U. S. Food and Drug Administration
U.S. Department of Agriculture
September 8, 2000

ROBERT BRACKETT, FDA, Presiding

SPEAKERS

Opening Comments/Welcome - Mr. Brackett
Peter Holt, ARS, USDA
Bailey Mitchell, ARS, USDA
Ahmed Yousef, Ohio State University
Doug Waltman, Georgia Poultry Lab
Richard Gast, ARS, USDA
Jean Guard-Petter, ARS, USDA
Charlie Beard, U.S. Poultry & Egg Association
Christine Bruhn, University of California
Eric Ebel, USDA

Panel Discussion

Public Comment Period:
Jill Snowdon
Karen Davis
Phyllis Bedford
Chuck Benson
Roberta Morales

Closing Remarks, Mr. Brackett

P R O C E E D I N G S

MR. BRACKETT: Good morning. Welcome to this public hearing. We have people here to address the research that is being done on this problem, and we have some who represent other interests.

This relates to the background in salmonella enteritidis illnesses that have increased over the past decades, and to the point where in your package you also have the Egg Safety From Production to Consumption Egg Action Plan, and this was published in 1999 as the long-range strategy to address this issue.

One of the -- there's a number of different objectives that are outlined in the action plan, but specifically one that we are interested in is research, that is how do we get the information that we need to make the policies and the decisions that we need to do to solve this problem.

And the specific areas which also are listed on your agenda is that there were four very broad objectives to this, and as I said they're on your agenda, and each of our speakers or group of speakers have been asked to sort of summarize and address what has been done and where things are going in these specific areas.

The specific topics range all the way from very applied, very on-farm practical type research all the way to molecular and genetic methods that would help us get to the mechanism of salmonella enteritidis illness in animals as well as in humans, and so it really spans the whole scope of what could be done in biological research.

As I mentioned, each of the speakers will provide sort of an overview, so this is by no means a comprehensive discussion; it is meant to sort of identify the gaps, and really that is the goal of this meeting, and what we hope to come away with at the end of the day, and that is to address sort of the state of the science regarding SE.

There have been many symposia over the years that have addressed SE, but this one is a little bit different than others in that we are specifically addressing those research items that were addressed in the action plan. And so the idea is to find out where we are right now, that is what has been accomplished that's in the action plan, where things are going right now -- we hope to hear a little bit about what research is going on now that perhaps has not been published yet, and more importantly to identify those research gaps so that we can find out what needs to be yet done in the future.

And so the outcome of this meeting, that is the research gaps, finding out what has been addressed will help to set, or at least allow both regulatory agencies as well as industry to focus their research dollars in a more effective way. That is identify the research gaps and set funding priorities as well as perhaps readjust the priorities that have already been set.

The format that we're going to use today is a little bit of a mixture of a variety of different techniques. The first groups addressing the different goals will be sort of symposium style, that is the speakers will give an overview; we will allow a few minutes if possible for technical questions, and we do ask that you limit these to technical questions. If you have other opinions or other questions, please wait until the end of the day during the public comment period.

Secondly, in the afternoon we will have a panel discussion with the speakers, and the goal of this is to get the speakers to answer some of the questions that were also identified in the Federal Register notice, which is to figure out what research and consensus looks like needs to be done

-- is where are the research gaps -- and perhaps some other questions, for instance what is the best way to get this research done. That is, who is to fund it, is it best done through private funds, is it best done through government funds; if so, how should that be done. Would it be best as a competitive grant? Would it be best as contracts? These are the sorts of questions that we would like to get some input on.

And then finally at the end of the day we will have a public comment period in which each person who wishes to can give a five-minute statement, or if they have written comments they can provide those also to Wendy Buckler.

Wendy Buckler for those of you who have not yet met her is the lady standing in the doorway, and the person who is really the person that gets the credit for organizing the meeting, and she will handle all of the audiovisuals for the speakers, as well as getting the information to the dockets.

Now, since this is a public meeting all of the comments will be recorded, and it will be part of the public record, and so anything that is said here has to be available to the public, and so during the public comment period that's why there's only five minutes, and if people have more to say they can send in written comments as well.

Finally, a little bit about the hotel. If you haven't already found them, the restrooms are all the way down the hall out the door to your right, and we will take several breaks, and we hope to keep those short and on time. And then also we will break for lunch. We are going to try to get a list of restaurants that are nearby. There are some right in the hotel here, there are some within walking distance although it's raining, and if you have a car there are some others just down the street, but there are a number of restaurants within five-minute drive, and some within a walk.

Okay. At this time I would like to also acknowledge the help that we've had from the Agricultural Research Service in their providing speakers, as well as the Food Safety and Inspection Service for helping to organize this. This has been a very cooperative effort that affects all of us, and so we try to do this in a concerted effort.

Okay. I guess we'll get started here. Our first speakers will be Peter Holt and Bailey Mitchell who are from ARS. They are ARS scientists who are specializing on salmonella enteritidis, and they are going to first address Objective 7, that is to ensure adequate current information is available to make decisions, but specifically 7.1, to develop and evaluate on-farm intervention strategies and technologies, and they are going to split their time, and first we'll have Peter Holt speaking.

STATEMENT BY PETER S. HOLT, SOUTHEAST POULTRY RESEARCH LAB, ATHENS, GEORGIA

MR. HOLT: Thanks Bob.

Bob has had me do the Objective 7.1 which is to conduct research, to develop and evaluate on-farm intervention strategies or technologies.

There's a lot of information to be given, so what I'm going to have to do is go fast and furious through a lot of it to get through everything, and rather than the long of it I'll give the short of it.

The first part of the Objective 7.1 is forced molting and other stress factors. The question that occurs is why molt in the first place.

Now, as a laying flock ages its ability to lay eggs decreases, and it reaches a point where it's no longer economically feasible to keep the flock in lay. A producer can send all his birds to slaughter and bring on a new flock, or he can recycle his birds.

Well, what the general trend is is most of the producers recycle their birds. This is a slide from 1987, and about 60 percent of the flocks were recycled at that time, and it's moved up to about 70 percent now.

When you put pen to paper figuring about 240 million birds in the U.S. that comes to somewhere between 144 and 168 million birds that are molted annually.

Now, there's a reason for this. Most of the early studies have shown that the effects of molting were primarily positive, that it increase productivity. Of course, that's the reason they recycle the birds in the first place.

Increased feed conversion, and actually on a number of the studies they actually had less mortality than their unmolted counterparts, but that's not always the case, so this equaled the rest of the rigors of daily egg lay.

Now, there are a number of ways to molt birds, but feed and nutrient restriction and feed removal are the two prevalent procedures to recycle the birds, and feed removal as shown in the green is the procedure that we looked at, and this is the primary procedure that we worked with. Generally dropped the photo period down to eight hours a day because egg lay is affected by photo period; take the birds off of feed and that drops our particular flocks' weight somewhere between 25 and 30 percent, and then start them back on the grower ration throughout the experiment.

Now, the first thing we looked at was the effect of molting on immunity, and we found that there were some pretty dramatic effects. While humeral immunity to antibody response was largely unaffected, cell mediated immunity was significantly depressed as indicated by three different parameters, and when we did photositometric analysis of the peripheral blood lymphocytes we found that the CT4+ T cells, the helper T cell subset was significantly decreased.

Now, the importance of the immune system is severalfold. First of all, in order to elicit to vaccination you need an intact immune system, but in birds this age that really doesn't play as big a factor.

Where it does play a factor is it affects their ability to fight disease, whether it be viral, protozoan, fungal, or bacterial, and so we focused in on a bacterial infection which is salmonella enteritidis and we found that molting did have a substantial effect on experimental infections, and I need to stress that that all the SE studies that we did were all experimental, done under controlled conditions with our specific pathogen-free flocks.

But birds that were infected during the molt, we had increased shedding, birds were infected for longer periods of time. If we infected the birds before the molt normally the normal-fed birds would generally clear the infection, but the molted birds a certain percentage of them would stay persistently infected, and that's shown in this slide.

You can see in the unmolted birds shown in green by day 24 they had essentially cleared the infection, but you can see that a certain percentage of the molted birds stayed positive throughout the experiment.

Molting also affected the susceptibility to infection. Generally it takes somewhere around five times ten to the fourth SE to infect a bird; it takes less than ten during the molt. So they're extremely susceptible to infection at this time, and because of that you get a very rapid horizontal spread to uninfected hens in adjacent cages.

And the way we ran this experiment is we had cages of molted and unmolted birds, eleven birds per row, and we infected just the center bird with a dose which is right around fifty percent of the infectious dose for unmolted birds, and you can see in the red that the unmolted birds had very little transmission; the molted birds you got a very rapid transmission. By day three about 35 percent of the birds were positive, and by day ten it's 85 percent, and they remained high from then on.

Now, all these studies were done in experimental conditions. There have been some studies looking out in the field, and this is from the SE pilot project, and they looked at the production of SE-positive eggs, and they did find that weeks zero to five post-molt there was an increase in the production of SE-positive eggs, and I think, Eric, you will probably be talking a little bit about that as well, so I won't dwell on it.

Now, what might be some of the causes that are affecting the SE infection. Immune depression is probably very prominent, but we saw on occasions effects occurring within 24 to 48 hours after infection, which is awfully fast for effects on specific immunity to play a role, so it had to be other factors, and depression of the immunity cropped up as a potential possibility, and Dr. Mike Cogan with the USDA lab down in College Station, Texas showed that heterophil function, the white blood cells were significantly depressed, so it looks like immunity is affected.

We thought because the birds were off feed that there would be an alteration of the intestinal microflora, and Dr. Don Coyer also from the lab at College Station, Texas, and unfortunately has recently passed away couldn't find any effects on the gut flora. It doesn't mean that they aren't occurring, it just means that they couldn't find them.

And finally there may be an effect on peristalsis and digesta. The combination of peristalsis and digesta are very effective in keeping the intestinal tract clean, and by removing the feed you very well may be eliminating one of the protective capacities.

Now, for some of the solutions, looking at the effect of digesta we ran a number of different what I call alternative molt procedures, molting the birds alternative to total feed withdrawal, and working in collaboration with the scientists at Poultry Science Department at University of Georgia they developed a low-energy/low-calcium diet which we then ran in comparison with total feed withdrawal, and we found that while the shed rate was largely unaffected, and that's a trend we normally observe, the amount of SE that's shed is significantly decreased, and that this is very important for transmission to other birds, for disinfection and cleanup in the house, and also for contaminating rodents and flies in the house as well.

Now, this experiment used a metered amount of feed. We generally gave them sixty grams per day, so that does make it a little bit more difficult for the producer, and the procedure never caught on.

We also looked at low nutrition/lower energy feed additives. Soybean hulls an cracked corn really didn't work all that well. We did see a decrease in the amount of SE being shed, but where we really saw effects were with what middlings, and wheat middlings are a byproduct of what processing.

And when we gave the birds ad lib amounts of wheat middlings we saw a very substantial decrease in the amount of SE being shed, actually back down to control levels.

I think very telling is the amount of SE that's disseminated extraintestinally, either the liver and spleen or the ovary, and actually with the ovaries we couldn't find any SE in the two fed groups, but 63 percent of the birds were ovary positive in the total feed withdrawal.

Now, the whole point behind the research is to try and find intervention strategies that may help on the SE infection, so we also looked at antibiotic therapy, and I'm saying right now I'm not an advocate for antibiotic therapy, but I thought it was important to look at it.

And working in collaboration with Baer Corporation we looked at the use of Baytril, an antibiotic, and eliminated the SE infection, and what we did was we administered the Baytril after the birds had finished up the feed removal period, and then after the ten-day regimen of Baytril when we put them on AviGuard which is their competitive exclusion culture to repopulate their intestinal tract.

And what we found was is that the Baytril did substantially decrease the percentage of birds that were SE positive in 33 to 4 percent by day 33, and from 25 percent down to zero percent by day forty. So it can be an effective way of eliminating SE infection after a molt.

