vaccinated with BCG alone and similar responses were found when the prime-boost regime was used. Animals vaccinated with the DNA vaccine alone did not give IFN-g or IL-2 responses until after challenge and then gave the strongest of any of the groups. In contrast to the results from the amount of IFN-g producing cells from peripheral blood mononuclear cells (PBMC) stimulated with bovine PPD gave different results. The IFN-g ELISpot assay showed that the DNA-vaccinated group had significantly higher number of IFN-g producing cells from bovine stimulated PBMCs than the non-vaccinated group. In addition, the DNA-vaccinated group had higher numbers from PBMC cultures stimulated with hsp65 and hsp70 than the DNA prime/BCG boost group and BCG alone groups (P<0.05). These results indicate that the DNA vaccine induced the antigen-specific IFN-g producing cells, although the total production of antigen-specific IFN-g appeared to be less than those groups vaccinated with BCG.

The skin test response after challenge with M.bovis was significantly less in the DNA/BCG group than in the non-vaccinated group (P<0.05), but there was no significant difference between the other two vaccinated groups and non-vaccinated controls (Figure 1). The DNA/BCG group also gave significant reduction in five pathological correlates of disease as shown in Table 1. There were fewer animals with lung or lymph node lesions, the mean lung and lymph node scores were less and there was a decrease in the mean number of lesioned lymph nodes per animal compared to the non-vaccinated controls. BCG alone produced a significant reduction in only two pathological correlates; fewer animals with lymph node lesions and a lower mean lymph node score. With DNA alone there was no significant reduction in disease. When the microbiological results were compared, all animals in the control and DNA alone groups were M. bovis culture positive, whereas 10 of 12 were positive in the BCG alone group and 9 of 12 were positive in the DNA/BCG group.

These findings confirm our previous work that DNA vaccines alone did not induce protection against challenge with M. bovis, even when they are administered by two routes, intradermally and intramuscularly. However, when DNA vaccines were used prior to BCG vaccination, the level of protection induced was greater than that seen for BCG alone. This is one of the first instances where a vaccine or vaccine combination has been shown to perform better than BCG alone in any animal species.

Table 1. Pathological findings following challenge of cattle with M. bovis


Vaccine	Proportion of animals with:		Mean lung	Mean LN	Mean No. of
Group	Lung Lesion	LN Lesions	score^a	score^b	lesioned LN per animal
Non-vaccinated	9/12	12/12	2.67	3.75	2.42
DNA	8/12	12/12	2.92	4.00	2.58
DNA/BCG	3/12*	5/12**	0.42**	1.08***	1.08*
BCG	5/12	7/12*	1.33	1.92**	1.33

^aLung lesion score:

0, no lesions; 1, 1-9 lesions; 2, 10-29 lesions; 3, 30-99 lesions; 4, 100-199 lesions; 5,

> 200 lesions

^bLymph node (LN) lesion score:

0, no lesions; 1, 1-5 small lesions (1-4mm diameter); 2, 5-19 small lesions;

3, > 20 small lesions; 4, medium size lesions (5-9mm diameter);

5, large lesions (> 10mm diameter)

*significantly different from non-vaccinated group (P<0.05); **(P<0.01); ***(P<0.001).

Protection of neonatal calves by vaccination with BCG (funded predominantly from The Sequella Global Tuberculosis Foundation, USA).

The Sequella Global Tuberculosis Foundation has provided funding to use calves as a model for improving the efficacy of tuberculosis vaccines for human neonates. The investigation is directed to identify the optimal timing of the first vaccination and the requirement for revaccination where individuals are sensitised to environmental mycobacteria. This study is relevant to the control of bovine tuberculosis, as we have no information whether BCG vaccine can protect young calves as we have used older calves, 5-8 months of age in all our previous studies. BCG vaccination has been less effective in these older calves when they were sensitised to environmental mycobacteria prior to vaccination.

A group of calves were vaccinated within 8 hours of birth with BCG and a second group of calves were vaccinated with BCG at 6 weeks of age. A third group of calves were vaccinated within 8 hours of birth and again at 6 weeks of age. A fourth group of non-vaccinated calves served as a control group. When the calves were between 14-17 weeks of age they were challenged intratracheally with a low dose of M. bovis. The calves were killed 4 months after challenge and examined for tuberculous lesions.

Preliminary results have indicated that the calves became responsive to environmental mycobacteria at a very young age (6 weeks), with the majority of the non-vaccinated calves producing IFN-g from avian PPD-stimulated blood cultures. This phenomenon in these young calves is more likely to be a non-specific immune stimulation rather than exposure to environmental mycobacteria as many of the non-vaccinated calves responded to M. bovis specific antigens such as ESAT-6 and CFP10, although, they had no exposure to M. bovis. Furthermore, in a longitudinal study involving tuberculosis-free calves from 10 herds, this sensitisation to avian PPD in young calves was at a maximum level when the calves were 6-9 weeks of age and reduced to a very low level by 15 weeks of age (G. de Lisle, unpublished data). Vaccination with BCG at birth or at 6 weeks of age induced strong IFN-g responses to bovine PPD. T cell sub-set analysis using FACS to measure up-regulation of CD25 expression revealed that BCG vaccination induced the activation of CD4, CD8 and gd T cells. Calves vaccinated at birth or at 6 weeks produced similar skin test responses to bovine PPD when tested at 11 weeks of age.

