General enquiries on this form should be made to:Defra, Science Directorate, Management Support and Finance Team,Telephone No. 020 7238 1612E-mail: research.competitions@defra.gsi.gov.uk

SID 5 Research Project Final Report

SID 5 (Rev. 3/06) Page 1 of 18

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code OZ0319

2. Project title

Salmonella pathogenesis in cattle and pigs.

3. Contractororganisation(s)

Institute for Animal HealthComptonNewburyBerkshireRG20 7NN     

54. Total Defra project costs £ 709,200(agreed fixed price)

5. Project: start date................ 01 July 2002

end date................. 30 September 2006

SID 5 (Rev. 3/06) Page 2 of 18

6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They

should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.

Salmonella enterica is an enteric bacterial pathogen of worldwide importance. Over 2500 distinct serovars of S. enterica exist and these can be divided into three broad types; ubiquitous serovars that can cause acute but self-limiting gastroenteritis in a broad range of hosts (for example, S. Typhimurium), host-restricted serovars that exhibit a preference for a narrower range of hosts (e.g. S. Dublin and S. Choleraesuis) and host-specific serovars that produce disease in only one host (e.g. Typhoid fever caused by S. Typhi in humans). Host-restricted and -specific serovars often cause only mild enteritis but possess the ability to disseminate from the intestines to the organs of the body, causing typhoid fever-like illnesses in animals that may be severe and life-threatening. Non-typhoidal Salmonellosis caused by ubiquitous serovars also poses serious welfare and economic problems for livestock producers. Moreover, Salmonella infection of pigs and cattle can lead to entry of the pathogen into the food chain and environment, providing an important reservoir of human infection. The molecular mechanisms underlying the ability of Salmonella enterica to colonise the intestines of food-producing animals, induce enteritis and in some cases translocate to distal sites, are poorly understood. Defra-funded research conducted under project OZ0319 aimed to understand these processes in greater detail and to use the knowledge to develop, test and refine novel methods of disease control. Such research has yielded insights that could not have been obtained in surrogate rodent or cell-based assays and led to the testing of subunit vaccines in pigs, live-attenuated vaccines in calves and analysis of the protective efficacy of small molecule inhibitors that disarm a key Salmonella virulence factor. Key scientific achievements are summarised below:

1. Identification of a portfolio of over 200 S. enterica serovar Typhimiurium genes required for colonisation of cattle and pigs by signature-tagged transposon mutagenesis (Morgan et al., 2004; Rowe et al., submitted). Factors that play conserved roles in colonisation of calves, pigs and chickens (e.g. Type III secretion systems), as well as host-specific colonisation factors (SPI-4 in calves; ccf in chickens) were identified, providing key information for vaccine design. This represents the most comprehensive survey of its kind and laid the foundations for a BBSRC-funded project to assign roles for every transposable S. Typhimurium gene in calves, pigs and chickens by a related method (D017556, commenced 1 October 2006).

2. Analysis of the basis of attenuation of mutant strains in vitro and in animals, including characterisation of several key virulence factors (Salmonella pathogenicity island-4, Type IV pili, ccf, fimbriae and ShdA)

3. Measurement of net growth in the porcine intestines of different S. enterica serovars and demonstration that systemic virulence of S. Choleraesuis correlates with increased persistence in intestinal lymph nodes and weaker induction of host innate immune responses compared to S. Typhimurium.

4. Demonstration that S. enterica rapidly clears from the enterocyte layer following infection of calves and associates with MHC-class II positive cells that have features of dendritic cells, but not macrophages. This tropism was found to be shared by serovars that differ in virulence.

5. Demonstration that the expression a key Salmonella virulence determinant (Type III secretion system-1) occurs at a lower level in S. Choleraesuis than S. Typhimurium, consistent with the magnitude of intestinal inflammatory and secretory responses induced by these serovars in animals. Other differences in the

SID 5 (Rev. 3/06) Page 3 of 18

repertoire, sequence and expression of key loci that may explain the differential virulence of S. enterica serovars were also identified.

6. Demonstration that the safety of live-attenuated Salmonella vaccines can be improved by introduction of mutations that reduce the activity of Type III secretion system. The resulting strains induced less enteritis in a bovine ligated intestinal loop model, implying that they are less likely to induce diarrhoea than the parent strains during animal and human use. AttenuatedsipAsseC mutant strains of S. Typhimurium and S. Dublin were well tolerated after oral infection of calves and conferred serovar-specific protection against Salmonella-induced enteritis.

7. Demonstration that a vaccine comprising Salmonella secreted proteins can reduce faecal excretion of S. Typhimurium in the acute stage of infection in pigs in a manner independent of Type III secreted proteins.

8. Demonstration that salicylanilide derivatives specifically inhibit the expression and secretion of Type III secretion system-1 proteins and can reduce bacterial invasion of cultured cells and the induction of enteritis in calves.

The project yielded high-impact publications and provided excellent training for the Postdoctoral Scientist and Research Technician involved. The skills instilled in these researchers were also put to use to define the mode and genetics of systemic translocation of S. Dublin in calves using a novel cannulation model and to examine the impact of a host stress hormone on the outcome of Salmonella infection. In this way, Defra project OZ0319 added value to other ongoing projects and created and maintained a nucleus of experts able to work at all levels from molecules to whole animals. The subsequent withdrawal of Defra support for research of this kind at IAH will have a detrimental impact on the UK science base in this area.

Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with

details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).

Project management.Project OZ0319 was conceived by Dr Tim Wallis and managed by him until 1 June 2005, when he voluntarily

left the Institute for Animal Health. Dr Stevens took responsibility for the Enteric Bacterial Pathogens laboratory at this time and prepared the objectives for a 1 year extension of the project, which ended on 30 September 2006. This report has been prepared by Dr Stevens without the input of Dr Wallis, but in consultation with the researcher’s involved and following inspection of annual interim reports and laboratory records. Progress against each of the scientific objectives of project OZ0319 is summarised below:

01/01. Assess non-colonising mutants for adhesion to epithelial cells in vitro.

By screening of signature-tagged S. Typhimurium mutants in calves, pigs and chickens we identified a locus encoding a putative Type IV pilus which appears to be required for host-specific colonisation of cattle. The pilV gene encoding the pilus adhesin is able to undergo recombination such that the resulting PilV proteins comprise a constant N-terminus and one of four possible C-termini. We constructed defined non-polar deletion mutants lacking the rci recombinase gene in which the pilV gene was locked in each of the four possible orientations. These mutants, alongside a pilO mutant, were compared with the parent strain for their ability to adhere to cultured Int-407 intestinal epithelial cells. No differences in the numbers of bacteria adhering to the cells were found. It remains possible that such pili facilitate colonisation by mediating inter-bacterial interactions (as with the bundle-forming pili of enteropathogenic E. coli) and/or attachment to epithelial surfaces in vivo.

01/02. Assess non-colonising mutants for invasion to epithelial cells in vitro.

SID 5 (Rev. 3/06) Page 4 of 18

Over 80 signature-tagged transposon mutants which were known to be impaired in intestinal colonisation of calves, pigs or chickens were screened for their ability to invade Int-407 cells in an attempt to define the basis of the attenuation. Only mutants with transposon insertions in components of Type III secretion system-1 (T3SS-1) or secreted substrates of this apparatus exhibited reduced invasion of Int-407 cells, suggesting that the non-T3SS-1 genes may contribute to pathogenesis in a manner other than primary invasion of the intestinal epithelium from which output pools were recovered. In addition, mutants with transposon insertions in the chicken-specific locus ccfA-D (see milestone 01/04) were compared with the parent strain for their ability to invade primary chick kidney cells and survive inside chicken (HD11) or murine (J774A.1) macrophage-like cells. No statistically significant differences were found.