And finally vaccination. Now, we worked up a collaboration with Megan Health using their live salmonella vaccine as a protection, potential protective capacity. This was requested by Gene Gregory from United Egg Producers to see what effect it would have, and what we did was that we vaccinated the birds two times with the Megan vaccine by aerosol two weeks apart, and then two weeks after the second boost, and challenged the birds.

Using the transmission study that I talked about before we had our groups of molted birds, and the center hen in each row got three times ten to the fifth SE, and then we followed the transmission down the line.

Now, this is day three post-challenge, and with the non-vaccinated birds we had about 25 percent of the birds were SE positive by day three; only 5 percent, one bird in the vaccinated group.

By day ten 75 percent of the birds were SE positive in non-vaccinated as opposed to 45 percent, but what you can look at is in that 45 percent it's very low numbers as opposed to like ten to the fifth in some birds, ten to the third, so the unvaccinated birds were also shedding substantial amounts of SE as well.

By day 17 the birds are starting to clear, but there are certain birds that are still shedding quite a bit of SE in the nonvaccinated group, and as far as internal organs go, the vaccination totally eliminated any extra-intestinal dissemination to livers and spleens or to ovaries.

So where do we go from here on molting? There is quite a bit that needs to be done. I think the wheat middlings show an awful lot of promise. I think that there are probably other possible procedures that need to be looked at, and once we settle on one we need to kind of determine just the total effect on the SE infection, looking at the 50 percent infectious dose pathology itself, et cetera.

Also more work needs to be done on molt as a stressor, and we have worked up a collaboration with a relatively new USDA lab, the Livestock Behavior Research Unit at Purdue, to look at the effect of molting on various neuroendocrine factors and behavior, and so what we plan on doing is once we get the initial studies with feed withdrawal done we'll start looking at the alternative molt procedures as well to see just how much of a stressor that is.

And last, but not least, is examine molt against SE in the field, and I really think this is an important variable. There has been very little work really done out in the field looking at the effect of molting on SE, but at the same time an awful lot of verbiage has been made about molting as a possible food safety situation, and the only way that this question could be put to bed is to actually go out and look at it, and that's what we plan on doing.

And what we want to do is go out and follow SE infections in flocks from before the molt, during the molt, and afterwards, and then look at a number of different parameters which may affect health science -- age of the flock, manure handing, and see if there is one or two or several different parameters which may enter into the equation.

And this is actually the number of the parameters we want to look at, and other salmonella -- and I'm going to have to thank Doug Waltman for this suggestion -- this very well may be a very important parameter, and not a negative parameter, a positive one that the presence of a number of salmonella very well may offer some degree of protection.

Now, there has been some work out in the field, the SE pilot project that I mentioned before, and also the NAHMS which are connecting the incidence of SE in houses with the molting procedure. Previous status of the house is unknown, so it's totally an epidemiological situation. And this is the questionnaire in kind of a nutshell in the NAHMS study.

Now, also in that Objective 7.1 is other stressors in SE. There has not been a lot of research that has been done. Disease is kind of the primary one. Phillips and Opitz showed in 1995 that infectious bursal disease increased the persistence of SE infection in birds and the number of SE-positive eggs.

Qin et al over in Japan -- this is a Japanese group that has done just a tremendous amount of work on coccidia and the effects on SE -- there have been a number of studies on environmental stressors, thermal, crowding, transport on salmonella infections in general, but nothing specifically on SE, and intoxication which generally would be like microtoxins, aflatoxins, T2 toxins that has also been known to affect salmonella infections.

Okay. The next intervention strategy would be vaccination and its effects on salmonella enteritidis infections. There are two primary types of vaccines. There are multiple different kinds of vaccines available, but the two primary ones that are available commercially are live which are attenuated salmonella which reduces the effectiveness for the host and for humans, and it's generally administered in the feed, water feed, or possibly as an aerosol, and inactivated which most everyone is familiar with, your standard vactarins which are injected.

As far as the live vaccines go, there is only one available commercially licensed in the United States, and that's Megan Vac from Megan Health, Incorporated, that's a double-dilution mutant, it's a cyclic AMP, a cyclic AMP receptor protein mutuant.

There are a number -- and this is just a small number of live vaccines that are out and available -- Zoosaloral, Zoosaloral H, and Salmonella vac T out of Germany. Fort Dodge is working with an Aral A, and there is a rough strain of salmonella gallinarum that was developed by H. William Smith back in the 1950s that's floating around.

There are currently three salmonella bacterins licensed in the United States, Layermune SE from Biomune of Lenexa, Kansas; Maine Biological Laboratories has an Inacti/Vac SE4; and Fort Dodge has recently come out with one Poulvac SE; and for those individuals who want to clear up their salmonella infections in their flocks there are autogenous vaccines that can be made by these companies as well.

Now, inactivated vaccines have worked pretty well in clearing up experimental infections, reduces clinical science and pathology, shedding is reduced, organ positivity, the A-positivity, studies showed that growth in egg contents was reduced.

The problem is vaccination can't be used in and of itself, it has to be used in combination with good management practices to help eliminate the SE problem in the flock.

Field work, most of the studies that come from the SE pilot project saw some reduction in positive environmentals and positive eggs. The Pennsylvania Egg Quality Assurance Program has showed that there was a substantial decrease in environmentals, and the eggs from environmentally positive eggs were 8 percent positive which were reduced to zero percent positive, so it does look like vaccination very well may have a role in reducing SE problems in the field.

Particularly telling is the inactivated vaccine in England. The producers over there, about 80 percent of them signed up to vaccinate their birds, they used a vaccine produced by Hoechst which was an iron-starved salmonella enteritidis which produces some iron scavenging proteins which they felt would be effective in a vaccine. They vaccinate the birds at hatch, and then when they are transferred to the layer facility, and they have seen a pretty substantial drop in salmonella enteritidis cases, and they feel that vaccination has played a very substantial role in that.

And protection by live vaccines, there has not been a lot of field data on live vaccines. It's still too new. This is experimental data, and essentially shows very similar results than the killed bacterin. There has been some observations of cross protection against different salmonella serovars, but that is variable with the vaccine, and as with the other -- with the bacterins this can only be, it needs to be used with good management practices.

What are the future directions for that? I think that we're going to see more live vaccines coming on the scene, and I would love to see them. I think live vaccines are a very important mechanism for helping to eliminate the SE problem.

Mucosal vaccinations, before I was redirected back into molting we had an active group going in that, and I think that can have a very major role in the future as well.

In ovo vaccination very well may play a role, and we've had some promising results from that as well.

And subunit/vectored vaccines and DNA vaccines are down the road.

Finally one last intervention strategy is competitive exclusion. The whole principle behind competitive exclusion is that very young birds lack an intact flora first week post-hatch, and Nurmi and Rantala in 1973 showed that if you took intestinal contents from adult birds and gave them to these newly-hatched birds it would help protect against salmonella infections, and there have been a number of studies that have shown it's been very effective to prevent colonization of chicks with different salmonellae, including salmonella enteritidis.

Now, what role does competitive exclusion play for SE? Just a partial role actually. It can be very important in preventing colonization in newly-hatched chicks, and this can be really important.

Richard Gast and I have done some studies where you infect very young birds, and they generally a lot of times will develop a persistent infection that lasts all the way out into egg-laying, so it's very important to try and clear up that infection as early as possible.

It has fairly limited utility in adult birds because they already have a well-developed intestinal flora. However, if the birds have been subjected to antibiotic therapy, then you can use competitive exclusion to repopulate the intestinal tract.

And finally there is only one commercial competitive exclusion product available or licensed here in the United States right now, and that's Pre-empt from Milk Specialties, but there are several other commercial products that are available, and hopefully the license will be approved in the not too distant future, Aviguard from Bayer AG, and Broilact from Farmos Orion.

The Poultry Microbiological Safety Research Unit in Athens, Georgia has also developed a mucosal competitive exclusion, and they are working for licensure as well.

And saccharomyces boulardii is actually not really a competitive is not really a competitive exclusion, it's more of a sponging type of organism which actually causes the salmonella to adhere to their surfaces, and then they just pull them on out of solution, or out of the intestinal tract. And that's it. And what I'll do is go ahead and pass the baton over to Bailey Mitchell who will be talking about negative air ionization.

STATEMENT OF BAILEY MITCHELL, USDA-ARS Southeast Poultry Research Laboratory, Athens, Georgia

MR. BAILEY: I want to look at a little different approach. From an engineering perspective there's also some things that we could probably do intervention-wise in dealing with SE. I basically want to go over some possibilities with electrostatic space charge.

Basically what I want to do in this approach is to reduce SE levels in the air by removing bacteria-laden dust, and there's also some killing effect that we might be able to use.

The results that we're looking for is to basically reduce SE transmission between birds, houses, poultry areas, and also to reduce SE-contaminated eggs, and cross-contamination, also a good potential for improving bird and animal caretaker health by improved air quality.

Basically what we're trying to do is introduce a strong electrostatic charge into an enclosed space. This will charge any kind of dust or particulate matter in the air in a negative direction, and then that dust would be attracted to room surfaces, or if you have in some cases specialized collectors that collect this dust off.

An interesting thing here, you can get a little extra bang for the buck by taking dust out because there have been some studies done that show for example if you take out half the dust in a room by various means that you can reduce airborne bacteria by a factor of a hundred or more, so a lot of bugs attach to dust.

Just a little quick video here in a small ionizer chamber, a small hatching cabinet, with the ionizer off you can see the smoke source just kind of dissipating here.

This is with it on, it's drawn to that grounded plate there.

A little closer up view with the ionizer off. That's on.

That just gives you a little visual picture of what you can do. You can draw materials for a foot or so in that manner.

This is looking at some feathers just to see what you can do with feathers, something that large. They come down through a tube that's got a grounded strip on the right without the ionizer. This is with coming up next. You see that stuff being drawn over to the ground strip on the right side.

We did some work in a room with caged layers, put an ionizer unit in the center of the room, and we had exhaust filters in the back that you can see here that are normally blue when they're clean. In this case the birds were infected with SE, mature laying hens. We ran the experiment for about ten days, and we found we were able to reduce the dust level by 52 percent with the ionization compared to an identical room without.

Notice after ten days this filter on the exhaust still looks basically clean on the ionizer room; the other room is starting to plug up with the chicken dust here.

Interestingly, right after that we ran the SE experiment and looked at SE levels in the air with plates spread around the room, and ran that for ten days with 24-hour samples, and found we reduced airborne SE by 95 percent, so that kind of reaffirms this concept that if you take dust out you'll get a little extra benefit on your bugs.

Another interesting study here, some folks in England looked at various ways of getting salmonella into eggs with salmonella typhilurium, and using an oral challenge they were able to get about 2 percent positive eggs. With the aerosol challenge, low-level aerosol they were able to get about 14 percent. That's about eight times more than the oral challenge.

With a little bit higher aerosol they were able to get 25.4 percent. That's about 15 times more than the oral challenge, so it does kind of suggest that the aerosol route is important more than probably a lot of folks might have thought.

We did some stuff, Dr. Gast and I did some studies with looking at airborne transmission in some special cabinets where we could isolate donor birds in the front part of the cabinet, air flowing from front to back, put susceptible birds in the back, and we started out with day-old birds up here, inoculated them with SE, and then we look at the transmission downwind.

Just to look at the results at day ten, surface contamination in the untreated -- I didn't say that, one of the cabinets had an ionizer in it and the other one didn't -- in the untreated cabinet there was a hundred percent surface contamination, cecal contamination about 30 percent, and then over here -- well, I'm sorry -- this is surface contamination on the treated birds, and then cecal contamination about 90 percent on the untreated birds, and we had none here at ten days on the treated cabinet, so it had a good effect on airborne transmission as indicated by surface, particularly by cecal contamination.