The level of protection seen in the calves vaccinated at birth or 6 weeks was the highest we had ever seen in any of our previous BCG vaccination experiments. BCG vaccination at birth had an efficacy of 100% for protection against development of macroscopic tuberculous lesions, while vaccination at 6 weeks had an efficacy of 90% (Table 2). There was no significant difference between these levels of protection, indicating that the responsiveness of the calves to environmental mycobaterial antigens at 6 weeks did not appear to interfere with development of protection against tuberculosis. In contract, animals vaccinated both at birth and 6 weeks of age had significantly less protection than those vaccinated only at birth. Detailed analysis of the immune responses is still to be completed and currently, we do not have an explanation for the poor efficacy of a double vaccination with BCG. The results of skin testing revealed that the protection induced by vaccination with BCG could reduce skin test responses to bovine PPD at 15 weeks after challenge. The standard interpretation for a positive reaction in the comparative cervical skin test is defined as increase in skin thickness for the bovine PPD site – corresponding response for the avian PPD site>4mm. Using this interpretation, 10/10 were positive for the non-vaccinated group, 5/10 for the BCG at birth group, 5/9 for the BCG at 6 weeks group and 9/10 for the BCG at birth and 6 weeks group. Culture of M. bovis from tissue samples is in progress. In summary, vaccination of calves in the first 6 weeks of age with BCG vaccine appeared to induce a higher level of protection than that seen when calves were vaccinated at 5-8 months of old when vaccine efficacy varied from 0-70% (results from eight previous AgResearch studies where BCG vaccine had been used in calves). Vaccination of young calves with two doses of BCG appears to be contra-indicated.

Table 2. Pathological findings following challenge of cattle with M. bovis


Vaccine Group	Proportion of animals with:		Mean lung score^a	Mean LN score^b	Mean No. of lesioned LB per animal
	Lung lesion	LN lesions
Non-vaccinated	10/10^a	10/10^a	2.4^a	3.75^a	3.0^a
BCG at birth	0/10^b	0/10^c	0^c	0^c	0^c
BCG at 6 weeks	0/9^b	1/9^bc	0^c	0.1^bc	0.2^c
BCG at birth and 6 weeks	3/10^b	4/10^b	0.9^b	0.5^b	1.2^b

^aLung lesion score:

0, no lesions; 1, 1-19 lesions; 2, 10-29 lesions; 3, 30-99 lesions; 4, 100-199 lesions

^bLymph node (LN) lesion score:

0, no lesions; 1, 1-19 small lesions (1-4 mm diameter); 2, >20 small lesions;

3, medium size lesion (5-99mm diameter); 4, large lesion (> 10mm diameter)

Figures in columns with the same superscript letter are not significantly different (P>0.05)

Frequency of IFN-g producing cells after DNA vaccination does not correlate with protection against M. bovis infection in cattle.

A cattle ELISpot assay for quantifying the frequency of IFN-g producing cells has been optimised with regard to the plates used, the antibodies employed and whether counting of the ELISpots is undertaken manually using a low power microscope or by an automated instrument. A standard method is now in operation and thanks to our collaborators at VLA Weybridge we are able to use instrumentation to count the ELISpots. This means that assays can be performed on large sample numbers generated during vaccine trials. The number of IFN-g secreting cells following stimulation of peripheral blood mononuclear cells (PBMC) with mycobaterial antigens was measured in the DNA prime/BCG boost vaccine trial and the results are shown in Table 3. Although the DNA vaccine alone stimulated the highest frequency of IFN-g secreting cells in response to mycobacterial antigens present in M. bovis there was no evidence of protection against tuberculosis (see Table 1). There was significant protection after vaccination with BCG or with the DNA prime/BCG boost regime although the frequency of IFN-g producing cells in response to mycobacterial antigens in these animals was similar to non-vaccinated controls. These results are in contrast to the levels IFN-g released from whole blood cultures which were highest for the calves vaccinated with BCG. Therefore, in this vaccine trial the frequency of IFN-g producing cells as detected by ELISpot did not correlate with protection but did indicate that DNA vaccination was successful in stimulating a cell-mediated immune response. Determination of both the number of IFN-g producing cells and amount of IFN-g produced will aid in our understanding of immune responses to mycobacterial antigens, an important area for the development of improved methods for the diagnosis of bovine tuberculosis.

Table 3 Number of IFN-g producing cells in PBMCs from vaccinated cattle

Vaccine group	Avian PPD	Bovine PPD	Hsp65	Hsp70
Non-vaccinated	1.98 (+0.15)^a*	1.62 (+0.21)^b	1.65 (+0.15)^ab	1.47(+0.19)^ab
DNA	2.15 (+0.16)^a	2.06 (+0.12)^a	1.76 (+0.17)^a	1.94(+0.16)^a
DNA/BCG	2.05 (+0.13)^a	1.97 (+0.14)^ab	1.32 (+0.19)^b	1.36 (+0.21)^b
BCG	1.9 (+0.15)^a	1.69 (+0.12)^ab	1.27 (+0.08)^b	1.25 (+0.09)^b

*Mean log₁₀number of IFN-g spot forming cells per 10⁶PBMCs (+SE). Figures in columns with the same superscript are not significantly different(P>0.05).

Immune responses in afferent lymph of calves vaccinated with BCG.

The aim of this study was to establish a technique to cannulate the afferent lymphatic duct of the prescapular lymph node in order to study regional immunity after vaccination with BCG. This may lead to a better understanding of what constitutes protective immunity and give better correlates of protection.

The prescapular lymph node on each side of the neck was removed surgically from six calves. The prescapular afferent lymph duct was cannulated and a heparin infusion pump inserted to minimise clotting and prevent cessation of lymph flow. Two animals with a good flow of lymph were vaccinated subcutaneously in the neck with BCG and pre and spot-vaccination lymph samples collected. IFN-g and IL-2 responses were measured in lymph cultured with PBS or bovine PPD using ELISA and bioassay, respectively. The expression of cytokine mRNA in both freshly collected non-cultured lymph (ex vivo lymph) and cultured lymph was also measured by real-time quantitative PCR using a GeneAmp 5700 sequence detection system. T cells from ex vivo and cultured lymph were analysed for phenotype (CD4, CD8 and gd) and activation status (IL-2R). For comparison, IFN-g and IL-2 responses were also measured in blood.