01/03. Assess colonisation of pilS and SPI-4 mutants following single challenge of calves.

We constructed a defined S. Typhimurium non-polar deletion mutant lacking pilO rather than pilS, since this is predicted to result in a complete absence of Type IV pilus assembly. The mutant and parent strain were separately inoculated into duplicate calves and the course of faecal excretion and severity of infection assessed. Although the pilO mutant was shed in lower numbers than the parent strain, induced weaker pyrexial responses and was recovered in lower numbers from intestinal tissues at post mortem examination, none of the differences proved to be statistically significant. A manuscript reporting these data, together with the construction and adherence phenotypes of locked PilV variants isolated after mutation of the Rci recombinase has been prepared for publication.

Defined S. Typhimurium non-polar deletion mutants lacking the SPI-4 genes siiE (putative large secreted protein) and siiF (putative inner membrane ATPase required for secretion of SiiE) were constructed using positive-selection suicide replicons and the mutants verified in vitro. The siiE, siiF and parent strains were each separately inoculated into 4 calves. Weak, but statistically significant, attenuation of both the siiE and siiF mutant was observed, with a greater reduction in colonisation of the ileum being detected relative to the wild-type compared to other intestinal sites. The possibility that SPI-4 may confer a tropism for ileal tissue in calves would be consistent with the results of screening signature-tagged S. Typhimurium mutants in calves, where multiple SPI-4 mutants were absent or poorly represented in the ileal wall (Morgan et al., 2004). Given the important role played by SiiE, a C-terminal fragment of the protein was cloned, expressed and affinity purified and used to raise monoclonal antibodies following immunisation of mice. The antibody specifically recognised a ca. 600kDa protein in the supernatant fraction of S. Typhimurium cultured in Luria Bertani medium, indicating that the siiE open reading frame is translated into a single large protein as predicted from the SPI-4 sequence that we generated (Morgan et al., 2004). Secretion of SiiE was found to rely on SiiF, since the siiF mutant accumulated intracellular SiiE but do not secrete the protein into the culture supernatant, consistent with the proposed function of SiiF as a component of a Type I secretion system (Fig. 1). Production of SiiE was also dependent on HilA, the master regulator of the SPI-1 locus which encodes T3SS-1 suggesting that SPI-1 and SPI-4 genes may be co-ordinately regulated. These data, together with analysis of the role of SPI-4-encoded genes in Salmonella invasion and the induction of enteritis (milestone 02/01) have been submitted for publication.

Fig. 1. Secretion of the ca. 600kDa SPI-4-encoded protein SiiE relies on SiiF. The protein was detected in total and secreted fractions of S. Typhimurium using an SDS-PAGE system optimised for resolution of high molecular weight proteins and a monoclonal antibody raised against a purified C-terminal fragment of SiiE.

SID 5 (Rev. 3/06) Page 5 of 18

SipASipB

SipCSipD

460

Mwt WT ΔsiiE ΔsiiF

500

240

97

40

500

240

97

40

Mwt WT ΔsiiF

460Anti-SiiE

Anti-SiiE

Secreted protein Total protein

SiiE

01/04. Assess colonisation of best candidate mutants following single challenge of poultry.

Screening of signature-tagged mutants of S. Typhimurium identified a cluster of genes (STM0080, -82 and -84) that were required for colonisation of chickens but not calves (Morgan et al., 2004) or pigs (Rowe et al. Submitted). The gene cluster exhibits features associated with pathogenicity islands and the role of the encoded products was probed by P22 transduction of the transposon insertions into the archived S. Typhimurium 4/74 strain and characterisation of the mutants in vitro and in chickens. A mutant lacking an insertion in a fourth gene of the locus (STM0081) was located in the mutant library by hierarchical PCR. Groups of 15 Light Sussex chicks were separately inoculated at 14 days old with c. 2 x 108 CFU of parent or mutant strains and faecal excretion of the bacteria monitored at 1, 2 and 3 weeks post-infection by swabbing of the cloaca. All the mutant strains exhibited statistically significant reductions in their ability to colonise the avian GI tract and as a consequence the genes were-renamed ccfABCD (chicken-specific colonisation factors A-D). Mutation of ccfA-D did not impair the ability of S. Typhimurium to cause systemic disease in mice (consistent with a host-specific role), invade primary chick kidney cells, or survive inside chicken (HD11) or murine (J774A.1) macrophage-like cells. The results of this study have been submitted for publication.

Further studies supported by the Defra-funded Research Technician have probed the role of fimbrial loci in the colonisation of chickens by Salmonella Enteritidis. Analysis of the raw unannotated sequence of S. Enteritidis PT4 strain P125109 identified 13 putative fimbrial loci. The predicted major subunits of each of these loci has been deleted by lambda Red recombinase-mediated linear recombination and the mutant strains verified by PCR and the defects P22 transduced into the archived parent strain. Each mutant has been assessed for its ability to adhere to HEp-2 human epithelial cells and chicken HD11 cells cultured in vitro but no statistically significant differences were observed. Screening of the mutant strains by oral inoculation of chickens has so far revealed roles for S. Enteritidis Stb and Stc fimbriae in colonisation of the avian alimentary tract, however chromosomal repair or re-introduction of the fimbrial loci in trans has still to be performed. The role of selected S. Typhimiurium fimbriae in intestinal colonisation of chickens was also probed by analysis of tagged-transposon mutants in oral challenge experiments and the data revealed roles for Stb, Std, Sth, plasmid-encoded and curli fimbriae in colonisation of the caeca of chickens. Fimbriae are attractive candidates for the development of vaccines as they are surface-exposed and comprise polymers of many repeated subunits that mediate initial interactions with host cells.

02/01. Assess effects of pilS and SPI-4 on Salmonella induced-enteropathogenesis in duplicate ligated ileal loop assay.

The effect of mutation of pilO and the siiE and siiF genes on the ability of S. Typhimurium to elicit intestinal inflammatory and secretory responses was assessed in the bovine ligated ileal loop assay. The mutant strains induced fluid secretion and the infiltration of 111In-labelled neutrophils at levels comparable to the parent strain. As these genes were found not to contribute to enteropathogenesis but have potential bovine-specific roles in intestinal colonisation their roles in invasion of the bovine ileal mucosa were also assessed using a gentamicin-protection assay in ileal loops. Strains with mutations in pilO, pilV, rci, siiE and siiF were compared with the parent strain for recovery from ileal mucosa 2 hours post-inoculation of loops. The siiE and siiF mutants were significantly attenuated for invasion (P < 0.005) but none of the pil mutants exhibited defects in intestinal invasion.

Ligated ileal loop experiments were also performed in pigs to assess the role played by the S. Typhimurium fibronectin-binding Type V secreted protein ShdA. ShdA has been implicated in long-term faecal excretion in mice however it has been reported that it is not a pre-requisite for long-term excretion of S. Typhimurium in orally inoculated pigs. Project OZ0319 supported a two-week visit by a scientist from the University of Ghent, during which time the phenotype of the shdA mutant was tested in porcine ligated ileal loops. No statistically significant differences in the induction of inflammatory or secretory responses were detected between wild-type and S. Typhimurium and an isogenic shdA mutant (Boyen et al. 2006. Vet. Microbiol. 115:284-90).

02/02. Assess SPI-1 effector gene expression in intestinal mucosa in vivo using antibodies to SopB and SopD by immunohistocytochemisrtry in duplicate ligated ileal loop assay.