We put these things in some commercial hatching cabinets also. This is a Jamesway cabinet, you can see the ionizer units here, they go on both sides of the fence. We put a grounded collector plate on each side. You can see it a little closer here, just a series of electrodes with high voltage DC applied to it to generate the ions.

Look at exhaust covers just to get a sense of the visual effect. After a hatch this is an exhaust cover from an ionizer cabinet. You see it looks quite clean here versus the control cabinet without any treatment. You can tell quite a difference there.

We were doing some plate sampling, auger plates. This a control cabinet in the upper part of the exhaust. This is the upper part of the exhaust on the ionizer cabinet, so I think you can see we're getting a good dust reduction.

We have also done a lot of plate samples using things like XLT plates, McConkey plates, and exhaust of hatching cabinets. In this case these were some XLT plates with the treatment cabinet with the ionizer versus a control cabinet without, so we're looking at ecol-I, maybe some salmonella here.

This is with the higher flow rate. You can see it's a little more dramatically on the treatment versus the controls, so we get usually somewhere in the neighborhood of 95 percent reduction in airborne pathogens by using this process in the hatching cabinet.

We have also done some studies just to look at the potential inactivation effect of this electrostatics. We have used, in a safety cabinet used a little chamber here with Argo plates in there, XLT plates, pump some air through a solution containing SE, pump that aerosol into the chamber, we've got a small ionization unit in there, and we look at how much SE we can recover with and without the ionizer.

I'll show you the results of the individual plates, but we go in and rinse everything out, take a sample of that rinse, we get something like this typically with a control plate it's all SE. There's the treatment plate.

So it looks kind of encouraging. We don't know exactly what level of charge it takes to get that, but that's a pretty high charge level. That's the next thing we need to look at is what kind of charge level it takes to get that.

We've also done some biofilm studies with Judy Arnold over at the Russell Center using broiler carcass rinses, taking a cocktail off of that, putting it on stainless steel coupons and treating those with electrostatic process. We got 99.8 percent reduction in three hours, 97.3 in two hours. This is consistent, and so that looks kind of encouraging as a potential non-chemical sterilizing technology that could be applied to SE as well.

Just something to give you a little relevance for this stuff. We did get recognized last year for tech transfer with the technology. It's also been listed in the President's Egg Safety Action Plan, it was listed as ion air scrubbers in hatchers. I would suggest a more appropriate name would be electrostatic space charge; it's not just a hatcher type thing.

Just like air quality, if you can clean up air it doesn't matter, you can do the same thing in a lot of places other than just hatchers.

Basically the system has been patented, it's been licensed to a company for manufacture and distribution. We've done a number of trials, commercial trials with it with commercial broiler folks, and we've got about three other commercial trials in progress. Still doing things back at the lab in the research setting.

Basically application areas would include continue to look at this inactivation process, airborne and surface SE. We've got a proposal pending on that.

We want to look at and see what we can do in a breeder house setting where you're feeding a lot of this material into the hatcher. We've got a grant proposal pending on that.

And then depending on how that goes we might want to look on out at production house, egg rooms, and we're already looking at hatching cabinets.

That's it.

MR. BRACKETT: Thank you Peter and Bailey.

We do have a couple minutes for any technical questions if there's something that either of the speakers did not make clear. First of all, if you do have questions we're going to ask you to go to the microphone and state your name as well as your affiliation for the record. In the meantime our next speaker is preparing his presentation.

Do we have any questions for either Dr. Holt or Dr. Mitchell?

[No response.]

MR. BRACKETT: Okay. Our next area of interest of course is Area 7.2 in the action plan, and that is to conduct research and provide additional information on commercial processing technologies and practices, so this goes from the realm of the farm now to food processing and more into food technology.

There are a number of investigators that are looking at this around the country. This morning we have with us Dr. Ahmed Yousef who is on the faculty in the Food Science Department at Ohio State University, and he will be providing an overview of some of the food technology type applications.

STATEMENT OF AHMED YOUSEF, Ohio State University

DR. YOUSEF: I will be talking about current and potential processing technologies and egg safety, so I will modify the topic a little bit.

Egg processing and safety, you can deal with two types of products, shell eggs and liquid whole eggs, but frankly because of time limitation I will focus basically on shell eggs, their safety and the processing and how the processing techniques affect the safety eggs.

The microorganisms of concern in shell eggs, of course we know that salmonella is one of them, but we know that other pathogens also can be important in shell eggs like microplasma viruses, and nonpathogenic microorganisms like pseudomonas proteus and even molds can be a problem.

With liquid whole eggs of course salmonella coming from shell eggs, but other microorganisms may be found in whole eggs that were not found in shell eggs like conceomoctogones [ph] and other gram negatives and spore formas which affect the quality of liquid whole eggs.

And the processing techniques that are meant to deal with microbial problems of shell eggs or liquid whole eggs include washing, in-shell pasteurization, or some alternative technologies that are coming up these days. These alternative technologies are not in practice, but they are coming pretty strongly, and I will comment a little bit o some of these.

I'm sure all of you know that salmonella gets into eggs through one of these three routes: if the ovary of the hen is infected, then there is a good chance that the egg coming from that ovary will be also containing salmonella. And the pathogen stays in the yolk in this case, and there is a chance for growth of the pathogen inside the yolk. However, trans-shell infection can happen. We call this horizontal, sometimes we call it horizontal. This happens through fecal contaminants. While the egg is being laid feces can be on the outside shell, and these may get sucked into the egg and contaminate the interior, the inside contents.

Improper washing may aggravate this problem, and the pathogen stays most of the time in the shell, but it may migrate trough the white and may eventually actually reach the yolk.

During egg-breaking if the shell is contaminated there is some chance of course that the pathogen will end up in the eggs.

So I will focus a little bit on processing shell eggs and how this affects the safety of the egg. These are the reasons that I think people should keep in mind while they are processing shell eggs. Of course, washing is done for visual reasons, aesthetic reasons, but freshness, shelf life, and the safety against external infection and internal infection should be in the minds of processors who are introducing new technologies.

So washing is done basically to remove fecal matter; this is the primary reason for washing. In fact, in some European countries they don't wash eggs, and they consider that washing is making eggs unsafe.

It all depends. This is a typical commercial egg washing process here. From the henhouses eggs are transmitted by a conveyor belt to the washing machines where the eggs are dipped in tanks containing chlorinated water and detergent. Usually the pH is pretty high, sometimes ten, sometimes eleven, and the temperature is mild, 110 degrees Fahrenheit, and this happens very quickly, one to two minutes.

Then the eggs are rinsed in hotter water, 140-150 degrees Fahrenheit for five seconds, very quick, dried with air because you want to remove as much water as you can, five to seven seconds, and then the eggs are candled, graded, packaged, and most importantly refrigerated during storage, because it has to be refrigerated at less than or equal to 45 degrees Fahrenheit, and goes through distribution.

The chlorine concentration, of course there are many variabilities in these washing operations, and people using different concentrations of chlorines, different temperature profiles, but we should understand that if we just soak an egg in water, a freshly-laid egg in water, we can be dissolving pathogens or fecal matter that may contain pathogens, and basically driving these pathogens into the egg through the pores in the shell.

Of course the regulations now, they inspect fecal matter from henhouses and should be free from salmonella, and if it isn't usually they follow up with certain actions.

So what washing is doing to the goals I just mentioned: For aesthetics, yes, it does remove visible fecal matter from eggs; freshness I would say questionable; shelf life probably; but egg safety I don't think this process really contributes much to egg safety, whether it is external infection or internal infection.

In-shell pasteurization came to take care of the infection problem, especially internal infections. The industry would like to define in-shell pasteurization as a precisely-controlled conductive thermal process, processes designed to effectively address salmonella egg safety concerns. They define that as at least 5 log degrees in the count of salmonella, while maintaining the appearance, texture, and functional characteristics of fresh high-quality shell eggs.

How this was developed originally, that is the patent that resulted in in-shell pasteurization, or one of them, basically they were inoculating the eggs with salmonella enteritidis, and initially they were really inoculating the eggs outside the yolk.

If you reach with the inoculum inside the yolk you usually puncture that membrane, and there may be a problem. So they stayed just outside the yolk and inoculated there. Later on, subsequently they did inoculation into the yolk, but this is originally how it was done.

Then eggs went through certain water bath at different temperatures. The temperatures they used, they are 56 to 60, and kept it at different times until they felt confident that they can reduce up to five logs, and they checked the produced eggs for counts of salmonella and quality like pH and other properties.

Now the process is practiced this way: They transfer the eggs to a pasteurizer, preheat, and that takes some time. The eggs are staying in water until they reach the hold temperature. Then once the internal temperature is about 56, they keep these eggs there anywhere from thirty to forty-five -- it should be thirty to forty-five minutes, I apologize for the mistake on the transparency. That translates to about a five-log reduction, then they are cooled, and the rest of the process. So it is a lengthy process, and it involves keeping the eggs in water for a long time.

How in-shell pasteurization meet these goals that I mentioned earlier. For aesthetics of course it will remove fecal matter and other problems. For freshness it has been claimed that it is close enough to fresh eggs, or nonprocessed eggs. Shelf life probably will improve. But the safety is the real concern, and we know that this can take care of internal infections, and of course it can take care of external infections. Heat does work, we know that.

Then there are alternative technologies that I would like to spend some time on that, still talking about in-shell processes.

Ozone can be used. Pulsed light, there is a technology there where they can pulse flashes of light to the eggs. These flashes are about 20,000 times the intensity of sunlight, and after a few flashes you can reduce the population, the internal population of salmonella in the egg white more than five logs, so it's pretty promising. But nobody really knows the quality of the eggs coming out of that. There's only one company, or a few people who are really playing with this. It's very hard to come up with equipment that you can test it and verify it.

Irradiation has been tried. It does work, but research shows the quality of the eggs are not that great.

High pressure, talking high pressure technology is coming up. We are talking about pressurizing things up to a hundred thousand psi or even more, and the engineers that I work with have convinced me that at hydrostatic pressure you can put an egg in there and it stays intact. I didn't believe them, and I tried that; unfortunately all the eggs cracked. They blamed it on the air cell inside the eggs, but I know that some others probably have tried it and succeeded.

The eggs that were contaminated with salmonella that has been high pressurized, they came out free from salmonella, but they were half cooked because the process also is high-pressure, nonthermal, but it does produce alterations in the properties of eggs.

Combination treatments are very promising. It's very nice to combine heat with something else. Since we know heat works, then you can use it at less intensity, but you combine it with other factors.

I'll spend a little more time on ozone since this is the work I have been doing over the past four or five years, and we would like to call this cold sanitization of shell eggs. We don't call it pasteurization because we know that we cannot really pasteurize eggs with a sanitizer, a strong sanitizer like ozone.

Ozone as you know is as natural as rain and thunderstorms. In fact, this is what you smell after thunderstorms because of the freshness of rain, and you smell it all the time if you are sitting like myself next to a laser printer or a Xerox machine. So it is not bad to use something that natural in a process like this.

We tried I would say hundreds of experiments, and I'm just presenting those that seemed to work really the best.

We contaminated eggs externally, we infected them externally. I mean by that is taking warm eggs that has been washed, dipping them in cold salmonella enteritidis solution, and let salmonella get sucked into the shell. Usually it doesn't pass the membranes, the shell membranes.

Then we take these eggs and subject them to gaseous ozone under a little pressure, ten to fifteen psi for ten minutes, and this is what we got with this experiment. The control was about ten to the sixth. We are inspecting and analyzing the shells only. We separated the shells from the contents, analyzed the shells. The control shells contained about ten to the sixth. After pressure with no ozone somehow we get less recovery. We know that ten, fifteen psi doesn't kill anything, but somehow we got lower recovery, but in the presence of ozone we got nothing on the eggs, which simply means we have eliminated more than five logs of externally-infected eggs.