Cytokine expression and T-cell activation in afferent lymph following vaccination with BCG After BCG vaccination, weak bovine-PPD specific IFN-g responses were detected in afferent lymph by ELISA. Real-time PCR also showed a weak but discernible bovine PPD-specific IFN-g response at 5-6 days after vaccination. IFN-g responses in peripheral blood, as measured by both ELISA and PCR, were weak at 7 days after vaccination, while strong IFN-g and IL-2 responses inblood were observed at 2.5 weeks post-vaccination. Real-time PCR was used to directly measure cytokine mRNA levels in lymph cells. The levels of mRNA in ex vivo lymph for IFN-g and IL-4 were increased at 2 days after BCG vaccination compared to pre-vaccination levels (Table 4).

Table 4. Relative expression of cytokine mRNA in ex vivo lymph

Time	Relative cytokine mRNA expression*
	IFN-g	IL-4
Pre-vaccination	0	0
2 days post-vaccination	4.4	2.1
5 days post vaccination	3.8	<0

^· Levels of cytokine mRNA (normalised to house-keeping gene) after vaccination relative to pre-vaccination levels, defined as 2^{-[ Ct 9post-vaccination) - Ct (pre-vaccination)}

T-cell activation in afferent lymph was determined by FACS. Lymphocytes in ex vivo lymph were double stained for bovine CD4+, CD8+ or gd T-cells expressing IL-2R was slightly increased at 2 days after BCG-vaccination compared to prevaccination. This increase was followed by a large decrease in IL-2R expression in all three T-cell subsets at 5 days after vaccination.

These results, suggest that following subcutaneous injection of BCG, T-cells in the prescapular afferent lymph which drains the vaccination site are rapidly activated and there is a concomitant increase in cellular levels of IFN-g and IL-4 mRNA. The subsequent decrease ion activation may reflect the migration of activated T cells from or to the vaccination site.

Gene expression profiling

A component of the work on the afferent lymphatics centred on using DNA microarrays to identify genes in lymphocytes and other cells present in lymph which are differentially regulated. The use of DNA microarray technology may lead to discovery of host genes, which play an important role in either protective immunity against tuberculosis or alternatively pathogenesis. Since DNA microarrays are printed with thousands of genes or ESTs, the expression of a very large number of genes can be measured in a single experiment. DNA microarrays printed with approximately 10,000 bovine genes or ESTs have been produced as part of an AgResearch gene discovery initiative. The arrays were also printed with a number of cytokine genes which we have cloned as part of our work on bovine tuberculosis. Lymph samples were collected from the afferent lymph of the two vaccinated calves before and after vaccination and RNA prepared from the lymph has been stored for future experiments with DNA microarrays.

Since the DNA microarray technology is relatively new, and preliminary experiment was conducted on cultured PBMCs from one of the BCG-vaccinated calves used in the afferent lymph studies. PMBCs were isolated at 2.5 weeks after vaccination and cultured for 20 hr with either PBS or bovine PPD. Highly purified RNA was prepared from the two populations of PBMCs (PBS and bovine PPD), labelled with Cy5 and Cy3 fluorescent dyes and hybridised to the DNA microarrays. Analysis of the data revealed a number of differences in gene expression between non-stimulated and antigen-stimulated PBMCs.

Development of a more specific test for the diagnosis of tuberculosis in cattle

The tightening of movement control regulations and the reduction in amount paid in compensation for reactors have increased the demand for more specific ancillary tests for the diagnosis of bovine tuberculosis. Preliminary results from a MAFPolicy/AHB funded trial in 2000/01 showed that a combination of two M. bovis antigens (ESAT-6 and CFP10) used in the IFN-g test markedly enhanced specificity compared to the standard IFN-g (BOVIGAM) test. The use of these specific antigens in the first year was aimed at the reducing false positive reactions in uninfected herds from areas where there have been no previous reports of bovine tuberculosis (Northland and Taranaki) and the results look very promising. Another application would be for the use of this new test in a tuberculosis endemic area, which has a history of non-specificity skin test reactivity (West Coast of the South Island). On the West Coast, a modified interpretation of the standard IFN-g test is used with a cut-off of bovine PPd — avian PPD > 0.3 rather than the usual bovine PPD — avian PPD > 0.1. This is necessary due to the animals having a high non-specific reactivity to avian and bovine PPD. The high cut-off means that some tuberculosis-infected animals are being missed. A more specific assay could prevent wastage from culling non-infected animals and still maintain a low cut-off for detecting the highest number of M.bovis-infected animals.

A project was started in this financial year to determine if an IFN-g test with enhanced specificity for detection of M. bovis infection could more efficiently identify tuberculous cattle on the West Coast. Blood samples were collected from cattle which had reacted positively in the skin test in the last 10-30 days. The bloods were assayed using the standard IFN-g test as well as with the ESAT-6/CFP10 antigen combination.

Blood samples have been submitted from 189 animals which reacted to a caudal fold skin test in the previous 10-30 days. The source of these animals have been the West Coast of the South Island (175 animals), Southland (5 animals) and Waikato (9 animals). Twenty-seven of these animals reacted positively in the standard IFN-g test (B-A>0.1) and 14 reacted positively to the two M. bovis specific antigens (ESAT-6 and CFP10 peptides) using a cut-off of Antigen — Nil >0.035 OD units x 1000. Six of these animals were positive in both the standard IFN-g test and ESAT-6/CFP10 tests. A total of 24 of animals which had reacted positively in one or both of these tests were killed in an abattoir and 7 had lesions which were suspicious of bovine tuberculosis. Samples from the suspicious lesions have been submitted for bacterial culture and from pooled lymph nodes of an additional 8 non-lesioned reactor animals. From the 7 animals with suspect lesions, 7 were positive in the standard IFN-g test and 3 were positive in the ESAT-6CFP10 test.