Experiments were carried out to detect SopB and SopD expression in the bovine intestines. Infected mucosa was generated by inoculation of bovine ileal loops with S. Typhimurium wild-type and Type III secretion system-1 mutant strains and examined by immunohistochemistry using monoclonal antibodies against the T3SS-1 effector proteins SopB and SopD. Despite repeated attempts no antibody staining attributable to these proteins could be detected, possibly as the quantities of these proteins produced in vivo are below the limit of detection by this method.

02/03. Assess SPI-2 effector gene expression in intestinal mucosa in vivo using SseJ reporter construct.An S. Typhimurium SseJ-HA tagged reporter construct was obtained and inoculated into bovine ileal loops as

above, however no staining attributable to the fusion protein could be detected with an antibody specific to the HA tag. As indicated in section 9 of the 2004/05 interim report, limitations on the sensitivity of detection of native or epitope-tagged effector proteins led us to focus our resources on the development of methods to quantify transcription of Type III secretion genes in tissues by real-time PCR analysis (milestone 08/01).

SID 5 (Rev. 3/06) Page 6 of 18

03/01. Assess net growth of S. Dublin, S. Gallinarum, S. Typhimurium and S. Choleraesuis in pigs by plasmid partitioning.

Previous studies in our laboratory have shown that the systemic virulence of S. enterica serovars Dublin in cattle and Choleraesuis in pigs does not correlate with the extent of invasion of the intestinal epithelium, the magnitude of inflammatory or secretory responses or enhanced ability to replicate and persist within primary macrophages compared to the enteritis-associated serovar S. Typhimurium. In order to determine the fate of different S. enterica serovars following oral inoculation we used a novel method reliant on a temperature-sensitive non-segregating plasmid (pHSG422) to quantify the net growth and killing of S. Dublin, S. Choleraesuis and S. Typhimurium in porcine intestinal mucosa. In parallel we also quantified host pro-inflammatory cytokine responses in the very same tissues as the replication of S. enterica serovars was being examined (milestone 03/05) in order that we could correlate the induction of innate responses with the fate of the bacteria. S. Dublin, S. Choleraesuis and S. Typhimurium strains harbouring pHSG442 were cultured at a temperature permissive for plasmid replication (30˚C) such that all the bacteria in the population could be expected to contain a copy of the plasmid. At body temperature the plasmid fails to replicate and thus with each cell division the plasmid is titrated out of the bacterial population owing to uneven partitioning. Thus analysis of the total number of plasmid-bearing cells as a proportion of the population provides a measure of the extent of bacterial replication in vivo. Furthermore, analysis of the total number of bacteria recovered can provide an indication of killing rates. For unknown reasons pHSG422 failed to segregate in S. Gallinarum, therefore it was not included in the analysis as originally intended. The method was first validated by analysis of the net growth in vivo of an S. Choleraesuis mutant harbouring a defect in aroA, which was known at the outset to affect bacterial persistence in vivo. The data indicate that S. Choleraesuis replicates better than its isogenic aroA mutant in all 4 tissues examined since the total number of bacteria are higher and a smaller proportion of the wild-type bacteria contain pHSG422 compared to the aroA mutant (Fig. 2).

Fig. 2. Replication rates of S. Choleraesuis pHSG422 and S. Choleraesuis aroA pHSG422 48 hours after oral inoculation of pigs. Triplicate samples were taken from each tissue and means were calculated to give a value per animal. Each bar represents the mean bacterial counts derived from 7 pigs infected with S. Choleraesuis and 3 pigs infected with S. Choleraesuis aroA and is presented with the standard error of the mean. The solid bars represent the number of bacteria carrying pHSG422 and the hashed bars represent total bacterial numbers. Triangles represent statistical differences in the replication rate between the two strains (P < 0.1, 0.05 and 0.01 for 1, 2 or 3 triangles respectively) and stars represent statistical differences in total bacterial numbers between strains (P < 0.1, 0.05, 0.01 and 0.001 for 1, 2, 3 or 4 stars respectively).

We next compared the replication rates of S. Dublin, S. Choleraesuis and S. Typhimurium strains harbouring pHSG442 following oral inoculation of pigs at varying times post-inoculation. The data for S. Choleraesuis (systemically virulent in pigs) and S. Typhimurium (gut-restricted) are shown in Fig. 3. The data indicate that S. Typhimurium (green bars) replicates more rapidly in the ileal and colonic mucosa than S. Choleraesuis (red bars) consistent with the induction of more potent pro-inflammatory cytokine responses (milestone 03/05) and the elevated enteropathogenic responses seen in these pigs and in porcine ileal loop assays. Conversely, S. Choleraeusis exhibited increased persistence in ileal and colonic mesenteric lymph nodes compared to S. Typhimurium, consistent with the systemic virulence of this strain. Rapid killing of S. Dublin was observed, with 90-99% of bacteria being eliminated from the mucosa between 14 and 48 h post-inoculation (data not shown), consistent with the low virulence of this serovar for pigs.

SID 5 (Rev. 3/06) Page 7 of 18

** *** ****

Ilealmucosa

Ilealnode

colonmucosa

colonnode

Ilealmucosa

Ilealnode

colonmucosa

colonnode

0

1

2

3

4

5

6

7

8

Recovery of Salmonella

(log10 CFU/g)

S. Choleraesuis S. Choleraesuis aroA

Fig. 3. Replication rates and bacterial numbers of S. Typhimurium pHSG422 (green) S. Choleraesuis pHSG422 (red) and up to 72 hours after oral inoculation of pigs. Triplicate samples were taken from each tissue and means were calculated to give a value per animal. Each bar represents the mean bacterial counts derived from 3-9 pigs per serotype, at each time point and is presented with the standard error of the mean. The solid bars represent the number of bacteria carrying pHSG422 and the hashed bars represent total bacterial numbers. Triangles represent statistical differences in the replication rate between the two strains (P < 0.1, 0.05 and 0.01 for 1, 2 or 3 triangles respectively) and stars represent statistical differences in total bacterial numbers between strains (P < 0.1, 0.05 and 0.01 for 1, 2 or 3 stars respectively).

03/02. Compare net growth of S. Dublin, S. Gallinarum, S. Typhimurium and S. Choleraesuis in calves by plasmid partitioning.

This is the only milestone not to have been completed. Please see SID5A section 13 and 2004/05 interim report.

03/03. Assess effects of mutations in SPI-2 and spvR on Salmonella net growth in vivo by plasmid partitioning.

Type III secretion system-2, encoded by Salmonella pathogenicity island 2 (SPI-2) has been shown by others enhance the ability of Salmonella to replicate intracellularly by delivering bacterial effector proteins into host cells that modulate the maturation and trafficking of the Salmonella-containing vacuole. We assessed the effect of disrupting TTSS-2 on the replication and survival of S. Typhimurium, S. Dublin and S. Choleraesuis in pigs. Mutants lacking structural components of the T3SS-2 apparatus (SsaJ and/or SsaU) were constructed in S. Typhimurium strain 4/74, S. Choleraesuis strain A50 and S. Dublin strain 2229. Partitioning of pHSG442 in the wild-type and mutant strains was then assessed following oral inoculation of at least 3 pigs per strain as

SID 5 (Rev. 3/06) Page 8 of 18

0

1

2

3

4

5

6

7

8

***

***

*

*

Small intestinal wall Small intestinal node

48 3624 36 48 72

Time after inoculation (h)

24 36 72 24 48 72 24 36 48 72

0

1

2

3

4

5

6

7

8

Recovery of Salmonella

(log10 CFU/g)

**

**

Colon wall Colon node

24 36 48 72

Time after inoculation (h)

24 36 72 24 48 72 24 36 48 7248 36

Recovery of Salmonella

(log10 CFU/g)

described above. Intriguingly, T3SS-2 mutations influenced replication kinetics in a serovar-specific manner. Within 24 hours of oral infection of pigs, disruption of T3SS-2 had no clear effect on the infection kinetics of S. Typhimurium, S. Choleraesuis or S. Dublin. However, by 48 hours post-infection disruption of SsaU was associated with a significant decrease in the recovery of total S. Dublin from porcine tissues, consistent with a role for this system in bacterial survival in the tissues (protection against killing). In contrast, disruption of SsaU in S. Typhimurium or SsaJ in S. Choleraesuis was not associated with loss of viable bacteria, but with reduced bacterial replication rates as assessed by analysis of the number of plasmid-bearing bacteria. This observation is consistent with a role for T3SS-2 in the replication of these serovars in porcine tissues. The common role of T3SS-2 in replication of S. Typhimurium and S. Choleraesuis indicates that it is unlikely that T3SS-2 alone dictates the differences in the net growth rate in vivo of the strains detected under milestone 03/01.