People are not happy with ten minutes of pressurization. A line of egg processing goes very fast, and they said "Can you do this in one minute?" so we tried externally-contaminated eggs again, in this case we have to combine the ozone with something else, and we tried UV light actually, copying those guys who are using flashes of light, but they had been using white light. We are using UV light, and we can see some reduction due to UV light, but there is a synergistic effect it seems to me between cells that has been exposed to UV light and then exposed later on to ozone.

In this case UV light was done for one minute, and ozone was done for another minute, so a total treatment time of two minutes. This gives the control ten to the sixth; ozone alone for such very short you find about one log reduction. UV alone about two and a half log reduction. The combination about four and a half log reduction.

So one can use such a thing maybe in sanitizing eggs, again taking care of all the external contaminants.

The summary of the results that we have, I probably don't need to go over every piece of information there, but extremely infected eggs we managed to get more than five logs in anywhere from ten to twenty minutes of exposure to ozone gas, and when we have a combination of UV light and ozone gas two-minute treatments produced about 4.3 log reduction.

How this process affects the goals that I set earlier, for aesthetics since we don't dip eggs in any water you probably -- if there are fecal matter on the eggs probably it's going to stay, but we advise that maybe you should wash it first in ozonated water before we do that process that we mentioned.

For freshness, we haven't tested that. Shelf life, we are in the process of testing for that. Safety, we know that we can take care of external contaminants and external infection, salmonella that comes through external means, but for internal infection we are still working on it. We are seeing pretty good results that I'm not presenting today.

These are, after four or five years of working with eggs, and after 25 years working with other pathogens I feel that this is the kind of challenges that are facing shell egg safety research right now, how to validate that a new technology is working.

The problem is facility. Can you go just with eggs that are highly-contaminated with salmonella and run it in any of these processes and say let us try this, then if eggs break then they have a contamination inside. It becomes a problem.

The other problem is naturally- versus artificially-contaminated eggs. We noticed that inoculation of eggs with salmonella doesn't produce exactly the same thing that happens naturally. Naturally-contaminated eggs, they have better distribution of cells in the yolk, and there are many other differences.

I believe also susceptibility of these cells inside the egg is different if you have naturally-contaminated versus artificially-contaminated. But to get naturally-contaminated eggs is very difficult unless you know that the flock is really infected, or if you infect some hens purposely to produce eggs infected with salmonella enteritidis, a very difficult task.

So if we are going for artificial contaminants what media should we suspend these cells in? Is it a buffer? Is it suspended in egg yolk? What do we do, and what phase of growth do we do. Do we stress these cells before we do that? There are all sorts of questions.

And when you inject this salmonella enteritidis into the egg, do you inject it into the white, or do you go all the way to the yolk, and when you inject it into the yolk what rupturing the membrane of the yolk will do to the experiment.

The other issue that also bothers me is disrupting the natural defenses of the egg, and I'll talk about this in a little bit more details. We know that this is approximately how the egg looks like. Forgive my poor drawing here.

And if you look at the shell, this is the first defense that any microorganism getting into the egg has to face, the cuticle, the little thin tiny layer outside the egg. It seems the shell because the shell is very porous, so the cuticle seals the shell and it does provide protection for at least a hundred hours or so. After that I think the protection of the cuticle is gone.

The membrane, shell membranes, there are two of them, even though my drawing says three it should be two. This functions as a physical barrier to prevent microorganisms, but many microorganisms really can handle this. It's a matter of time and concentration of cells. If you have enough of these cells they will break this membrane and get inside the egg.

After that the albumen, the white has lysozyme which is known to be antimicrobial, it breaks the walls of gram positive bacteria. There may be antibodies coming from vaccination or other means in the white that should provide some protection.

Avidin which combines biotin, biotin is needed for the growth of some microorganisms. If you combine biotin you're probably preventing these microorganisms from growing in the white, other compounds like ovotransferin which binds iron which may be needed by many gram negatives, so that the white is quite hostile to invading microorganisms.

These defenses, they grow weaker and weaker as the egg gets older, but if you weaken any of these defenses during processing of the eggs it may not be a good idea.

The yolk itself is very, very rich in proteins, fats, minerals, vitamins, an ideal medium for growth of microorganisms. Luckily it is that innermost layer of the egg; otherwise would have more problems.

There is a physical barrier around the yolk which is the membrane, but the yolk itself may contain antibodies that is coming through vaccination.

So all these natural defenses, what do we do to these defenses when we process eggs. That is the question I think we should be asking and we should be focusing on.

We know that if we inject salmonella in the white and incubate this egg for three days the green ball shows that salmonella dies over this period of time, and we inject the salmonella in the yolk and incubate it at the same period of time we have growth of salmonella. There's no secret about that. So white provides a defense line for the egg.

Other challenges like new practices that is coming, and we need definitely to study these -- continuous washing. If the line shows signs of fecal matter on washed eggs, maybe just divert the line and go back and do another round of washing.

Reminds me with the reworking which has been criticized heavily in other industries, like in the dairy industry and the meat industry, reworking is the cause of many, many problems. Is that reworking causing any problems? Is that the first time around if you didn't wash right you may have salmonella getting deeper into the egg, and the second wash would not really do much. We don't know.

So there are potential problems. Also stress adaptation which I'm very, very interested in makes me worry about how much stress we are giving the microorganisms the first round, and if we go the second round are these microorganisms responding at all to that reworking process.

Repackaging which is basically before expiration date take the eggs and we wash them again, I would say this is bad practice but should be studied before I make my judgment.

So in conclusion current practices and new processing technologies for shell eggs should be evaluated against clear goals, and hopefully these practices will allow us to maintain or even benefit from the natural defenses in eggs, and we better also use some new technologies in microbiology to address it to the egg safety that I didn't see much of that research recently.

Stress-adaptive response, sustaining and visualizing techniques, it may provide new answers for old questions that you see in literature all the time, and facility for running egg safety research with similarity tot real world there is a huge need for that, and trying to build one it's very difficult.

Any questions?

Yes, sir. Can you use the microphone, please?

DR. MITCHELL: I was wondering on your ozone treatment what levels, you know, how many ppm of ozone you were using for that.

DR. YOUSEF: We tried also some ozone concentrations, and we ended up with ozone in the gas at more than 10 percent of the gas mixture. That's pretty high.

DR. MITCHELL: It's going to be a few thousand ppm?

DR. YOUSEF: In the gaseous peers. In the water phase we can get 20, 25 parts per million and get about similar results.

DR. MITCHELL: Okay. I didn't mention, my name is Bailey Mitchell, I'm with the Southeast Poultry Research Lab. Thank you.

DR. YOUSEF: Other questions?

[No response.]

MS. SNOWDON: Thank you for summarizing things. I'm Jill Snowdon with the Egg Nutrition Center, and I need some clarification. I'm not sure I was understanding one of your points, and that will lead me to a comment that I want to make sure that it's clear, and that is when you did that nice evaluation and taking a look at the different components and how different technologies impact either the aesthetic qualities or the external or interior safety, and you were talking about washing, the general washing and sanitation practice that's going on in the industry now, were you coming to the conclusion that that was not contributing to external safety aspects?

DR. YOUSEF: Well, I'm saying that washing as far as affecting the natural defenses, it definitely eliminates the outside cuticle, but if the wash water contains high enough sanitizer any dislodged fecal matter will be taken care of before they have a chance to cause internal contamination.

How much that washing process eliminates salmonella that got already in the egg by other means, fecal matter sucked into the egg while the egg is being laid, we don't really know the answer to that, and I doubt if it affects that, but microorganisms on the outer surface of the eggs should be taken care of by the high levels of chlorine and the temperature combination, and the high pH is a very, very important factor in eliminating salmonella on the outside of the shell.

MS. SNOWDON: That's what I wanted to hear, because we don't want to lose any of the gains that we have gained on public health protection with the washing and sanitation that we're doing in terms of the external, so your concern is there might be something in the shell itself or the interior.

DR. YOUSEF: My concern is the wash water, if I'm trying wash water that doesn't have any sanitizer first I think that's not right, because I may be dissolving fecal matter and getting it into the holes of the egg, and that can be a problem.

MS. SNOWDON: Thank you.

MR. BRACKETT: Are there any other questions for Dr. Yousef?

MS. CURTIS: Pat Curtis, North Carolina State University.

In your schedule or your diagram where you're showing the wash process, was the washing time and temperatures that gave, was that for a single wash system, a double wash system?

DR. YOUSEF: It was for a single wash system using Diamond washer.

MS. CURTIS: Most of the processors now use double wash systems, and there's a little difference in time there. It looked like the rinse temperatures were also a little bit high, but the comment I wanted to make about the wash water, the pH, the wash water is recycled and the pH is mainly to take care of bacteria that would come off of the egg in the recycled wash water, and there's a number of studies that have been conducted regarding wash water, and temperatures, and cold water washing, and a number of those areas that weren't brought out in this that I think are important aspects that we need to consider, because when we look at temperatures of those eggs during that process that's an important concern is how much temperature is being picked up in those eggs during the process.

DR. YOUSEF: What I presented is an example of the wash process. Maybe I shouldn't have said it is a typical wash process.

MR. BRACKETT: Do we have any other questions?

[No response.]

MR. BRACKETT: There was one technology that has been studied a lot in the last year that was not addressed yet, and since we have one of the people who have worked on that I would like to ask Pat Curtis to come back up again and sort of summarize some of the chilling technologies that have been done at North Carolina State.

MS. CURTIS: Actually there's two universities that have worked on rapid cooling of shell eggs, and that's North Carolina State and the University of California, and I'll mention both of those.

North Carolina State has spent a lot of time looking at initial processes from washing to the point of packaging and trying to cool the eggs down, and what we have found is that we went around and did a lot of surveys looking at egg temperatures, and we found that the temperature of the egg during processing rises from twelve to fourteen degrees before we put that egg in the carton, and so it then peaks and rises another five to ten degrees after we package them, put them into pallets, and then either put them into coolers or ship them out.

So you've got an extra little peak there before we actually start any cooling process, and if you actually put a pallet of eggs, thirty cases of thirty dozen eggs in a pallet and in the center of that pallet you measure the length of time it takes that egg to actually cool down to ambient temperature can be anywhere from five to fourteen days, depending upon ambient temperature and air movement and, you know, coolers, and those types of things.

And this is important from the standpoint that we know that salmonella enteritidis will grow if the temperature is above 45 degrees there. So both NC State and the University of California have looked at ways to speed up that process of getting the internal temperature of the egg down to 40 to 45 degrees, and what we have done at NC State is we used carbon dioxide as a coolant, and we cool down the eggs, we have a process, we've worked with PraxAir, Incorporated out of Chicago to run eggs through before they are put into the carton, and it takes less than two minutes, and we're getting them down to about 48 degrees, and then they'll continue to cool where that shell was hot and it was going to peak because it was going to continue to heat, at this point the shell is cooler so it's going to continue and in about fifteen minutes after they have been processed they're down to 45 or 41 depending on what your temperature was at that point that you sat.

So that process will be commercialized later this year. It should be at the international show here in Atlanta in January, a regular unit.

The University of California -- and I'll just comment very briefly on this -- has done some research where they're taking and putting them into coolers, and then drawing cold air through the eggs, and it's a little bit slower process, but it is still more rapid than a traditional mechanism. You have to double-stack the eggs because you have to put them into a line and cover them, and then pull the cold air through there, but it does have some potential there of speeding up the cooling of the eggs as well.

So we have worked on the standpoint that if there did happen to be contamination we could control that contamination growth by getting the eggs cooled as fast as possible.

MR. BRACKETT: Thank you. Jill, did you have a question?

MS. SNOWDON: I just wanted a point of clarification, and that is that Humphrey's work at least indicates that SE is not going to grow below 68 degrees for about three to four weeks, so the 45-degree concept I think has to be in that context.

I think I know what you were saying when you said they don't grow below, you know, and I understand the goal there. I'm not arguing that, but I wanted to bring that little detail out that we do have the natural protective mechanisms in the location of the SE in the membrane I think is the current hypothesis in the white next to the yolk.