The majority of the animals have come from non-infected herds. There have been fewer non-lesioned animals reacting in the ESAT-6/CFP10 test than the standard IFN-g test, but the special antigen test has missed some animals with suspect tuberculous lesions. We have sufficient EST-6/CFP10 peptides to sample a further 80 animals and it is hoped that these will come from tuberculosis-infected herds.

Vaccination of possums with a recombinant strain of BCG which expresses TNFa for protection of possums against bovine tuberculosis.

Vaccination of possums with BCG induces significant levels of protection against an experimental challenge with virulent M.bovis. The efficacy of BCG may be enhanced if the bacilli are engineered to express a cytokine such as TNFa. TNFa has the ability to enhance granuloma formation and increases protection against mycobacterial infections. A strain of BCG has been made which contains an expression plasmid with the gene coding for possum TNFa. This recombinant BCG expresses the TNFa protein on the cell surface. A series of studies have been undertaken to measure the immune responses induced by the BCG-TNFa strain and determine its efficacy for protecting possums against tuberculosis.

Virulence of the recombinant BCG

Possums were inoculated with BCG-TNFa to determine the virulence of the strain and persistence of the bacilli in vivo. The induction of immune responses by BCG-TNFa was compared with wild-type BCG. Two groups of possums (n=8) were inoculated intratracheally with 10⁶CFU of either BCG or BCG-TNFa. Four animals from each group were sacrificed at 3 weeks and 4 animals at 6 weeks. Lungs were examined for pathology and samples taken for bacteriology and histochemistry and the animals were bled at 0, 3 and 6 weeks to measure the immune responses to mycobacterial antigens using a lymphocyte proliferation assay (LPA). Macroscopic lesions were observed in the lungs of all the BCG-inoculated possums sacrificed at 3 weeks with fewer lesions observed in lungs of the BCG-TNFa-inoculated possums at the same time point. Histologically, possums inoculated intratracheally with BCG had a marked consolidation of some lung lobes at 3 weeks post-inoculation. The lung alveoli were filled with a mononuclear cellular exudate, consisting of macrophages and lymphocytes with a few neutrophils and plasma cells. Possums inoculated with the recombinant BCG-TNFa had a milder consolidation with few cells infiltrating into the alveoli. For possums killed at 6 weeks post-inoculation, there was very little lung consolidation for either the BCG-inoculated or BCG-TNFa-inoculated animals. BCG was isolated from the lungs of all BCG-inoculated possums sacrificed at 3 weeks (mean number of bacilli in lungs was 135 CFU/g tissue) and from the lungs of 1/4 animals at 6 weeks (50 CFU/g tissue). No mycobacteria were isolated from the lungs of possums inoculated with BCG-TNFa.

Both the BCG and BCG-TNFa strains induced strong lymphocyte proliferation responses to bovine PPD at 3 weeks after inoculation with no significant differences between the strains at 3 and 6 weeks.

This experiment has shown that BCG-TNFa is not more virulent in possums than BCG. The replication of this train may in fact be restricted in comparison with BCG, although both BCG-TNFa and BCG induced comparable immune responses particularly at 3 weeks post-inoculation.

Growth of BCG-TNFa in macrophages

The ability of the BCG-TNFa strain to grow in possum lung macrophage cultures was assessed and compared with BCG. Lung alveolar macrophages were obtained from possums and infected with BCG, BCG-TNFa or a virulent strain of M. bovis at a multiplicity of infection (MOI) of 5:1 or 2.5:1. The growth of strains within macrophages was assessed by uptake of ³H-uracil (metabolic labelling of mycobacteria) and expressed as a percentage of the amount of labelling of bacteria when grown in mycobacterial growth media (TAB). The growth of both BCG and BCG-TNFa within macrophages was suppressed in comparison to the rapid growth of virulent M. bovis. There were no significant differences in intracellular growth between the BCG-TNFa and BCG strains suggesting that the ability of the recombinant BCG to survive in macrophage is not impaired compared to BCG.

Efficacy of BCG- TNFa as a vaccine strain

An animal experiment was performed to determine the effectiveness of BCG- TNFa for inducing protection in possums against challenge with virulent M. bovis. Two groups of possums (n=6) were immunised subcutaneously with 10⁶CFU BCG or 10⁶CFU BCG-TNFa and a third group was non-vaccinated. Immune responses were measured by LPA and possums were challenged at 8 weeks with a low dose of virulent M. bovis administered by aerosol.

Vaccination of possums with BCG-TNFa induced strong proliferative responses to bovine PPD by 4 weeks after vaccination and these responses were not significantly different to responses induced by BCG.

Both BCG and BCG-TNFa induced significant protection against challenge with virulent M. bovis. Animals in both groups lost significantly less body weight compared to the non-vaccinated animals (Table 5). The mean spleen bacterial counts in both vaccinated groups were lower than counts in the non-vaccinated group, although this difference was only significant for the BCG groups.

Table 5. Pathological and microbiological findings at sacrifice

Vaccine group	Change in body weight/challenge weight	Lung weight/PM body weight^a	Lung bacterial count^b	Spleen bacterial count^b
BCG-TNF	-0.015+0.029*	1.630+0.201	5.695+0.193	1.577+0.391
BCG	-0.012+0.055*	2.007+0.386	5.384+0.427	1.270+0.309*
Non-Vaccinated	-0.147+0.048	2.410+0.350	6.048+0.166	2.553+0.465

^a Lung weight (g)/body weight at post mortem (g) x 100

^b Bacterial count, CFU log₁₀/g of tissue

· Means significantly less than means of non-vaccinated group (P<0.05)

The recombinant form of BCG which expresses possum TNFa on the cell surface appeared to be less virulent than BCG and was as effective at inducing immune responses to mycobacterial antigens in possums. The level of protection against tuberculosis induced by the BCG-TNFa vaccine strain was similar to that achieved by BCG. Work is underway to produce a recombinant BCG which secretes the TNFa into the extracellular milieu and future experiments will measure the immune responses induced by this strain.