03/04. Compare net growth of S. Dublin, S.Typhimurium and S. Choleraesuis in pigs by plasmid partitioning.

This milestone was entered in error by Dr Wallis and is the same as 03/01.

03/05. Determine cytokine responses of intestinal mucosa following infection with S. Typhimurium, S. Dublin and S. Choleraesuis.

During analysis of the net growth in vivo of strains of S. enterica serovars Typhimurium, Choleraesuis and Dublin, biopsies of ileum and colon collected at post mortem examination 24, 28 or 72 h after oral inoculation were snap frozen in RNA Protect reagent for subsequent analysis of the transcription of host (03/05) and bacterial (08/01) genes. Tissues were thawed and disrupted using QiaShredder columns and total RNA extracted using RNeasy columns and subjected to on-column DNase digestion. The quality of RNA samples was determined by agarose gel electrophoreses and spectrophotometry. Real-time PCR assays for the detection of porcine TNF, IL-6, IL-8 and IL-18 were developed using primers based on the pig genome sequence designed to distinguish between products amplified from RNA (lacking introns) or contaminating genomic DNA. The sensitivity of detection of mRNAs was determined over a range of RNA concentrations, with data normalised to host 28S ribosomal RNA.

Transcript levels for the pro-inflammatory cytokines TNF, IL-8 and IL-18 were observed to be markedly elevated during infection of porcine ileal mucosa by both S. Typhimurium and S. Choleraesuis (Fig. 4). However at 24 h after oral inoculation of separate pigs with matched doses of the organisms, the pro-inflammatory cytokine responses to S. Typhimurium were significantly higher than those raised against S. Choleraesuis (Fig. 4). The data represent the mean data from at least three different pigs infected with each strain and an identical trend was observed in infected colonic tissues from the same animals (data not shown). By 48 h and 72 h after oral inoculation transcript levels for TNF, IL-8 and IL-18 were comparable in tissues infected with S. Typhimurium and S. Choleraesuis. TNF and IL-8 are key pro-inflammatory cytokines induced in intestinal epithelial cells by Salmonella and play a crucial role in the recruitment of neutrophils and macrophages to the site of infection. IL-18 also promotes inflammation, stimulates IFN production and is required for Salmonella clearance in mice. The data indicate that the potent early pro-inflammatory cytokine response to S. Typhimurium may reflect the rapid net growth of the organism in the intestinal mucosa compared to S. Choleraesuis detected under milestone 03/01. It is possible that induction of such potent early responses contributes to control of S. Typhimurium locally in the intestines, limiting the translocation of bacteria to other sites. We speculate that S. Choleraesuis induces weaker pro-inflammatory cytokine responses by virtue of its slower growth rate in the intestinal mucosa and that this may facilitate bacterial persistence in the intestines and aid its systemic dissemination by stealth. These data are being prepared for publication alongside observations of net growth in vivo of the strains used.

SID 5 (Rev. 3/06) Page 9 of 18

Fig. 4. Pro-inflammatory cytokine responses induced by S. Typhimurium strain 4/74 or S. Choleraesuis strain A50 at daily intervals following oral inoculation of groups of at least 3 pigs. Measurements were made by quantitative real-time PCR (normalised to 28S rRNA) from the same tissues as used to monitor net growth of the strains in vivo (milestone 03/01).

03/06. Identify intestinal cell populations targeted by S. Typhimurium, S. Dublin and S. Gallinarum in cattle by immunohistochemistry.

Duplicate calves were orally inoculated with strains of S. enterica serovars Typhimurium, Dublin or Gallinarum and intestinal mucosae and mesenteric lymph nodes (MLN) recovered 24 h post-inoculation for histological analysis. Samples were also collected from ligated ileal loops from duplicate calves 12 hours after they were filled with the different serovars. Bacteria were labelled with O-specific antiserum and eukaryotic cell populations separately stained with antibodies against markers of different immune cell subsets. Bound antibodies were detected with appropriate fluorochrome-conjugated secondary antibodies. In the ileal mucosa and MLN all three serotypes were associated with phenotypically similar host cell populations. In the ileum the majority of bacteria were found within the lamina propria and bacteria were only rarely found associated with the enterocyte monolayer. No bacteria were found within the lymphoid follicles associated with the Peyer’s patch tissue, the muscularis mucosa or the submucosa. All of the bacteria within the ileal lamina propria were associated with cells expressing MHCII and CD68, the latter belonging to a family of proteins expressed in macrophage-related cells (Fig. 5). In contrast there was no association of bacteria with cells labelled by antibodies recognising CC76 which is expressed only by B cells or CD14, which is mainly expressed by macrophages.

SID 5 (Rev. 3/06) Page 10 of 18

S. Typhimurium S. Choleraesuis

0

5

10

15

20

25

24h 48h 72h UI0

510

1520

25

24h 48h 72h UI

TNF

0

5

10

15

20

25

24h 48h 72h UI0

5

10

15

20

25

24h 48h 72h UI

IL-8

0

5

10

15

20

25

24h 48h 72h UI0

5

10

15

20

25

24h 48h 72h UI

IL-18

Fig. 5. Association of S. Typhimurium (red) with lamina propria MHC class II-positive cells (green) 24 hours after oral inoculation of calves. S. Dublin and S. Gallinarum strains exhibited an identical cell tropism (data not shown).

Additionally we assessed whether the presence of different serovars in the mucosa was associated with differences in the number of apoptotic cells. Nuclei containing fragmented DNA were detected by TUNEL staining using the ApopTag in situ apoptosis detection method. No differences in the proportion of apoptotic cells were noted in response to Salmonella infection compared to control mucosa. Furthermore, the proportion of TUNEL-positive cells was similar in mucosa infected with the different serovars. The observation that all three S. enterica serovars associate with a population of cells with features of phagocytic antigen-presenting cells, without an apparent increase in apoptosis, is novel and the data are being prepared for publication. No differences in host cell tropism or survival that could explain the differential virulence of the S. enterica serovars tested could be detected.

04/01. Generate aroA mutations in combination with SPI-1 and SPI-2 mutants.

Live-attenuated strains of Salmonella show promise both for the control of Salmonellosis in food-producing animals and humans, but also as carriers of heterologous antigens for the control of other diseases. A major limitation of the use of such strains is that they induce mild enteritis some vaccinees, likely as a consequence of deployment of the key enteritis mediator T3SS-1. Drs Wallis & Galyov participated in a joint EC-funded project (QLK2-CT-1999-00310) in which attenuated Salmonella strains were constructed that contained mutations which impair the function of T3SS-1 and/or T3SS-2 and/or secreted substrates of these systems. The selection of genes to be mutated was based on the knowledge accrued at IAH on the role of translocon and effector proteins in the induction of enteritis. Constructs and expertise were transferred to a consortium of researchers that constructed a panel of single and double mutants, some of which were assessed for virulence in mice (04/03) and some tested for safety and efficacy in cattle (04/02 and 04/04).