MR. BRACKETT: And that was Jill Snowdon from the Egg Nutrition Center.

MS. CURTIS: And just one comment on that, and you'll hear Richard Gast a little bit later, but the studies that Richard has done and some of the studies that have been done at Auburn University of inoculated eggs has not shown the same thing that has happened with Humprhey.

We have seen that you have been able to maintain the live salmonella within that, and that in some cases it has grown according to some things that we have seen at Auburn.

MS. SNOWDON: The Egg Nutrition Center has a request for proposal out to take a look at it so we can get the data published.

MR. BRACKETT: Any other technical questions that we have for the last speaker?

[No response.]

MR. BRACKETT: Okay. Fortunately, we are a little bit ahead of time, which I think is fine. We will take a break now, and then reconvene back in here at ten-thirty.

Again, we have coffee as well as drinks in the back, as well as donuts and that sort of thing in the back. Please avail yourself to them, and be back here promptly at ten-thirty.

[A brief recess.]

MR. BRACKETT: Okay. It is ten-thirty, if you could begin finding your seats we will get started with the next section.

The next section of information that we are going to receive deals with Objective 7.3, and that really is involving the research to improve testing methodology of SE on the farm and in the eggs, and we would like to stress that the testing that is being looked at is both for individual foods as well as environmental.

This morning to speak about methodology we will have Doug Waltman who is with the Georgia Poultry Lab to discuss some of the methodologies.

STATEMENT OF DOUG WALTMAN, GEORGIA POULTRY LAB

MR. WALTMAN: Thank you.

I appreciate the opportunity to share with y'all an area that is a passion of mine, although my technicians would use the word obsession a little more than passion.

I have been asked to address Objective 7.3 which deals with the research to improve the testing methodologies for SE both in the environment of the farm and in the eggs, and there's five components of this objective dealing with the sampling protocols -- this is the section and collection of the samples themselves, the screening tests or how we detect SE, the development of rapid tests which would greatly help the turn-around time, molecular methods for subtyping which would deal with the epidemiology, and then the identification of virulence factors.

I'm going to specifically address the first four of these, and hopefully Dr. Gast and Dr. Petter will address the virulence factor aspect of this in a following talk.

If we first look at the sampling protocols as they deal with the environment we can look at several programs that have been well established, for example the Pennsylvania Egg Quality Assurance Program. Normally they focus on the manure areas, whether it's pit or scrapers, and we'll talk a little bit more about these housing types, the egg machineries, and these walkway samples.

Now, there is published data from the SE Pilot project which preceded the Pennsylvania program, and they summarized their SE isolations from these various sample sources, and it really didn't make a lot of difference, from fans which was their lowest isolation of SE of about 12 or 13 percent to the walkways which was around 18 percent.

But the way I would like to have seen it analyzed was on a per-house, or a group-of-house basis. For example, if the walkways were all positive out of twenty houses, then you wouldn't need to do any of these other sampling types. But on the other hand if say five of the houses were positive by the walkway, ten by the manure pit, and then another five by egg belt then we would need to do all of these different sampling types, and to my knowledge this data was not analyzed with regard to that procedure.

Another resource that we have is the NAHMS study that is in its final stages. I understand that the final report will be out hopefully this fall, and again they looked at the very similar sampling areas, and I hope that they when they analyze their data that they will do it with respect to particular sources so that we can begin to determine which ones may be more effective than others, and if we can get by with just one.

The FDA trace-back data is another, in my mind a very good resource because it would be on a national level. There's a number of houses that are in this. They have also looked at a few other sampling types, and as I understand it that data has not been analyzed by source, but it would serve to hopefully answer some of these questions about the sample source.

Now, when we talk about sampling layer houses we have a fundamental problem, and that is because of the tremendous diversity of these housing types and the equipment in those, and let me just illustrate this this way: Let's say for example over here we have a house with five or eight thousand birds, here we have one with 80,000, and over there we have one with a quarter of a million birds. Here we have a high-rise deep-pit house, this one we have a shallow pit one-tiered house, we have a manure belt with scraper system here, we have a flush system here. We might have one that is completely environmentally controlled over here, this one doesn't even have walls. This one is hand gathered, whereas this one is completely automated.

So this tremendous diversity can happen even in one state, so when we get to the point where how do we sample these kind of houses the situation is that there's no way presently of developing a standard protocol for sampling, and that is a concern.

For example, the FDA trace-back folks list all of these different housing types, and each of them has its own sampling protocol. And trying to put that on an equivalent basis that we're sampling these houses equally is very difficult at best, so there is a need to come up with a better method to sample these houses, hopefully to put them on the same plane.

Now, there is another problem of major concern at least in my mind, and it's that first one right there, the high-rise deep-pit house, and this could be the most common housing type. This is a situation where the birds are actually on for example the second story, their manure falls down to ground level, the manure domes up, and that's what's called the manure pit, and you have fans and ventilation down there that dries that out, and you have some type of composting that goes on.

But in order to sample that you actually have to get down in that pit and you drag these gauze pads the complete length of the house on top of those domes of manure. Now, if you have never been there that is an experience, I'll call it hazard because it's very dim down there, and to some extent that's not bad because there's things down there that you don't want to know about.

But of primary concern is water accumulation, whether that's from rain, or a leaky drinker, or even the evaporative cooling system that regulates the temperature where the birds are, you can get set up areas that are analogous to quicksand, only this is with manure, and I personally have been in over my knee, and I know of an individual that went in over his head, and it's a very dangerous situation, and there's other hazards down there as well, so from my perspective I'm going after something that I can replace that sample with, and that's my focus, what my focus has been on.

I did a study that was funded through U.S. Poultry and Egg, and that full report is available through Dr. Charlie Beard, and we looked at a variety of different sample sites, even more than what is listed here, trying to determine what would be the best sources either singularly or plural.

What we found was that as other people have shown it's not difficult to find salmonella in layer houses. That is fairly common, and you would expect that given the fact that these birds have been in that house, if this is the end- of-lay testing which this is, they've been in there about two years without antibiotic treatment, without a lot of cleaning and disinfection going on because they are food producers.

Now, disconcerting to me from a research standpoint, but good news for the layer industry here in Georgia, we didn't find SE. As extensively as we looked at it we didn't find SE in any of these houses, and so the data that I'm presenting is for generic salmonella.

Now, I don't have any reason to believe that salmonella enteritidis would respond differently, but I cannot confirm that aspect. And we can see from this that the walkway swab for example detected all of the positive houses as well or better than the manure areas, and just slightly less on a per-sample basis than the manure pit, and certainly both of those were better than the egg machinery swabs and these other dust-type of samples.

If we consider the research needs as I see it we still need to have a valid comparison of SE positive houses to determine what source or sources that we actually need to sample.

If we can get away with just the walkways, or just the egg belts, then we don't need to be sampling these other things. From a cost and labor standpoint that would be very beneficial. Also, from a hazard standpoint it would be nice if we didn't have to get down there in those pits.

A subnote of that is that it would be nice to be able to find a sample that would allow us to put all of these different housing types on the same equivalency, such as it would be nice if the walkway sample panned out.

Along these lines we need to determine the optimum number of samples. There is one program that you can go into a house of 80,000 birds and come out with five samples. Is that enough to tell you the true situation in that house?

And then determine the effectiveness of cooling samples. Most of these studies that I've shown use individual samplings, but I know the California groups have looked into this area of cooling samples, and this would cut down on the number of samples that are tested.

Now, we are going to be switching back and forth between environment and egg, and keep in mind these are totally different situations from a microbiological standpoint. The sampling protocols for eggs are pretty set. It's just the number of eggs that varies. The trace-back when they test eggs it's a thousand eggs, I think the Pennsylvania program depending on the situation is 480, a thousand, or 4,000 eggs.

These are collected usually at random through the house, the environmentally-positive house, the shell is sanitized, aseptically cracked, and the contents, the entire egg contents are pooled into a bag, and twenty eggs to a pool, and we'll show you how these are done shortly.

That's the sampling protocol. If we look now at the screening test, and we keep in mind that a screening test is a very sensitive, usually not a specific test, that then after we screen we then do something in addition to that to actually confirm it, and the question some people ask is "Well, why are you looking for SE in the environment when the egg is what we are concerned about?"

Well, again, this is a screening test. We're using the environment to show the likelihood, or to increase the likelihood of the houses where the eggs might be contaminated.

Now, before I get into the environmental sampling which is what is being done now let me just touch base a little bit on antibody testing, because this is another at least technically feasible way of screening for SE. The problem is it hasn't panned out yet. There are several reasons why antibody testing has not been useful.

One, there's a lot of other salmonella in these facilities as was already said, there's a lot of cross-reactivity that goes on with salmonella so you can have some specificity problems, but also because a lot of these salmonella are not very invasive, or tend to localize very quickly you get a very marginal antibody response in many cases, so not only do you have specificity problems but also sensitivity issues as well.

Now, as part of that NAHMS work they looked at an antibody test I conjunction with that survey, and perhaps they will have some different data to share with us later.

But for the most part everyone is looking at the environment, or environmental culture for the screening test for SE, and this is pretty standard. This is the Pennsylvania program, you add a selective enrichment which is usually tetrathionate, you incubate that overnight, and then you inoculate selective plating media, and then you screen salmonella-suspect colonies. That's typical of most programs, for example the FDA group.

Again part of my research project was not only to get some insight into the samples that may be better than others, but also the culture method, the actual way that we can isolate salmonella, and we looked at eleven different isolation protocols as you can see here. Again this is generic salmonella, we did not find SE, so the data has to be looked at from that viewpoint.

These computer-generated slides didn't like this slanted version, so I'll have to share with you what some of these are.

Notice that there are variations in these different methods as to the percent of salmonella they recovered. This procedure right here is similar to what many laboratories are using. At least in our hands this tetrathionate conya is somewhat inhibitory.

The best procedure incorporates a delayed secondary enrichment aspect, but a problem with that is that this procedure takes ten to twelve days, and in most settings that is unacceptable because a quicker turn-around is needed, especially for example in C&D where they need to know now if it's still there so they can do the process again.

So we looked at the possibility of combining a couple of these, and these over here on the left are the singular versions, and then we combined the initial tetrathionate with the delayed, and once again it did produce the highest recovery.

This so-called BAM method incorporates the preenriched tetrathionate with the preenriched rappaport, and this is similar to the LAC-approved method for heavily contaminated raw poultry, and we see that it's in sort of the same ball park, but we can cut out five to seven days with this method over this one, and more than likely this will be the accepted method for culturing these environments.

If we look at specific research needs just to sort of go back over these, as I've pointed out it would be nice to confirm whether or not SE does respond like all these other salmonellas. There is a need for rapid detection of SE, and I specifically mean a rapid detection for salmonella enteritidis. We don't need a rapid method for salmonella because salmonella are present on these farms; we need one that will be specific for salmonella enteritidis.

Now, that's sort of the challenges that we have to deal with with the layer farm. We've got this huge level of background flora from which relatively speaking salmonella is in very low numbers, and even perhaps stressed or sublethally injured, so we've got to pull these few salmonella from among the forest, and even then we've got to then screen or identify whether or not they're SE or some of the other two thousand serotypes.

Well, the situation is different with the eggs. We don't have that massive background flora, but we do have several challenges with the eggs as well.

Early work in Britain by Humphrey showed that the vast majority of contaminated eggs were contaminated with very low numbers, less than twenty, or even less than ten organisms per egg.

Now, you add to that the USDA risk assessment said that one in 20,000 eggs are contaminated here in the United States, very few in number. And then several studies have documented that SE contamination is intermittent and sporadic. So you get a situation such as this: You've got a hundred thousand hen layer farm producing 80,000 eggs a day. Today you might get one positive egg, tomorrow none, the next day three, no more for say seven days, you might get one, and that's the kind of situation you have. You don't have a situation where five thousand eggs are being produced each day with salmonella enteritidis. I think you begin to start seeing this needle-in-haystack scenario that is developing.