Encapsulation of BCG vaccine in lipid for oral delivery to possums.

BCG vaccine administrated orally or intragastically to possums does not consistently induce protection against challenge with M. bovis. It is important to determine a method of protecting the BCG from degradation in the stomach of the possum as well as protection the BCG vaccine from deterioration in the environment prior to the possums eating the vaccine bait. Frank Aldwell from Otago University has developed a technique for encapsulating a broth culture of BCG in lipid. When BCG was encapsulated in lipid, there was minimal drop in viability over a 3 month period at 4^oC. Otago University studies in mice have shown that the oral administration of BCG in lipid induced strong IFNg and IL-2 responses to bovine PPD and a significant level of protection against an aerosol challenge with M. bovis. Frank Aldwell has now filed for a provisional patent on the lipid encapsulation of live mycobacterial vaccines. A series of experiments were undertaken in possums during the past year to determine the efficacy of BCG vaccine in lipid, the effect of using different types of lipids for encapsulation of BCG and a dose response study.

Efficacy of BCG vaccine encapsulated in lipid

Twenty-one adult possums were trapped in the Upper Hutt region and were housed individually in cages at AgResearch Wallaceville. Groups of possums (5/group) were fed with either 10⁸ CFU of BCG mixed in lipid (1 g pellet), 10⁸ CFU of BCG + 10⁸ CFU of killed Mycobacterium vaccae mixed in lipid (1g pellet) or 10⁸ CFU of BCG mixed in 1 g of jam. Previous studies have shown that BCG did not lose viability when stored in jam for 24 hours. The possums were fed the same vaccine bait on two consecutive days. A fourth group of six possums were fed 1 g of lipid on two consecutive days to serve as a control. All the lipid preparations contained a small quantity of anise oil which acted as an attractant for the possums and all animals readily ate the baits usually within 3-4 hours. The possums were challenged with a low aerosol dose of virulent M. bovis in an aerosol generating chamber at 7 weeks after vaccination and were killed and examined for tuberculous lesions at 8 weeks after challenge. Blood samples were collected at 0, 4 and 6 weeks after vaccination and at 4 weeks after challenge.

Oral vaccination of possums with BCG in lipid induced peripheral blood lymphocyte proliferation responses to bovine PPD at 6 weeks after vaccination (Figure 4). The addition of killed M. vaccae to the BCG did not enhance the immune responses. The level of protection against challenge with M. bovis is shown in Table 6. Possums vaccinated with BCG in lipid with or without M. vaccae had significantly smaller lung lesions compared to that for possums fed control lipid as indicated by lower mean lung weight/body weight at post mortem as well as a significant reduction in the mean bacterial count from the lungs compared to the control group. Only the possums fed BCG in lipid had a significant reduction in the spleen bacterial count. Possums fed BCG in jam showed no indication of protection from challenge with M. bovis indicating that the lipid encapsulation was important in protecting the BCG and enhancing vaccine efficacy. In a previous study killed M. vaccae administered intranasally together with live BCG significantly enhanced the effectiveness of BCG compared to BCG or M. vaccae alone. However, in the current study, when killed M. vaccae was added to BCG in lipid and administered orally, the efficacy was not enhanced.

Table 6. Pathological and microbiological findings for vaccinated possums challenged with M. bovis

Vaccine group	Change in body weight/challenge weight^a	Lung weight/PM body weight^b	Lung bacterial count^c	Spleen bacterial count^c
BCG/lipid	0.007(+ 0.044)	1.61*(+0.27)	6.085*(+0.438)	2.058*(+0.291)
BCG jam	-0.118(+0.078)	2.87(+0.38)	6.558(+0.175)	3.053(+0.543)
BCG + M. vaccae.lipid	-0.119(+0.067)	1.60*(+0.25)	6.144*(+0.217)	2.664(+0.462)
Control lipid	-0.087(+0.050)	3.00(+0.49)	6.956(+0.055)	3.685(+0.379)

^a Change in body weight between post mortem and challenge (g)/body weight at challenge (g).

^b Lung weight (g)/body weight at post mortem (g) x 100.

^c Bacterial count, CFU log₁₀/g of tissue.

* Significantly different to non-vaccinated group.

Comparison of different types of lipid for the encapsulation of BCG.

Four different types of lipid were used for the encapsulation of BCG (lipids C, K, N and F). These lipids were of plant or animal origin. Four groups of possums (6/group) were fed a 1 g pellet containing BCG (10⁸ CFU) mixed in either lipid C, K, N or F. The possums were fed the same vaccine bait on two consecutive days. A fifth group (6/group) was vaccinated with 10⁶CFU of BCG subcutaneously and a further 6 possums were not vaccinated to serve as controls. The possums were challenged with a low aerosol dose of virulent M. bovis in an aerosol generating chamber at 8 weeks after vaccination and were killed and examined for tuberculosis lesions at 8 weeks after challenge. Blood samples were collected at 0, 4 and 6 weeks after vaccination and at 4 weeks after challenge.

Possums vaccinated subcutaneously with BCG produced strong lymphocyte proliferation responses to bovine PPD by 4 weeks after vaccination. IN contrast, the proliferation responses of possums orally vaccinated with BCG in lipid C, K or F did not increase until 6 weeks after vaccination, however, by 8 weeks after these responses from these groups were only slightly lower than those from the subcutaneous BCG-vaccinated group. Possums vaccinated with BCG in lipid N and the non-vaccinated animals did not produce any proliferation responses to bovine PPD until after challenge with M. bovis.