04/02. Assess enteropathogenicity of mutant strains in ligated loop assay.

Mutations that reduce (but do not abolish) the activity of Type III secretion system-1 ( invH or sipA) or impair intracellular replication by disrupting a component of Type III secretion system-2 (sseC) were introduced into S. Typhimurium. The induction of intestinal inflammatory and secretory responses against such strains was then compared in bovine ligated ileal loops (triplicate determinations in 3 separate animals) and compared to the parent strain and a live-attenuated vaccine candidate strain lacking aroA (Fig. 6).

SID 5 (Rev. 3/06) Page 11 of 18

Fluid secretion (volume/length)

0

2000

4000

6000

8000

10000

12000

LB WT

sseC aroA invH sseCinvH

sipA sseCsipA

aroAsipA in

vH

Tota

l CP

M

Neutrophil influx(Mean CPM)

Fig. 6. Secretory and inflammatory responses to aroA and sseC vaccine strains of S. Typhimurium and isogenic double mutants containing invH and/or sipA mutations in calves.

Consistent with other observations, the aroA and sseC vaccine strains induced weaker enteropathogenic responses compared to the wild-type, but still appreciably higher responses than sterile medium or a T3SS-1 mutant (Fig. 6). Indeed the secretory responses observed for the sseC mutant were not significantly different from the wild-type strain. The introduction of additional mutations in sipA or invH completely abolished the ability of the strains to induce enteritis and responses were comparable to sterile medium (Fig. 6; yellow stars). It may therefore be inferred that the strains are safer for use in animals or humans. Double sipA sseC mutants of S. Typhimurium and S. Dublin were selected for further studies in cattle (04/04)

04/03. Assess systemic pathogenicity of mutant strains in mice.

The virulence of S. Typhimurium vaccine strains containing single or double mutations affecting T3SS-1 and T3SS-2 translocon or effector proteins was assessed in mice in collaboration with the group of Carlos Guzmán (Link et al. 2006. Microbes & Infection 8:2262-69). Interestingly, introduction of a mutation affecting the enteritis-associated T3SS-1 effector sopB into an S. Typhimurium sseC mutant improved the efficacy of the vaccine following oral immunisation of mice, apparently as a consequence of relieving a SopB-mediated mechanism by which Salmonella evades host immunity.

In addition to studies to improve live vaccines, we tested ability of numerous attenuated mutants of S. Typhimurium identified by screening signature-tagged mutants in calves, pigs or chickens, to cause systemic disease mice. Of 20 mutants which were tested following intra-peritoneal infection of NrampR mice, one mutant with a transposon insertion in the slsA gene was found to be attenuated. Of 11 mutants which were tested following intra-peritoneal infection of NrampS mice, two mutants with transposon insertions in SPI-4-encoded genes were found to be attenuated. These observations, together with the screening of our S. Typhimurium mutant library in streptomycin-pretreated mice (in collaboration with Prof. W.-D. Hardt. Swiss Federal Institute of Technology, Zurich), indicated that surrogate rodent models do not always identify genes that are required for infection of livestock species, reinforcing the value of studies in target host wherever feasible.

04/04. If strains are attenuated assess vaccine potential of mutant strains in orally infected cattle.

Having demonstrated that a live-attenuated vaccine strain of S. Typhimurium lacking sipA and sseC was less enteropathogenic than existing aroA- or sseC-based vaccine strains in the bovine ligated ileal loop model (04/01), a group of 3 four-week old Friesian bull calves was dosed orally with the mutant strain to evaluate safety and to determine if vaccination conferred serovar-specific or broad protection against Salmonella-induced enteritis. Three further calves were also challenged with a S. Dublin sipA sseC mutant for this purpose. Pyrexial responses, scour scores and the magnitude and duration of faecal excretion of the vaccine strains was monitored daily for five weeks after infection. Both vaccine strains were very well tolerated, with no marked induction of pyrexia or diarrhoea being observed. The organisms were persistently shed in the faeces, with ca. 104 CFU/g still being detectable in the faeces of 2/3 calves in each group by 5 weeks post-inoculation. This complicates interpretation of protective effects, since any immunity cannot be considered sterile and a possibility remains that protection is conferred by competition and/or the induction of inflammatory responses by the vaccine strains that may control incoming bacteria. Indeed, studies at IAH have suggested that gnotobiotic pigs infected with S. infantis are protected from lethal challenge with S. Typhimurium a day later by this mechanism (Foster et al. 2003. Infect. Immun. 71:2182-91). Samples of venous blood, saliva and nasal washes were collected at weekly intervals after oral inoculation to monitor seroconversion. These studies were conducted by Dr Miranda Batchelor at the Veterinary Laboratories Agency, Weybridge. Six weeks after oral infection with the vaccine strains, two calves from each group were used for ligated ileal loop experiments in which two strains of S. Typhimurium (4/74 and 12/75) and two strains of S. Dublin (SD3246 and 2229) were tested for their ability to induce fluid accumulation and recruitment of 111In-labelled neutrophils. The extent to which these strains induce enteritis in calves has been extensively studied (Paulin et al. 2002) and the S. Dublin strains typically induce secretory and inflammatory responses that are approximately half the magnitude of those seen with the S. Typhimurium strains. In calves that were vaccinated with S. Typhimurium sipA sseC, the responses observed in loops inoculated with S. Typhimurium were comparable to those seen against S. Dublin (ca. 3 mls fluid accumulated per cm of loop and ca. 10 fold more radio-labelled PMN compared to medium only control in each case). Thus, whilst complete protection against S. Typhimurium-induced enteritis was not observed, it appears that the magnitude of the response to S. Typhimurium was reduced to the lower level typically observed with S. Dublin. In calves vaccinated with S. Dublin sipA sseC, S. Typhimurium still induced potent secretory and inflammatory responses of a magnitude seen in previous experiments (Paulin et al. 2002), whereas responses to S. Dublin were not as strong as in naïve animals (approximately one quarter of the magnitude of the responses seen with S. Typhimurium). The data indicate that the vaccine strains do not offer complete protection against Salmonella-induced enteritis and imply that the protection observed may be serovar-specific. Further studies are needed to assess protection against oral challenge and to determine the immunological basis of protection.

05/01. Screen spiked pools of S. typhimurium mutants in cattle, pigs and poultry.

Of the mutants found to be attenuated for colonisation of calves, pigs or chickens by signature-tagged mutagenesis, many were assembled in three ‘spiked’ pools for re-testing. These spiked pools comprised mutants

SID 5 (Rev. 3/06) Page 12 of 18

for re-testing and ‘fillers’ of mutants which were found not to be attenuated for cattle, pigs or chickens in the primary screens. The re-spiked pools included 31 mutants which were found to be attenuated for colonisation of calves, pigs or chickens and a further 17 mutants which were not present in the STM pools screened initially, but which were isolated from a library of tagged mutants by hierarchical PCR. These mutants had transposon insertions in loci linked to the mutants for re-testing, for instance in the various Salmonella pathogenicity islands, or in the same operons. The mutants were chosen in order to assess the role genes on Salmonella pathogenicity islands 3, 4, 5, 9, the csg, stb, sth, pil, saf and pef fimbrial operons, and several other virulence genes clusters that could constitute new pathogenicity islands or islets.

The 3 spiked pools were screened by oral dosing of one 4-week-old Friesian calf, five 2-week-old Light Sussex chickens and two 6-week-old Landrace x Large White pigs per pool. The mutants were recovered from intestinal sites 3-4 days post-infection and the composition of the output pools compared with the composition of the inoculum by hybridisation. Although the majority of the results concurred with the findings from the primary STM screen, several of the mutants showed a different pattern of host-specific attenuation than was revealed in the primary screen. This reinforces the need for validation of the findings of STM experiments and has helped to identify clearly attenuating mutations for further study.