It actually gets worse than that, and Dr. Benson can verify what I'm about to share with you. Remember we're pooling twenty eggs, each egg has roughly 50 mls of egg contents, depending on the size, so after pooling twenty of these you've got a liter of egg material, and you've all broken eggs and you know how viscous and what you have to do in order to homogenize or to mix that up, you've got to beat the daylights out of, so it's very difficult to mix that, get a homogenous mixture, and then you have to incubate that, so we've got fifty of these bags, or jars, or whatever container this liter of eggs is in, and we've got to incubate that, and the standard way is room temperature because there's few labs in this country that have the incubator space for this volume of material. So we do it at room temperature for at least three days.

We then inoculate two selective plating media from this, and again we are going into this liter of material usually with a swab, and then we streak these plates.

So if for example we had one contaminated egg with ten cells in it, we put it in a liter of material, that would have to multiply to roughly around one to ten million organisms in order for us to be able to detect it on these plates, so the old needle-in-a-haystack scenario becomes very probable when you compare it with the situation of trying to find SE in eggs.

Dr. Gast looked at the sensitivity of these procedures. This direct process is what I have been describing, and you can increase the sensitivity of detecting salmonella from eggs through enrichment methods, but each time you go through these processes you increase the labor involved, the cost involved, and the turn-around time as well, and with eggs especially we need to be able to get an answer as quickly as possible.

So that's the screening test, or how we're actually detecting the salmonella right now, and there is a need as I have already mentioned for these rapid tests, and there are a slew of them on the market. I could use not only my fingers but most of my toes in telling you of all these commercial kits that are already available for detecting salmonella.

The problem is that they were developed for food and food products, and they have the LAC approval, and they work very well with that setting and in that situation, but the environment of layer houses are entirely different.

The Arkansas group looked at three, the reveal, the bind, and a filter method, and their conclusion was that they did not recommend these rapid detection methods in their present state of development.

This group looked at a genus-specific PCR, and we don't even have to look at the results of their research to tell you that it's not going to help us, because remember we said that salmonella is present in these layer houses, so a yes-no test is not beneficial because we've still got to culture it and then determine whether or not it's enteritidis or not, so this is not of any use for us.

We also looked at six different kits. This is the best isolation here culturally, this is the BAM from which all of these are sort of evaluated against, and you see that they performed comparable to the BAM, but again all this is telling us is yes-no salmonella, and it really doesn't help us because of the level of salmonella that's there.

In a food where .1 percent of the samples may be positive for salmonella, rapid kits are very effective because they screen out the negatives, you only culture the positives, but in a situation in a layer house where 50, 75, maybe even a higher percentage of samples are positive these rapid kits that are generic do not help us.

So the research needs, we do need the rapid kits, we need to increase our turn-around time, but they must be specific for salmonella enteritidis.

With eggs there has been some work looking at kits from the antigen capture elisa formats, as well as PCR, but again just because of the matrix that eggs, those massive pools, all of these require some type of enrichment, and even to antigegen capture elisa, most of these require a level or around ten to the four or higher in order to detect them, so when we talk about rapid it's not in the sense that we normally think of rapid. We may be cutting out a day or so with these, but again they are salmonella specific.

And then the final component is molecular methods for subtyping SE. Again, salmonella is a huge group of organisms antigenically. There's well over two thousand serotypes of salmonella, of which salmonella enteritidis is one. We already have one method of dividing SE, and that's the phage type, and you've heard phage type 4, phage type 8, et cetera.

But within those we don't have, or it would be nice to have a method of dividing those isolates out for epidemiological purposes. For example, if we had forty isolates of phage type 4 it would be nice to know if they were all clonal or if they were of diverse origin.

So various methods have been used, plasmic profiling, ribotyping, pulse field gelalit, electrophoresis, and random amplified polymorphic DNA typing, or rapid.

All of these have been shown to work in various laboratories to be able to discriminate between isolates of SE. One of the problems, though, is that if I have forty isolates of SE and I do ribotyping on them for example I may get eight different groups, and if I do the rapid procedure I may get 25 groups, and this group and this group don't have any relationship at all, so then it's almost like apples and oranges how I compare this, and certainly when different investigators try to compare their results with one another it's a confusing mess.

So it's not that they don't work. What we need is the acceptance of some kind of standard. Whatever we choose, if we standardize it then we can start comparing the results from various locations, from various laboratories, and start making some broader epidemiological statements, and it perhaps would be beneficial to have one laboratory that's doing this testing. That way you don't have reproducability problems, you don't have differences in perhaps the way the method is being done; you have a central repository that is looking at this data.

So what I have tried to share with you briefly -- it's not really briefly I guess -- is the current status of our detection, our monitoring and detection program, and to also try to share with you at least from my opinion of what some of these research needs may be.

And if there's time I'll take any questions.

MR. BRACKETT: Thank you.

Yes, Peter.

DR. HOLT: Peter Holt, USDA, Southeast Poultry in Athens.

Could you go over the BAM technique that you talked about which seems to be the accepted procedure?

MR. WALTMAN: Well, again, the BAM, this is an FDA protocol, it's what is used for testing food and food products. It's the accepted standard for example that everything else is judged against these rapid kits or whatnot.

And depending on the food type, that method may be different, but for example for heavily-contaminated raw poultry the procedure that is recommended is preenrichment followed by tetrathionate and rappaport baccilioitis. Okay. You preenrich the sample, and then you inoculate both tetrathionate and rappaport, and then you go through the plating and the processing from there. It's a dual enrichment procedure.

MR. GODFREY: David Godfrey, Georgia Tech Research Institute.

Has there been any interest or any studies of airborne sampling for either generic or salmonella enteritidis specific?

MR. WALTMAN: Well, in layer houses there has sort of been word of mouth detection. It has been shown that SE is airborne transmitted, the group Peter was mentioning, or Bailey was mentioning where you can infect birds by an airborne route. I don't know that anyone has shown that in a field situation. I have talked with the CDC individuals, and they know of no situation where for example a worker in a layer house has been infected with SE by that route.

MR. BRACKETT: Thank you, Doug.

As you can see, we're moving from the more practical into the much more theoretical, and in some cases much more difficult questions.

The final research objective actually dealt with more of the fundamental questions that affect all of the other previous ones, and that is Objective 7.4 which is to conduct research to understand the ecology and the epidemiology of salmonella enteritidis in the hen and farm environment.

Again, we will have two individuals from the Agricultural Research Service to talk about that. First is Richard Gast who will discuss more of the ecological aspects.

STATEMENT OF RICHARD GAST, ARS, USDA

DR. GAST: Good morning.

Actually as Bob alluded to it is sort of interesting in approaching these objectives in the order that they're published we're ending up possibly at this point in the program having what might turn out to be sort of an introduction, because much of what I'm going to do is really along the lines of an overview.

I'm a little bit raspy this morning. I was commenting to my colleagues on the way over here, having already passed out copies of my slides to all of you we could just turn the lights on and do this like and eighth grade social studies class and go around and have each of you read one of them out loud.

Just a stray thought, and one that I suppose will be disposed of immediately.

Over the past few years as most of your quality assurance and risk reduction programs have been developed, if we look at these we can see that these programs tend to involve what I guess we might call the broad spectrum strategy of approaching the problem of controlling SE in eggs by applying a coordinated series of responses across the entire continuum that goes from breeder flocks, to egg-laying flocks, on to the processing, storage, and preparation of eggs.

And although it has been argued with I think considerable effectiveness and considerable merit that the post-production arena is an area in which there are many particularly cost-effective responses available to us, nonetheless historically and still at the current time the laying house remains one of the primary battlegrounds for our war on SE.

Accordingly, an assortment of questions, understanding how SE gets into the laying house environment in the first place, how it survives there, where it survives there, the nature of the interplay between the pathogen, the laying house environment and the biology of the chicken that ends up resulting in the production of contaminated eggs, all these kinds of questions are important pieces of the SE control puzzle.

It's interesting looking at the title of this this morning, all of these things in some vague way, all of these kinds of questions are what we tend to end up referring to as the epidemiology and ecology of SE.

To be truthful, I think most of us realize these words are pulled in from other disciplines, they're not probably exactly precise or applicable to this situation, but we do have a general sense of what we're talking about, and at least this provides us a common vocabulary.

What we're really talking about here is put in simplest terms what goes on in the laying house, and how it results in birds becoming infected and contaminated eggs being produced.

What I'm going to try to do this morning is review in a very superficial way some of what I see as the principle issues related to this topic, and then provide some personal opinions about what I think are valuable research areas that are worth pursuing.

It's kind of interesting looking around the audience here, I'm probably not the most appropriate person to give this talk. It's interesting to me looking and seeing people like Andy Rohr and Eric Ebel that were in the trenches quite literally, and unfortunately for them, years ago dealing with this situation, Marilyn Baumer who's dealing with it today, people like Chuck Benson who is probably arguably the father of research on SE in the U.S. -- if his beard gets any longer we'll have to start calling him the grandfather. You don't get the microphone until later, Chuck.

But there are a lot of the rest of you that can and should contribute to this, and I'm somewhat of an outsider to this issue of what goes on in the laying house. I'm a research laboratory guy, but I hope that the perspective I provide at least as a laboratory researcher and as a student of the literature may have some bearing.

I think the principle issues on this topic can really be grouped into three broad categories, first those things that relate to the course of infection in individual birds and how contaminated eggs end up being produced by infected hens; secondly, what the sources of SE are, and; third, the reservoirs that enable the organism to persist, and the mechanisms by which it is transmitted between birds within houses.

I'm going to divide these up into three broad questions and try to look at a little bit of what we may know and what we think we need to know according to those three categories.

First, how does SE infection of laying hens result in the production of contaminated eggs. Much of what we know about SE infections doesn't different them a lot from other paratyphoid infections of poultry. They establish intestinal colonization quite effectively.

This is an experimental infection study from some years ago where we gave laying hens very large oral doses of SE, and you can see we established -- these results are from the period during the first month after we infected the birds -- colonized the intestinal tract quite nicely, spread to internal tissues including the liver and spleen, and of greatest significance for egg contamination, also makes its way to reproductive tissues such as the ovary and the oviduct.

It's also interesting if you look at the birds that are listed here as contact exposed, those are birds that were not orally infected, but simply placed in cages in the same room with the infected birds. You see relatively similar results. This organism is horizontally transmissible, and those perhaps provide us a little bit better model of a natural infection.

As you would expect from the fact it gets into reproductive tissues it also of course makes its way into eggs. It's distinctive about SE that unlike other salmonella serotypes which are at a fairly reasonable frequency historically known to be deposited on the shells of eggs, largely because eggs and feces exit the bird via the same opening, SE is also at a considerably elevated frequency compared to the other serotypes found in the contents of eggs.

We can see this again from an experimental infection study where we sampled shells, yolks, and albumens and found it in all three. Actually if you look at second week there we're finding it in yolks and albumens at frequencies higher than we're finding it in shells, suggesting that internal contamination doesn't seem to be very strongly related to external contamination.

Some good work that was done at the University of Pennsylvania a few years ago corroborated that by finding SE in the developing contents of eggs before the shell was even secreted around it.

A couple things about this that I think you have to keep in mind that are distinctive and in regard to the real-world situation, remember these are experimentally infected hens given giant doses of SE, so there are some things here that are unrealistic.

First of all, the responses are exaggerated. You will never find these kind of responses in naturally-infected, in eggs from naturally-infected birds. The risk assessment data from USDA from several years ago suggested that a one-in-twenty-thousand kind of incidence nationally is more realistic.

Also the kinetics of this are pretty artificial as well. We see all the birds infected at the same time, they're all producing contaminated eggs, or they're producing contaminated eggs predominately for about a two- to a two-and-a-half-week period after infection, then it drops off.