The level of protection against challenge with M. bovis is shown in Table 7. Possums orally vaccinated with BCG in lipid C, K and F had a significantly lower mean spleen bacterial count than that for the non-vaccinated group (P<0.05). These three BCG lipid groups also had lower mean lung bacterial counts, change in body weight after challenge/challenge body weight and lung weight/body weight at post mortem, but these differences were not significantly different. Possums subcutaneously vaccinated with BCG had significantly lower lung bacterial counts and change in body weight after challenge/challenge body weight (P<0.05).

Table 7 Pathological and microbiological findings for vaccinated possums challenged with M. bovis

Vaccine group	Change in body weight/challenge weight	Lung weight/PM body weight^b	Lung bacterial count^c	Spleen bacterial count^c
Lipid C	-0.060(+0.033)	2.254(+0.263)	5.634(+0.385)	1.301*(+0.373)
Lipid K	-0.067(+0.030)	1.549(+0.212)	5.481(+0.428)	1.321*(+0.289)
Lipid N	-0.122(+0.037)	2.347(+0.285)	6.038(+0.273)	2.200(+0.498)
Lipid F	-0.035(+0.031)	1.628(+0.411)	5.342(+0.290)	0.934*(+0.230)
BC G (s/c)	-0.012*(+0.055)	2.007(+0.386)	5.384(+0.427)	1.270*(+0.309)
Non-vacc.	-0.147(+0.048)	2.410(+0.350)	6.048(+0.166)	2.553(+0.465)

^a Change in body weight between post mortem and challenge (kg)/body weight at challenge (kg).

^b Lung weight (g)/body weight at post mortem (g) x 100.

^c Bacterial count, CFU log₁₀/g of tissue.

* Significantly different to non-vaccinated group (P<0.05).

In summary, BCG encapsulated in three of the lipid preparations (C, K and F) induced protection against M. bovis infection, which was similar to that induced by subcutaneous vaccination with BCG.

Dose response of BCG for oral bait vaccines.

Considerable progress has been made in establishing a method for encapsulation of BCG for oral bait vaccine. It is now important to determine the optical dose of BCG to be incorporated in these vaccines. BCG was formulated in lipid K at doses of 10⁶, 10⁷ and 10⁸ CFU per 1g pellet (low, medium and high doses, respectively). Groups of 6 possums were fed one pellet of one of the three doses of BCG and repeated next day. The baits were formulated by Frank Aldwell, but in contrast to the previous studies the formulations were undertaken at AgResearch Wallaceville rather than at Otago University. At 6 and 9 weeks after vaccination only 1-2 animals from the medium and high dose groups and no animals from the low dose group that produced lymphocyte proliferation response to bovine PPD of >4. The animals were challenged with an aerosol dose of M.bovis at 9 weeks after vaccination and were killed 7-8 weeks later. None of the BCG vaccinated groups showed a significant level of protection against challenge with M. bovis.

There are a number of possible reasons for the failure of oral BCG to protect in the current trial. Most of these relate to standardisation of methodology between laboratories. In particular, the stringent conditions (particularly temperature) required for formulating BCG in lipids. Steps have now been implemented to overcome these problems. Plans are in place to repeat the dose response study in possums using a modified lipid C which has been demonstrated to induce more consistent immune responses in mice and possums.

Production of recombinant possum transforming growth factor beta (TGF-b) for use as an adjuvant for mucosal immunity in possums.

Transforming growth factor-beta 1 (TGF-b) has been shown to stimulate antibody production and promote 1gA switching at mucosal surfaces. This cytokine may be a useful adjuvant for stimulation of antibody responses to antigens used for biocontrol of possums. The gene encoding possum TGF-b had already been identified and the cDNA for the mature protein (alanine²⁷⁰ —serine³⁹⁰) cloned into the E.coli expression vector pET-22b. Recombinant protein was recovered from IPTG-induced culture in inclusion bodies and purified using immobilised metal affinity chromatography. Biological activity of the purified recombinant cytokine was measured using a bioassay for TGF-b which measures TGF-b induced inhibition of mink lung cells. Attempts to refold the TGF-b from the inclusion bodies and produce a biologically active protein were not successful. A tightly regulated strain of BL21 E. coli, BL2(DE3)pLysS, was used to express the TGF-b and the culture conditions during the induction phase were modified in order to maximise the possibility of producing recombinant cytokine in a soluble form. As before, the cytokine was only produced in an insoluble form in inclusion bodies.

Work is underway to express the possum TGF-b in the yeast Pichia pastoris, an expression host which we previously used to produce recombinant proteins in a secreted form. The cDNA encoding the mature TGF-b coding region was cloned into the EcoR1 and XbaI sites of the P.pastoris expression vector pPICZa A. Provided soluble active TGF-b is expressed from the yeast the TGF-b will be tested in possums for ability to enhance antibody responses to a model antigen or an antigen relevant to biocontrol.

Location of T cells in Peyer’s patches and mesenteric lymph nodes after oral immunisation with BCG.

A protocol was optimised to identify lymphocytes in possum Peyer’s patches and mesenteric lymph nodes by immunohistochemistry. The best enzyme digestion system for antigen retrieval was identified, an appropriate negative control antibody was found and both primary and secondary antibodies were titrated to ensure low background staining. The staining of possum B cells using rabbit anti-possum Ig was unsuccessful and another antibody has now been identified. T cells could be clearly identified in the appropriate areas using a human anti-CD3 antibody. Peyer’s patches and mesenteric lymph nodes were collected from 6 possums immunised either 3 or 6 weeks previously with oral BCG formulated in lipid. Peyer’s patches and mesenteric lymph nodes were also collected from 3 non-vaccinated possums. Sections were stained for T cells. No differences in the staining of T cell areas in sections of Peyer’s patches and mesenteric lymph nodes from the BCG vaccinated animals compared to non-vaccinated animals was apparent. Other antibodies have been identified which may be useful for studying gut associated lymphoid cells in the possum.