05/02. Assess role of Saf and Stb fimbriae in colonisation of pigs by S. Typhimurium.

Mutations affecting two fimbrial loci (saf and stb) were observed to be attenuating upon screening of signature-tagged S. Typhimurium mutants for their ability to colonise the porcine intestines. Transposon insertions in safA and stbD were transduced into the archived S. Typhimurium 4/74 strain using bacteriophage P22 to eliminate the possibility that the attenuation observed in the original mutant strains may be the result of second-site mutations. The transductants were then used to orally dose groups of two 6 week old Landrace x Large White pigs by the oral role in comparison to the parent strain. The course of faecal excretion of the wild-type and mutant strains was followed for one week and clinical signs including diarrhoea, pyrexia and anorexia were monitored. No statistically significant reduction in faecal excretion was observed with the mutant strains (data not shown). This may reflect a subtle role for Saf and Stb fimbriae in intestinal colonisation that is only apparent on screening of the mutant in competition with other colonisation proficient S. Typhimurium strains. Alternatively, it may indicate that a small proportion of the mutants observed to be attenuated by STM are attenuated for reasons other than the transposon insertions that were mapped.

06/01. Survey the prevalence of conserved and host-specific colonisation factors among Salmonella isolates from humans, animals and the environment.

The production and secretion of SiiE via the Type I secretion system encoded on SPI-4 was evaluated in 28 serovars of Salmonella obtained from the Salmonella Reference Collection B (SARB) which represent serovars of S. enterica subspecies I, which are most commonly associated with infections of warm-blooded hosts. Secreted proteins were isolated from the culture supernatants and SiiE was clearly visible by SDS-PAGE analysis in the secreted protein fractions from all serovars studied which included Agona, Anatum, Brandenberg, Choleraesuis, Derby, Dublin, Duisberg, Enteritidis, Emek, Gallinarum, Haifa, Heidelberg, Indiana, Infantis, Miami, Montevideo, Muenchen, Newport, Panama, Pullorum, Reading, Rubislaw, Saintpaul, Senftenberg, Stanley, Stanleyville, Typhisuis and Wien. This suggests that although SPI-4 influences intestinal colonisation in a bovine-specific manner, the distribution of this island is not restricted to bovine-specific serovars.

We also evaluated the conservation of the pil operon, which is carried on a high molecular weight plasmid in S. Typhimurium strain 4/74. As this locus appears to influence intestinal colonisation of cattle, we examined the carriage of this locus in field strains of the most prevalent serovars from cattle, pigs and sheep, which were obtained from the Veterinary Laboratories Agency, Weybridge. Of the 230 strains which were tested, only eight strains were positive for the pil locus by PCR using primers to detect two genes within the pil operon. The eight strains represented four different serovars of Salmonella and did not exhibit a host-specific distribution.

Implicit in the finding that Type III secretion systems-1 and -2 are key colonisation factors of Salmonella in cattle and pigs is the notion that the proteins delivered by these systems into host cells have the potential to alter the outcome of infection. We reasoned that differences in the repertoire, sequence or expression of Type III secreted effector proteins may explain the differential virulence of S. enterica serovars. We therefore determined the nucleotide sequence of the genes encoding the T3SS-1 secreted effectors SopA, SopB, SopD and SopD2 in strains of five serovars (Typhimurium, Dublin, Choleraesuis, Gallinarum & Abortusovis). No differences were observed in sopB, sopD or sopD2, however sopA was found to be truncated in the strains of S. Choleraesuis and S. Abortusovis used. Western blot analysis using SopA-specific antisera confirmed that the truncated proteins are still produced by these strains. Analysis of available Salmonella genome sequences also indicated that sopA is a pseudogene in S. Typhi. SopA associates with mitochondria and influences Salmonella-induced enteritis in calves and the significance of the truncation of SopA in some serovars merits further investigation. In silico analysis of the complete genome sequences of S. Typhimurium, S. Typhi, S. Enteritidis, S. Choleraesuis and S. Gallinarum identified differences in the prevalence or sequence of other Type III secreted effectors (including PipB2, SifB, SlrP, SrfB, SseI, SseK1, SseK2 and SspH2), with a marked increase in the number of effector pseudogenes in host-restricted serovars compared to ubiquitous serovars. Such changes may explain the narrower host range of host-restricted serovars, although it should be noted that the attrition of other genes may explain why the bacteria are able to occupy fewer niches. Analysis of the repertoire, sequence and expression of Type III secreted

SID 5 (Rev. 3/06) Page 13 of 18

effectors in a wider sample of epidemic and non-epidemic variants may identify traits associated with strains that have epidemic or zoonotic potential.

07/01. Assess in vitro phenotypes of non-colonising S. Typhiumurium mutants.

The library of signature-tagged transposon mutants screened in calves, pigs and chickens was also submitted to a number of in vitro assays to detect defects in growth rate, bile resistance and acid tolerance and resistance. Of the 1045 mutants that were screened in this assay, 6 mutants were found to be sensitive to bile, and these mutants had transposon insertions in genes involved in lipopolysaccharide or capsule genes, consistent with the role of these structures in controlling the integrity of the bacterial cell surface. A further 5 mutants were sensitive to acid shock (pH8 to pH3) and 18 mutants were unable to grow under conditions designed to mimic acid tolerance (pH8 - pH5 - pH3). The majority of these mutants were not attenuated for intestinal colonisation as assessed by screening of the signature-tagged mutants in vivo.

08/01. Quantify transcription of SPI-1 and SPI-2 genes by S. Typhimurium and S. Choleraesuis during infection of macrophages in vitro and porcine ileal mucosa in vivo.

Given the difficulty encountered in detection of native (02/02) and epitope-tagged (02/03) Type III secreted effector proteins during infection, we developed quantitative real-time PCR methods to monitor the transcription of genes encoding key translocon components of T3SS-1 (sipC) and T3SS-2 (sseC) in the S. Typhimurium and S. Choleraesuis strains of well defined virulence in our animal models. Bacterial mRNA levels were normalised to an internal control (the conserved housekeeping gene yejA, which is expressed at constant basal levels in mid-logarithmic and stationary phase cultures and following infection of epithelial and macrophage-like cells). As a positive control for infection experiments mgtC was included, since this gene is known to be strongly induced in macrophages. Having examined the sensitivity of the method over a range of RNA concentrations we confirmed that induction of sipC, but not mgtC or sseC, occurs maximally in the mid-exponential phase of growth, consistent with previous observations. Having validated the technology, we next established that sipC is transcribed at a significantly higher level in three S. Typhimurium strains examined than in two representative strains of S. Choleraesuis (p<0.001), a fact that was further supported by the detection of increased levels of SipC protein in the S. Typhimurium strains by Western blotting (Fig. 7). Given that T3SS-1 is a key mediator of Salmonella-induced enteritis, it is tempting to speculate that elevated levels of expression of T3SS-1 components in S. Typhimurium may explain the higher levels of enteritis observed to be induced by this strain in vivo compared to S. Choleraesuis. This in turn may explain why S. Typhimurium induces stronger pro-inflammatory cytokine responses in porcine ileal mucosa (03/05).

Fig. 7. Expression of the T3SS-1 translocon component SipC in S. Typhimurium and S. Choleraesuis detected at the level of transcription by real-time PCR and translation by Western blotting with SipC-specific monoclonal antibody.