In the field of course every bird isn't going to be exposed on the same day, it's going to roll through the flock, so you won't see that same kind of trend, but you would expect a wider distribution over time.

Interestingly, though, in the field there still tend to be little clusters over time where something is going on that seems to trigger a little burst of egg contamination, whether that means a new group of susceptible birds are being infected at that time, or some management factor has changed susceptibility to the infection and allowed contaminated eggs to be produced.

I should note just a -- actually I can go back -- just a real footnote, and this is an area that that was alluded to earlier, and I don't have time within this presentation to get to, note that in our experimental infection study history both -- and this is an old study, and we have repeated these things several times -- even in very recent studies we see both yolk and albumen being contaminated experimentally at comparable frequencies.

Now, when we see yolk contaminated it's almost invariably associated with external structures of the yolk, the membrane. It rarely seems to be in the contents of the developing eggs. But this is somewhat in distinction to what Tom Humphrey has reported in England in that whole issue of is it in the albumen where it's unlikely to grow, or is it near the yolk where it's more likely to grow is I think not quite as simple as it may be portrayed on the basis of Tom's research.

I don't know if either his lab research or my lab research accurately reflects the reality. I think this is a black box area. It's not part of this presentation, but this issue of where is SE being deposited in eggs is critical to understanding what might happen subsequently during refrigeration.

An interesting characteristic we've seen consistently of SE infections, they are pretty persistent. When we infected as Pete mentioned earlier day-old chicks with moderate doses, around ten to the sixth, you could see that although it was cleared out of internal tissues after the first month post-inoculation it stayed in the intestinal tract for quite some time, even considerably longer than this graph shows. Actually at 24 weeks of age most of these birds had grown -- by that point they were laying hens and were still infected, more than half were still carrying the organism in the intestinal tract at that point. Some of those birds also laid contaminated egg infected as day-old chicks.

Also you can sometimes see very considerable persistence in laying hens, although usually not quite at the same frequency.

Finally, as Doug was alluding to, the antibody response of birds is a characteristic of infection that we keep coming back to wondering what to do with it. Experimentally-infected birds, again given large doses, do produce very large antibody titers easily detected for a long time, six months or more post-infection.

It's very tantalizing to believe that those are out there. The same thing for the contact-exposed birds, by the way, although you can see the elovars, it takes longer for the response to develop, but again significant titers, long duration. It's a very attractive target, but as Doug alluded there are a lot of other factors that influence whether that response is a meaningful target for protection.

Some of the things that I think that are still worth doing, that we still need to know in regard to this first category of SE infections, I think we need to know still more about bacteriological and serological characteristics of hens that are useful for detection.

When we think about detection we're often focusing on the technology itself of finding new primers in PCR tests, things that are related to the details of the test, but we also need to know the details of the infection, what does the bird, what antigens does the bird express, where is the organism found in the bird, and so on, so that we know what to go looking for. This is particularly I think a consequence in the antibody tests.

Secondly, the same sorts of information, details about the infection, how the host responds to it, so that we can look at some of those intervention strategies.

When we think about vaccination the same thing applies. We need to understand how the infection proceeds in the bird, and how the bird responds to natural infection so that we're better able to develop strategies that circumvent that process and prevent infection from leading to contaminated eggs, or, better yet, preventing infection from happening in the first place. But some of that depends on better understanding the infection process itself.

Third, and this is one I guess of more personal interest to me as I alluded to a few minutes ago, better understanding of how SE is deposited in eggs -- where, when, how, what kind of numbers, because that's terribly relevant for understanding all of those post-production intervention strategies such as refrigeration and what effect they're going to have.

The second broad question, how is SE introduced into laying flocks in the first place. An assortment of potential sources, all of which we have discussed many times over the years, the first one that always comes to mind since the organism we know is deposited in eggs it's not surprising to us that it's also vertically transmissible from parent to offspring, and even worse is the fact -- and there's been a nice body of research done by my colleagues at the Russell Research Center in regard to broiler chickens that hatchers are places where there is immense opportunity for rapid and prolific spread of salmonella.

You have birds at the most susceptible point in their entire life crowded together with rapid air circulation, lots of dust and moisture circulating around, it's an excellent opportunity if it was there in the egg, and if any of the material in there is contaminated for a large number of birds to become very quickly contaminated.

However, let's also keep in mind, although this is an immensely important potential source, this is one of the areas in which we know probably both the most information about what's actually there, and have probably arguably the best control program already in place, and both of those, the information and control program are in the guise of the National Poultry Improvement Plan which specifically targets this issue, so we've got a good data base of what's out there, we have constant ongoing monitoring, and we know to what extent I think fairly precisely that this contribution at least from the chick standpoint is leading into SE, or leading the SE problem in laying flocks.

Second, the poultry house environment from previous flocks looks to be a major player in the ongoing problem. There was a very good Dutch study some years ago that looked at when flocks became positive over time, and a very, very significant percentage of them first became positive after transfer and placement into the laying house. And I think there's an increasing emerging consensus that more of the battle, more of the issue that we're dealing with today has to do either with flocks becoming infected because they're put into laying houses that were contaminated previously, or there's some other environmental source introducing into the laying house.

Third, all kinds of things -- I mean every invertebrate and vertebrate that we know of seemingly can carry salmonella either on its legs, or inside itself, and so on, and it's so easy to get trapped into one-dimensional thinking about the flow of salmonella, chickens to eggs to humans in our human arrogance. Because we're the ones that we care about most getting sick we tend to forget that we are also just an intermediate arrow in some other pathway.

Human workers can bring it back into laying houses, humans can transmit it to each other, and so on. We have a very complex picture of the transmission of salmonella amongst all of the potential hosts, including poultry.

Finally, feed is always a potential source. Certainly we know that many feedstuffs, especially those that contain animal products, are potentially contaminated. Feed sampling almost invariably has failed everywhere in the world to show significant levels of SE. It's hard to pin down feed as a source, but feed sampling is another of those needle-in-a-haystack situations. You get a little cluster, a little bolus of contamination somewhere in there, you know, two grams of it in a silo that might be responsible for a problem, but even though everybody agrees that feed is a potential source it's almost been impossible to really identify contaminated feeds.

The Pennsylvania Pilot Project some years ago is still at this point until we have the NAHMS data probably our best available field study that relates to some of these sources issues.

Some of the principal things we learned from that include the fact that first environmental samples, or the presence of the organism in the environment is indeed relevant to whether it ends up in eggs. Looking for it in the environment correlated very strongly with whether it showed up in eggs.

Secondly, above and beyond every other thing that they looked at mouse infestation in houses looked like a major issue. Heavy mouse infestations were very, very consistently associated with a higher likelihood of the environment being contaminated. Lots of nice connections have been shown subsequently by David Hensly and other people in Pennsylvania between the organisms found in these mice and the organisms that showed up in the flocks, and subsequently in eggs, and so on.

And third, one that's kind of ominous for us when we start looking at what we're actually achieving in the laying house, in the Pennsylvania study only 50 percent of the time was cleaning and disinfection effective in cleaning up a contaminated environment. That's extremely critical if we think of that issue the Dutch are arguing that it's the laying house environment that's introducing it to subsequent flocks anyway.

Where do I think can we go with this in the future? First of all, I think we need to continue looking for what the prevalence of SE really is in all those potential sources -- breeders, chicks, rodents, insects, feed, environment after C&D and so on. That's a little bit different from a lot of questions we're asking.

We have been, logically I think, often very interested in asking where can we find SE most efficiently in order to detect infected flocks, so a lot of our questions have been related really to sampling methodology. We wanted to know the best sources in order to identify an infected flock.

That's not the same as identifying which of those environmental sources are in fact the ones bringing it into the flock in the first place, and some attention to looking at those sources and where it is I think may help us understand where it's coming from.

An epidemiological approach secondly, looking for the relationship between the isolates in the different sources, looking at the input sources and the output in terms of eggs, chickens, and/or eggs is important.

We still this far into the game are struggling to find good epidemiological markers that will really distinguish which sources matter. That's I think a really critical point of issue.

Third, some geographic questions and some management questions I think are relevant, because as Doug alluded to a few minutes ago there's considerable diversity of what's going on out there, and it would be nice to know if the kinds of sources that are involved are in fact the same say in California as they are in Pennsylvania, or Ohio, or Indiana, or Georgia, or any place else, and in addition whether they're the same in these very drastically different types of management systems.

Fourth, we do need to know the effects potentially of all the kinds of intervention strategies that we might apply in the laying houses -- C&D, testing plans, rodent control, feed treatments, and so on, on the sources of SE and on the resultant possibility of egg contamination.

The third question is how does SE infection spread within flocks once it gets there. We've got an assortment of potential natural routes that we know of by which birds might become infected.

I mentioned vertical transmission before. The classic mode of salmonella infection is via oral ingestion of organisms from all kinds of sources.

Third, inhalation of either aerosols or dust particles certainly seems increasingly like a possibility, not only because inhalation might lead to respiratory infection, but inhalation in the case of a bird because the nasopharyngeal connection there might be simply another way, or an effective means of ultimately infection via the upper part of the gastrointestinal system as well.

And fourth there are some Japanese folks who have put a lot of emphasis on this one, at least in an experimental context -- I don't know how relevant it is in the field -- is ascending infection up the other direction, either through the gastrointestinal tract or up into the reproductive tract. That works very nicely in the laboratory; I don't if that's any kind of a real world thing or not.

But considering those ways of infecting a chicken you've got an assortment of means it can be transmitted around in laying flocks. Direct bird-to-bird contact of course is a major issue; all kinds of factors, many of the ones I mentioned earlier, both biological ones that are infected themselves, the mechanical ones that just carry it around.

Insects are certainly commonly shown to be carriers of SE in poultry facilities. Mice look like the principal target, though, I think for the most part.

All the things that we might call fomites for lack of a better word, basically every physical thing in there, all the surfaces and equipment in the house can certainly be physical sources of transmission.

And finally one that's been of interest to us in our laboratory as Bailey was showing you earlier, air circulation.

We're going to get a sense of how transmissible these things are horizontally. When we did an experiment a little while ago where we infected two birds in a group of twelve -- these were relatively young chicks, and they were given very small doses, the two that were infected were only given a dose of ten to the third, and then we monitored the other birds after about a week, and with an assortment of isolates the vast majority of the other birds eventually became infected as well. This organism, if the birds are in close enough quarters it's very, very transmissible.

Interestingly enough, this is just a total aside point on here, it's interesting to us that the phage Type 4s that we were so terribly worried about seemed to be a little less transmissible than some of the other phage types. I don't know if that's a meaningful comment or not.

Bailey told you about some transmission cabinet studies that we did in which he was looking at the effect of the ionizers, but in some earlier studies we did just looking at the possibility of airborne transmission we saw that when we infected upstream birds and then looked in the downstream cabinets when we did surface rinses of the exterior feathers from those birds in the downstream cabinets 77 percent of them ended up being contaminated.

Remember, these are birds that there's no possibility of contact. The only contact between the groups is air flow. A third of those birds ended up carrying it in the intestinal tract, and a fair percentage of them ended up having it in internal organs.

I don't know if that 11 percent that showed up in the lungs if that's inhalation into the lungs, or the lungs may also be an end point tissue like spleens. That may be the result of oral ingestion followed by systemic dissemination to tissues including the lungs. I'm not offering this as some sort of defense of inhalation respiratory infection.

In fact, I tend to think, my own gut reaction here is more that the surface contamination is probably resulting from a lot of oral ingestion; hence, the higher level of cecal contamination. Nonetheless, airborne movement is clearly I think a very relevant mechanism for dissemination.

Some of the needs I think for further work in the transmission area include determining the prevalence associated with the various transmission mechanisms -- dust, moisture, rodents, insects -- I think those are all relevant. We should add moisture as an issue. Colleagues at the University of Maryland have been very interested for some years in moisture and water activity levels as a means of perpetuating salmonella survival in broiler houses, and I would presume that many of those same issues apply in laying houses as well.