Combined vaccine for biocontrol and tuberculosis-development of model system.

A model system is being developed to determine whether BCG can be an appropriate vehicle for delivering an immunosterilising agent to possums. A recombinant BCG strain was constructed which expressed the Bol antigen of the sheep louse. This antigen was expressed on the surface of BCG but not secreted. The rationale for using this type of recombinant BCG was that the Bol antigen and both adjuvant and vaccine components of BCG would be delivered to the same antigen presenting cell. Four possums were immunised and boosted intranasally with the recombinant BCG but there was no evidence from serology or lymphocyte proliferation assays that an immune response to the Bol antigen had been induced. There were immune responses to mycobacterial antigens. Work is in progress to construct a recombinant strain of BCG which secretes the Bol antigen.

PUBLICATIONS

Referred Journals

Cockle, P.J., Gordon, S.V., Lalvani, A., Buddle, B.M., Hewinson, R.G. and Vordermeier, H.M. 2002, Identification of novel Mycobacterium tuberculosis antigens with potential as diagnostic reagents or sub-unit vaccine candidates by comparative genomics. Infect. Immun, (submitted).

Doolin, E.E., Wedlock, D.N., Buddle, B.M. and Skinner, M.A. 2002. Mucosal immunity in the brushtail possum (Trichosurus vulpecula). Detection of antibody in serum and at female reproductive sites after intranasal immunisation. Immunol. Cell. Biol. (in press)

Corner, L.A., Buddle, B.M. and Morris, R.S. 2002. Infection of the brushtail possum (Trichosurus vulpecula) with bovine tuberculosis by the conjunctival route. Vet. J. (in press).

Corner L.A., Pfeiffer, D.U., Buddle, B.M. de Lisle, G.W. and Morris, R.S. 2002. Natural transmission of bovine tuberculosis in captive brushtail possums. N.Z. Vet. J. (in press).

Corner, L.A., Norton, S., Buddle, B.M., and Morris, R.S. 2002. The efficacy of bacilli Calmette-Guerin vaccine in wild brushtail possums (Trichosurus vulpecula) Res. Vet. Sci. (in press)

Buddle, B.M., Skinner, M.A., Wedlock, D.N., Collins,D.M. and de Lisle, G.W. 2002. New generation vaccines and delivery systems for control of bovine tuberculosis in cattle and wildlife. Vet. Immunol. Immunopathol (in press).

Young, S., O’Donnell, M., Lockhart, E., Buddle, B., Slobbe, L., Lyo, Yi and Buchan G. 2002. Manipulation of immune responses to Mycobacterium bovis by vaccination with IL-2 and IL-l8 secreting recombinant baccillus Calmette Guerin. Immunol. Cell Biol. (in press).

Buddle, B.M., Wards, B.J., Aldwell, F.E., Collins, D.M. and de Lisle, G.W., 2002. Influence of sensitisation to environmental mycobacteria on subsequent vaccination against bovine tuberculosis. Vaccine 20: 1126-1133.

Wedlock, D.N., Buddle, B.M., Skinner, M.A. and G.W. de Lisle. 2002. Mycobacterium bovis infection implications for human populations and progress in eradication of bovine tuberculosis. Microbes and Infection 4: 471-480.

Wedlock, D.N., Keen, D.L., Aldwell, F.E., Anderson, P. and Buddle, B.M. 2002. Effect of adjuvants on immune responses of cattle vaccinated with culture filtrate proteins from Mycobacterium tuberculosis. Vet. Immunol. Immunopathol 86:79-88.

Corner, L.A.L., Buddle, B.M., Pfeiffer, D.U. and Morris, R.S. 2002. Vaccination of the brushtail possum (Trichosurus vulpecula) against Mycobacterium bovis infection with bacille Calmette-Guerin: the response to multiple doses. Vet. Microbiol. 84:327-336.

Skinner, M.A., Keen, D.L., Parlane, N.A., Yates, G.F, and Buddle, B.M., 2002. Increased protection against bovine tuberculosis in the brushtail possum (Trichosurus vulpecula) when BCG is administered with killed Mycobaterium vaccae. Tuberculosis 82:15-22.

Corner, L.A.L., Buddle, B.M., Pfeiffer, D.U. and Morris, R.S. 2001. Aerosol vaccination of the brushtail possum (Trichosurus vulpecula) with bacille Calmette-Guerin: the duration of protection. Vet. Microbiol. 81: 181-191.

Doolin, E.E., Midwinter, R.G., McCarthy, A.R. and Buddle, B.M. 2001. Purification of secretory immunoglobulin A from milk of the brushtail possum (Trichosurus vulpecula). N.Z. Vet. J. 49: 181-186.

Abstracts at conferences and general publications

Buddle, B.M., Skinner, M.A., Wedlock, D.N. de Lisle, G.W. Vordermeier, H.M. and Hewinson, R.G. 2002. DNA prime/BCG boost vaccination strategy for enhanced protection of cattle against bovine tuberculosis. 4^th World Congress on Tuberculosis Washington, DC USA, 3-5 June 2002.

Hewinson, G., Vordermeier, M and Buddle, B. 2002. Use of the bovine model of tuberculosis for vaccine development. 4^th World Congress on Tuberculosis, Washington, DC USA, 3-5 June 2002.

Aldwell, F.E., Tucker, I.G. and Buddle, B.M. 2002. Oral delivery of BCG in a lipid formulation enhances immune responses to tuberculosis. The Experimental Biology meeting, New Orleans, USA, 21-25 April, 2002.