We next quantified the transcription of sipC, sseC and mgtC (normalised to yejA) during infection of murine macrophage-like cells (J774A.1) and primary porcine alveolar macrophages by S. Typhimurium strain 4/74 and S. Choleraesuis strain A50. These experiments confirmed that mgtC is markedly induced inside macrophages (Fig. 8 panels A-i and -ii), validating the qRT-PCR approach. Transcription of S. Typhimurium sipC was observed to decrease with time following infection of macrophages, whereas S. Choleraesuis sipC was transcribed at a consistent low level at all time points following infection (Fig. 8 panels B-i and ii). Transcription of sseC in S. Typhimurium 4/74 in both J774-A.1 and alveolar macrophages increased by 4 h post-infection and thereafter declined. In contrast, transcription of S. Choleraesuis A50 sseC increased significantly over time, reaching a peak of more than 100-fold induction 24 h post-infection of J774A.1 cells (Fig. 8 panel C-i). In contrast to J774-A.1 cells, no significant increase in sseC transcription could be detected in alveolar macrophages infected with S. Choleraesuis A50 (p<0.001), (Fig. 8 panel C-ii). These data indicate that both strain- and cell type-dependent differences in the transcription of SPI-1 and SPI-2 genes exist during infection, providing an important cautionary tale that the behaviour of a single gene in a single strain in isolated cell populations cannot be taken as representative of the behaviour of an entire serovar. The intracellular survival of the strains was compared over time, however we were not able to correlate persistence inside macrophages with the patterns of transcription of SPI-1 or SPI-2 genes (data not shown).

SID 5 (Rev. 3/06) Page 14 of 18

sipC qRT-PCR

SipC W. blot

Typhimurium

Choleraesuis

Fig. 8. Quantative real-time PCR analysis of the transcription of mgtC (control) sipC (SPI-1) and sseC (SPI-2) genes during infection of immortalised and primary macrophages by S. Typhimurium and S. Choleraesuis.

Having validated the quantitative real-time PCR method we applied it to evaluate the transcription of mgtC, sipC and sseC in porcine ileal mucosa that had been separately infected with S. Typhimurium and S. Choleraesuis. The tissues were the same as those used to measure net growth of the strains in vivo (milestone 03/01) and the RNA prepared from such tissues was also used for quantification of host pro-inflammatory cytokine responses (milestone 03/05). Thus it was hoped that we would be able to directly correlate the transcription of key Salmonella virulence loci with the induction of host responses and the fate of the bacteria in the same tissue. Despite the fact that RNA extracted from such tissue was competent for detection of porcine cytokine mRNAs by RT-PCR (03/05), we were unable to successfully amplify RT-PCR products for the bacterial genes under study. It was anticipated that adequate numbers of bacteria would be present from the bacterial replication data (03/01), however bacterial mRNAs have short half lives and are highly unstable and they may have rapidly fallen below the limit of detection during post mortem procedures. It should be stressed that such experiments are at the cutting edge of our understanding of bacterial pathogenesis and our laboratory is not alone in having to optimise methods of RNA recovery from infected intestinal tissues.

SID 5 (Rev. 3/06) Page 15 of 18

09/01. Assess the protective efficacy of a subunit vaccine comprising of S. Typhimurium Type III secreted proteins in pigs.

Screening of signature-tagged transposon mutants of S. Typhimurium in pigs confirmed a key role for Type III secreted proteins in colonisation of the porcine intestines (Rowe et al. Submitted). Vaccination of cattle with secreted proteins from E. coli O157:H7 is known to elicit responses against Type III secreted proteins and to be protective against experimental challenge of calves (Potter et al. 2004. Vaccine 22:362), therefore we tested the efficacy of a secreted protein preparation of S. Typhimurium in pigs. Bacteria were cultured in vitro under conditions known to activate Type III secretion and secreted proteins prepared by tricholoracetic acid precipitation and confirmed to contain acceptable levels of endotoxin for vaccination (Fig. 9A). Groups of 4 Landrace x Large White pigs were vaccinated by the intramuscular route with 100g protein at 5 weeks old and 200g protein 3 weeks later in Freund’s incomplete adjuvant. Groups of 4 pigs were separately immunised with matched quantities of protein from a prgH mutant at the same times to assess the role of T3SS-1-secreted proteins in protection and 4 animals were given mock antigen (PBS) in adjuvant. All vaccinated pigs were challenged 2 weeks after the final boost with 5x109 CFU S. Typhimurium 4/74. ELISA and Western blotting experiments confirmed that pigs immunised with secreted proteins from wild-type S. Typhimurium mounted potent IgG responses against Sip and Sop proteins and that such responses were absent when a vaccine from the prgH mutant was used (Fig. 9C). The course of faecal excretion of the bacteria in vaccinated animals is shown in Fig. 9B. The results indicated that statistically significant protection against faecal excretion occurs at peak onset of enteritis (day 2 post-inoculation; p <0.05) when proteins from wild-type S. Typhimurium or prgH mutant are compared to animals immunised with mock antigen (Fig. 9). No significant long term protection exists against experimental challenge. The fact that protection during the acute stage of infection is unaffected by mutation of prgH implies that it is not conferred by T3SS-1 secreted proteins, even though responses against such proteins can be detected by Western blotting using sera from pigs immunised with proteins from the wild-type. It is considered likely that other components in the secreted fraction, notably lipopolysaccharide, may mediate the weak protection observed. The results merit further optimisation of the vaccine preparation, adjuvant, delivery regime and experimental model.

Fig. 9. A. SDS-PAGE analysis of secreted proteins prepared from wild-type S. Typhimurium and an isogenic prgH mutant. B. The course of faecal excretion of S. Typhimurium upon experimental infection of vaccinated pigs. C. Sero-conversion of pigs vaccinated with secreted proteins from wild-type or prgH mutant S. Typhimurium, detected by Western blotting of sera against proteins from the wild-type strain.

10/01. Assess the ability of small molecule inhibitors of Type III secretion to block Salmonella-induced enteritis.

In collaboration with Dr E. Galyov at IAH and Innate Pharmaceuticals AB (Sweden), we examined the ability of salicylanilides and their derivatives to inhibit Type III secretion in Salmonella, as well as bacterial invasion in vitro and the induction of enteritis in calves. Recent studies identified a molecule (designated Compound 1) that specifically impaired Type III secretion in enteropathogenic E. coli (Gauthier et al. 2005. Antimicrob. Agents & Chemother. 49:4101). Compound 1 is composed of a halogenated salicylaldehyde derivative condensed with a 3-aminoacetophenone derivative via a Schiff base linkage. Remarkably, Compound 1 was recently reported to impair Type III secretion in Chlamydia trachomatis (Wolf et al., 2006. Mol. Micro. 61:1543) and a molecule with a similar Schiff base core has been found to decrease transcription of T3SS-related genes and effector protein secretion in Yersinia pseudotuberculosis (Kauppi et al. 2003. Adv. Exp. Med. Biol. 529:97). Such molecules inhibited Type III secretion without affecting bacterial viability, indicating that they are anti-infective rather than anti-microbial in action, with the advantage that the pressure for selection of drug resistance is expected to be lower and escape mutants may lose the function of a key virulence factor.

We have demonstrated that several salicylanilide derivatives inhibited the production and secretion of S. Typhimurium Type III secretion system-1 secreted proteins (Sips and Sops) during growth in laboratory medium without impairing bacterial viability (Fig. 10A). Further, the molecules inhibited contact-dependent haemolysis of

SID 5 (Rev. 3/06) Page 16 of 18

WT prgH

0123456789

1 2 3 4 5 6Days post-inoculation

Mea

n lo

g10 C

FU/g

faec

es

prgH WT mock

Secreted protein fromA. B. C.

SipA

SipC

Pre-imm. Post-imm.

erythrocytes, a process which is dependent on the pore-forming activity of Type III secreted Sip proteins. Invasion of Int-407 cells in vitro was shown to be inhibited by the two most promising lead compounds (INP007 and INP403), albeit only when S. Typhimurium was pre-cultured with the agents and not if the chemicals were added at the time of infection (Hudson et al. Submitted). The two compounds were tested for their ability to inhibit Salmonella-induced enteritis in the bovine ligated ileal loop model. Each treatment was tested in triplicate in three separate calves and the mean volume of fluid accumulated and influx of 111In-labelled neutrophils was quantified after 12 hours. As expected, S. Typhimurium wild-type induced potent intestinal inflammatory and secretory responses whereas a prgH mutant induced markedly lower responses, confirming that the model detects T3SS-1-mediated enteritis. The chemicals or solvent carrier did not induce enteritis per se. When S. Typhmiurium was pre-grown in the presence of the chemicals a significant reduction in the enteropathogenicity of the bacteria was detected (Fig. 10B; p values <0.05), however the inhibition was not detected when bacteria and chemicals were mixed just prior to loop inoculation, indicating that there will be significant challenges ahead to ensure the delivery of pharmacologically active concentrations of drug to relevant intestinal sites prior to Salmonella infection. The inhibition data nevertheless indicate that studies on the molecular basis of Salmonella virulence can yield translational research with the potential to lead to new methods of disease control.

Fig. 10. Effect of salicylanilide derivatives INP007 and INP403 on production of T3SS-1 secreted proteins in vitro (A) and Salmonella-induced inflammatory responses in a bovine ligated ileal loop model of infection (B). The effect of co-injection of inhibitor at the time of inoculation with S. Typhimurium (ST) was examined (Co-inj.) as well as the effect of culturing the bacteria in the presence of inhibitor (Cultured).

Added value of Defra funding.

The Research Technician supported by OZ0319 trained in surgery and anaesthesia (Institute of Biology Module 4 for personnel working under the Animals (Scientific Procedures) Act). In addition to her commitments to OZ0319 she used her surgical skills to demonstrate that the host neuroendocrine stress hormone norephinephrine augments Salmonella-induced enteritis in a bovine ligated ileal loop model of infection, providing a possible molecular explanation for the increased faecal excretion of Salmonella in animals subject to stress (BBSRC project C518022). In addition, she used a novel lymphatic cannulation model to demonstrate that systemic translocation of the host-restricted S. enterica serovar Dublin in calves occurs via efferent lymphatic vessels in an extracellular niche in a manner that requires Type III secretion system-1, but not T3SS-2 (BBSRC-Defra project C50964X; Paulin et al., 2002; Pullinger et al., in preparation). These high-impact observations contradict received wisdom from the murine typhoid model and emphasise the need for studies on Salmonella pathogenesis using natural pathogens in their target animal hosts. Defra funding also allowed Drs Stevens and Wallis to participate in the development of in vitro organ culture methods to study the interaction between Salmonella and human and bovine intestinal epithelia (Haque et al., 2004). Drs Wallis and Morgan also contributed detailed reviews on the molecular basis of Salmonella virulence in recognition of their important Defra-funded contributions to the field.

PublicationsPublished and submitted manuscripts arising from project OZ0319 are listed below. Dr Stevens assumed

responsibility for such outputs only during the last 15 months of the project and wishes to stress that every effort is being made to bring insights delivered under Dr Wallis to publication. Manuscripts have now been submitted on several aspects of the research supported by OZ0319 (SPI-4, ccf islet, pig STM screen and vaccination, salicylanilide inhibitors of Type III secretion) and manuscripts reporting net growth in vivo of S. enterica serovars and systemic translocation of S. Dublin are in an advanced stage of preparation.

SID 5 (Rev. 3/06) Page 17 of 18

Mwt DMSO 007 403

- SipA

- SipC- SipD

A B

PMNInflux

LBDMSO

STDMSO

prgHDMSO

ST007

Co-inj.

ST403

Co-inj.

ST007

Cultured

ST403

Cultured

LB007

LB403

References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.

Hudson D, Field T, Bowen A, Bispham J, Wolf-Watz H, Stevens MP, Galyov EE. 2006. Inhibition of Salmonella Type III secretion by small-molecule inhibitors. Submitted to Antimicrob. Agents & Chemother.

Rowe S, Bowen A, Morgan E, Maskell DJ, Wallis TS, Stevens MP. 2006. Identification of Salmonella enterica serovar Typhimurium genes required for colonisation of the porcine intestines and application of this knowledge for disease control. Submitted to Microbiology.

Morgan E, Bowen A, Wallis TS, Stevens MP. Salmonella pathogenicity island 4 encodes a Type I secretion system required for secretion of a ca. 600 kDa protein (SiiE) that influences host-specific intestinal colonization, invasion and the induction of enteritis in calves. Submitted to Infect. Immun.

Morgan E, Rowe SC, Bowen AJ, Barrow PA, Maskell DJ, Wallis TS, Stevens MP. Identification of a Salmonella enterica serovar Typhimurium pathogenicity islet required for host-specific intestinal colonization. Submitted to Infect. Immun.

Morgan E, Wallis TS. Salmonella pathogenicity islands. In Salmonella pathogenesis, molecular typing and epidemiology. Horizon Scientific Press. In press.

Boyen F, Pasmans F, Donne E, Van Immerseel F, Morgan E, Adriaensen C, Hernalsteens JP, Wallis TS, Ducatelle R, Haesebrouck F. 2006. The fibronectin binding protein ShdA is not a prerequisite for long term faecal shedding of Salmonella typhimurium in pigs. Vet. Microbiol. 115:284-90.

Wallis TS, & Barrow PA. 2005. Salmonella epidemiology and pathogenesis in food-producing animals. In EcoSal-Escherichia coli and Salmonella: cellular and molecular biology. Module 8.6.2.1. Ed. J. Kaper [Online.] http://www.ecosal.org. ASM Press, Washington, D.C.

Wallis TS. 2004. Vaccination against Salmonella, enterohaemorrhagic E. coli and Campylobacter in food-producing animals. Dev Biol (Basel). 119:343-50.

Morgan E, Campbell JD, Rowe SC, Bispham J, Stevens MP, Bowen AJ, Barrow PA, Maskell DJ, Wallis TS. 2004. Identification of host-specific colonization factors of Salmonella enterica serovar Typhimurium. Mol. Microbiol. 54:994-1010.

Haque A, Bowe F, Fitzhenry RJ, Frankel G, Thomson M, Heuschkel R, Murch S, Stevens MP, Wallis TS, Phillips AD, Dougan G. 2004. Early interactions of Salmonella enterica serovar typhimurium with human small intestinal epithelial explants. Gut 53:1424-30.

Thomsen LE, Chadfield MS, Bispham J, Wallis TS, Olsen JE, Ingmer H. 2003. Reduced amounts of LPS affect both stress tolerance and virulence of Salmonella enterica serovar Dublin. FEMS Microbiol. Lett. 228:225-31.

Morris C, Tam CK, Wallis TS, Jones PW, Hackett J. 2003. Salmonella enterica serovar Dublin strains which are Vi antigen-positive use type IVB pili for bacterial self-association and human intestinal cell entry. Microb. Pathog. 35:279-84.

Paulin SM, Watson PR, Benmore AR, Stevens MP, Jones PW, Villarreal-Ramos B, Wallis TS. 2002. Analysis of Salmonella enterica serotype-host specificity in calves: avirulence of S. enterica serotype gallinarum correlates with bacterial dissemination from mesenteric lymph nodes and persistence in vivo.Infect. Immun. 70:6788-97.

SID 5 (Rev. 3/06) Page 18 of 18