Parlane, N. 2002. Bovine interferon-g, the inside story. NZ Flow Cytometry meeting, Auckland, 8 April 2002.

Buddle, B.M. Skinner, M.A., Wedlock, D.N., de Lisle G.W., Vordermeier H.M. and Hewinson, R.G. 2002. DNA prime./BCG boost vaccination strategy for enhanced protection of cattle against bovine tuberculosis. Formulation and Delivery of Bioactives 54^th Annual Conference, February 20-21, Dunedin, 2002.

Skinner, M.A., Wedlock, D.N. and Buddle, B.M. 2002. CpG DNA as an adjuvant component for new generation tuberculosis vaccines. Formulation and Delivery of Bioactives 5^th Annual Conference, February 20-21, Dunedin, 2002.

Skinner, M.A., Wedlock, D.N., Doolin, E.E., Hamel, K., Mahoney, T., McLeod, B. and Buddle, B.M. 2002. Stimulation of mucosal immune responses in the possum. In: Biological Management of Possums. Ministry of Agriculture and Fisheries, Wellington, New Zealand, pp40-45.

McCarthy, A.M., Ryan, T.J., de Lisle, G.W. and Buddle, B.M. 2001. A highly specific blood test for the diagnosis of bovine tuberculosis in cattle. N.Z. Microbiological Society. 10-13 December 2001, Wellington.

Skinner, M. 2001. Prime-boost strategies for vaccination against bovine tuberculosis. 31^st Scientific Meeting of the Australasian Society for Immunology, Canberra, 2-6 December, 2001.

Buddle, B.M., Doolin, E., Wedlock, D.N., Skinner, M.A., Collins, D.M., de Lisle, G.W. 2001. Development of vaccines for the control of bovine tuberculosis in cattle and wildlife. NZBA meeting, Napier, December 4-7, 2001.

Buddle, B.M., Skinner, M.A., Wedlock, D.N. 2001: Improved the efficiency of BCG vaccination for neonates. Sequella Global Tuberculosis Foundation TB vaccine conference, Montreal, Quebec, Canada, November 7-9, 2001.

Buddle, B.M., Skinner, M.A., Wedlock, D.N., Collins, D.M., and de Lisle, G.W. 2001. Vaccination of cattle against bovine tuberculosis. The Animal Health Board’s Vaccine Workshop, Victoria University, Wellington, 7-8 August, 2001.

Buddle, B.M., Skinner, M.A., Wedlock, D.N., Collins, D.M., and de Lisle, G.W. 2001. AgResearch Wallaceville studies on vaccination of possums against bovine tuberculosis. The Animal Health Board’s Vaccine Workshop, Victoria University, Wellington, 7-8 August, 2001.

Buddle, B.M., M.A. Skinner, D.N. Wedlock, D.M. Collins, L.A. Corner and G.W. de Lisle 2001. New generation vaccines and delivery systems for control of bovine tuberculosis in cattle and wildlife. 6^th International Veterinary Immunology Symposium, Uppsala, Sweden. 14-20 July, 2001.

Wedlock, D.N., D. L. Keen, A.R. McCarthy, P. Andersen and B.M. Buddle, 2001. Effect of adjuvants on immune responses of cattle vaccinated with culture filtrate proteins from Mycobacterium tuberculosis. 6^th International Veterinary Immunology Symposium, Uppsala, Sweden, 14-20 July, 2001.

Rich, F. 2001. A prime/boost strategy for vaccination against bovine tuberculosis. Immunet meeting, Wellington 2-3 July 2001.

Skinner, M.A. 2001. New generation vaccines for bovine tuberculosis in cattle and possums. Immunet meeting, Wellington, 2-3 July, 2001.

Contact for Enquiries

Farm Monitoring Programme Manager
Monitoring and Evaluation
MAF Policy
PO Box 2526
Wellington
NEW ZEALAND
Phone: +64 4 894 0623
Fax: +64 4 894 0741
Contact this person

5. Possum Biocontrol Category

5.1 PBC 220

Population responses of wild possum to fertility control

Summary

Background

Approach & Outcomes

5.2 PBC 221

Blocking Embryonic Development in Brushtail Possums

Summary

Background

Approach & Outcomes

PUBLICATIONS

5.3 PBC 222

To isolate, characterise and genetically manipulate possum viruses for the biological control of possums

Summary

Background

Approach & Outcomes

Publications

5.4 PBC 223

Control of reproduction in possums by targeting the pituitary gland through the use of GNRH-toxin conjugates

Summary

Background

Objective 1 – Production And Testing Of Gnrh-Pap

Approach & Outcomes

Table 1 Concentration at which quinones become non-toxic. Average of 3 trials

Cytotoxicity of a GnRH-quinone conjugate in immortal cell lines

Publications

5.5 PBC 224

Control of reproduction in possums by targeting the oocyte-specific growth factors, GDF-9 and GDF-9B

Summary

Approach & Outcomes

OBJECTIVE 1 – IMMUNISATION OF POSSUMS AGAINST GDF-9 AND GDF-9B

PUBLICATIONS

5.6 PBC 225

Controlled oestrus and synchronised breeding in possums to facilitate the testing of the new biocontrol agents

Summary

Background

Approach & Outcomes

Achievement of Milestones

5.7 PBC 226

Oral delivery of Bioactives to Possums

Summary

Background

Approach & Outcomes

PUBLICATIONS

5.8 PBC 227

Transgenic worms for the control of possums

Summary

Background

Approach & Outcomes

5.9 PBC 228

Bovine TB

Summary

Background

Approach & Outcomes

PUBLICATIONS

Contact for Enquiries

Search Rural NZ

Keyword Search