In fact, any of you who are on Ed Malinson's mailing list, Ed has been very evangelical in making the point that controlling water levels, moisture levels in poultry houses is a very affordable technology, it doesn't require significant rethinking of how we're managing our flocks, and if it's indeed relevant to the salmonella levels it offers us an opportunity for progress without completely technologically changing what we're doing.

I say determine the prevalence associated with the mechanisms here because realistically in field studies it's going to be very hard to document which mechanisms are actually responsible for transmission, but all we can do is look at the potential target mechanisms and try to see which of them seem to be identifiable as places where the contamination is.

Secondly, looking at how current laying house management practices are affecting these kinds of things, affecting the sources and transmission.

And third, I think looking at potential intervention strategies for disrupting transmission, rodent control, dust control by the kinds of things that Bailey is looking at, moisture control, I think those are the three things that come to mind most quickly.

A quick summary here -- this is actually just sort of the general theme of what I've been saying all along -- I think that research to better characterize and understand all these kinds of things going on in the laying house can indeed directly help us come up with some usable tools.

However, as much as I do believe this I think that coming up with some better tools that can help us deal with SE infections in laying houses is a really worthy goal.

We need to remember, nevertheless I think, that what goes on in the laying house is still only a small part of I think our overall targets for how we want to try to control SE.

I think that that broad spectrum strategy is still our best overall option for having long-term success in reducing the problem.

If you look at this realistically, if we look at both the technologies we have available to us, or are proposed in ideas we have for research here, and we look at the technologies available to us for managing egg production, for controlling pathogens in egg-producing flocks, completely eliminating salmonella or any other food-borne pathogen from egg production flocks anywhere in the very near future is not I think an attainable or a reasonably-attainable objective, and so I think we need to think of these things very specifically and only as a component in that broader spectrum strategy.

What I'm going to do now is introduce my colleague Jean Guard-Petter who is going to take this I think to a level of a little bit more intense focus and look at some very specific issues related to epidemiology and ecology questions.

STATEMENT OF JEAN GUARD-PETTER, ARS, USDA

DR. GUARD-PETTER: Thank you for having me here today, and what I think we're all realizing as time goes by is just how integrated the chicken, the egg, and the environment is, and unfortunately with SE I think the devil is in the details in understanding specific control measures, specific things that we can do to reduce the current problem. The topics in microbial pathogenesis with special relevance to egg contamination that I work with are outer membrane complex carbohydrates, primarily a molecule you've heard me speak about before called lipopolysacrite.

I worked on a process that some bacteria can go through called high cell density growth, and for those of you who aren't familiar with this, this involves cell-cell communication between bacterial cells through chemicals that they release in the environment when they're growing.

So this is one thing that we know that enteritidis can do that typhimurium so far has not been demonstrated for typhimurium, but enteritidis definitely can grow to high cell densities.

Whether or not it's doing it in the classical method that relies on a certain thing called the ACL-homoserinlactum [ph] we don't know yet, but I am collaborating with people at Iowa State, Peter Greenburg, to answer that question.

Finally I also studied proteotomes, and this is the changes that occur in protein expression that appear on the surface of the bacteria in response to environmental stimuli, so you can have a major change in proteotome without having a change in the genetics of the organism, because what you're doing is you're entering different modes of gene expression, so this has special relevance to vaccine development.

So proteotomes have a lot of relevance to vaccine development. The high-cell density growth work has a lot of relevance to the development of science-based regulations for better control, and finally also for improved epidemiological monitoring we have proposed analysis of lipopolysacrite structures as a method of subtyping SE.

Now, I just want to show you -- this is using a genetic approach to analyze the contribution that different proteins make to virulence in birds, and it's not that anything we have done here is too different from what's been done from typhimurium. In fact, I rely heavily on the immense amount of work that's been done with typhimurium.

But here we're asking a very specific question, we're asking what is the relative contribution of flagella in this case to virulence of SE. Now, this has been studied in a number of different ways, but what had not yet been done was incorporation of new information that when you have flagella genes in different classes, and there are three classes of flagellin genes that are required to interact with each other to wind up with the molecule for motility, if you have a mutation in the Class 1 flagellin master operai, it actually is integrated then into other regulatory circuits in the cell, and so people had not asked specifically what if you mutate a Class 1 gene, and then compare it to a Class 3 gene, which is the structural gene for flagellin.

And what it turns out is that when you mutate a Class 1 gene, here it's fldD, what we found was a hundred fold increase in oral invasiveness of the organism, and that has not yet been reported.

So what we're finding is that flagellation which is a major out-of-membrane marker on salmonella, on all salmonella except the avian-adaptive ones, flagellation is not contributing to oral invasion, it's absolutely required for what happens afterwards, so once the organism has gotten into the bird it appears to be directly linked into the ability to grow high-cell density.

And a way, another way that we know that these are two separate compartments of virulence, in other words issues involving oral invasion may be quite different from what you need for control of something that has already gained access in the bird and is now growing like crazy.

We also took a look at SipD. Now, SipD is a proT that is a salmonella invasion proT, and in typhimurium they know it's involved in virulence, they have a whole slew of these salmonella invasion proTs. They have investigated extensively, and this is a lot of Jorge Galon's work.

Well, again we go to enteritidis and we find a slightly different picture. Yes, SipD is absolutely required for oral invasion, and if you knock it out you won't get any salmonella in the birds, but if that organism has some way of gaining parenteral or internal access to the animal -- in this case we just inject it -- what we find is that the SipD mutant is not attenuated at all. In fact, it grew in organs a little bit better even than our wildtype strain did.

So we're starting to see ten- and hundred-fold differences between the oral invasion compartment and what happens afterwards.

Now, enteritidis as far as we can tell differs in only one major way from typhimurium in regards to oral invasion, and that has to do with another class of genes that we're working up, and it's called the glucoseal transferasis. Again, the devil is in the details with SE, so I'm not going to go into that, but just to let you know we're now investigating very particularly how typhimurium differs from enteritidis which has a different epidemiological pattern.

So I mentioned that we worked with lipopolysacrite, and I'm not even going to show the structure, the detailed structure to this group. I'm just going to show you what we're doing with it, and mainly we're concerned about how this molecule contributes to the virulence of enteritidis, but not to typhimurium's virulence, because we now know typhimurium doesn't make it. Only enteritidis makes this particular form of what's called lypopoly high molecular weight, lypopolysacrite.

Now, there is another important organism that also makes high molecular weight lypopolysacrite, and that's salmonella typhi, so we know at certain times as enteritidis is going through these bouts of infection and depending on where it is in the bird that it's actually converting and looking more like salmonella typhi at times, so what is the role for high molecular weight LPS, and what we're seeing is mitigation of clinical signs in hens of active infection.

Now, here's what we did. We took wildtype SE and a mutant of SE that cannot make high molecular weight LPS, but in all other aspects is a highly virulent strain, and we challenged some hens, and we did use an intravenous route because we wanted to produce a cluster of contaminated eggs.

And what we found was that at this dose -- and there is some dose specificity to doing this sort of experiment -- at this dose both groups of birds produced about 10 percent contaminated eggs in the size of cluster of eggs that we collected here.

Now, one interesting thing popped up about these eggs. If the organism could not make high molecular weight LPS we saw a huge peak of soft-shell eggs that correlated with egg contamination. If it was making high molecular weight LPS, the shell remained in good shape as far as we could tell, because within my little lab we don't have anything fancier for judging eggshell quality than our technicians' subjective assessment.

So here's what we found was that about 39 percent of the eggs following the day one of challenge from those receiving the mutant that can't make high molecular weight LPS were soft, and it was so obvious they were soft, some of them were like lizard eggs, some of them just smashed as they were collected, and so what we're wondering is you see a tiny, tiny little bump here. Now, that may or may not mean anything at all, but what if strains that are making high molecular weight LPS are sub-clinically altering shell quality just enough that maybe we could use improved technology on a high throughput basis to assess egg shell quality.

Now, this is a different sort of correlation with a change in shell quality than the classic we cracked the egg shell and the organism got in. This is a change in egg shell quality that comes about from the bird having picked up an infection.

Now, you'll see here that we get another little cluster of soft shell eggs. In this experimental model what is happening is these birds are suddenly beginning to increase egg production, and so that may be an artifact of the experiment to model, but what we definitely see is at least an uncoupling of contamination with the change in egg shell quality if the bacteria can make that special form of LPS that I've talked about before.

Now, the other thing we know is that the egg is a selective environment for strains of SE that produce typhi-like LPS, and let me tell you how we determine this.

We do a lot of chemical determination of serotype, not immunological. We are not interested in that one little sugar that determines Group B or Group D; we look at all the other sugars, which there are a lot more sugars on that LPS, so what we're doing here is we're plotting the amount of rhamnose against the amount of glucose in LPS.

Now, typhimurium produces an average LPS structure that clusters right here where you actually see a cluster of structures from SE. Now, all of these data points represent a different SE isolate, and so here we see a nidus or a focus of structure, and from that structure then there is a diaspora of structures coming out of it.

Now, this is a quadrant analysis where we're actually correlating where the LPS structure falls with the virulence outcome in birds.

Now, this typhi structure here has been the one associated most with high levels of egg contamination in our experimental animal challenge model, and what we have found, though, is that if you take an egg isolate and store it for more than a year it starts losing a lot of these sugars on the LPS, and it will fall down into this range. But all you've got to do is pass it back through the bird and the egg will eventually select back out then for the typhi form.

Why is that? Well, the typhi-like LPS that we deal with acts as a capsule for this bacteria. I think everybody here is pretty familiar with, or most everybody here is familiar with the fact that capsules always impart some sort of survival advantage to SE.

Now, remember what I said, SE makes this, typhimurium does not. So we have this now molecular marker for strains that have particular ability to do this.

Now, we actually now know that we can manipulate this glucosilation -- as you can see here it's glucose on this Y axis -- we can manipulate glucosilation by the growth conditions by letting, by giving some stresses and it will pop up, and most of these have to do with something that happens in the egg, which is the egg has a very basic pH, the white does, so we apply stresses that are the same as either oxidative stress, or alkaline stress. They both induce the same set of genes that kind of overlap and intermingle, and so we have a feel now for what it is in the egg that might be contributing to the problem.

We know that what winds up is that the molecule on the surface changes to resemble something that is associated with human adaptation of salmonella to people, because as you know typhi is adapted to the human population.

Now, all of these little red marks here, these all came from mouse organs, mouse spleen. The squares came from chicken organs. We don't see the chicken organ isolates popping up into the high glucose range. The mouse spleen is the richest source of LPS structural diversity I've ever seen.

Whereas the egg we know we've got a good fifty-fifty chance of recovering the typhi molecule, the organs of chickens we know it's probably just going to be the average typhimurium structure, the mouse is all over the board. It is spreading all sorts, forms of isolates out there in the hen house.

So one of the research needs I can visualize is someone actually studying mouse populations specifically following the -- as much as we do say of chicken population. It could be there are some dynamics of salmonella infections in mice that are escaping our detection methods in the way that we view mice right now. Nobody is doing epidemiology in mice. However, I do think there's going to be some ways to do it.

So anyway, some applied research needs. I think that we can modify some existing equipment, some existing technology to assess shell quality, and it has been suggested to me that laser air puff technology would be appropriate.

Now, this is amazing technology, and it would be for assessing shell quality. A huge flat of eggs could come through, this tiny little jet stream of air under high pressure comes out, and it puts a dimple where it hits the egg.

Now, a good shell should barely be affected at all, but the laser comes along and measures this dimple. Okay. So you get a readout, a digital readout, and this could be very high throughput, you know, thousands of flats coming through here at a time, and then you see a flat coming through perhaps from a flock or a farm, and if all of a sudden that baseline pops up then perhaps maybe we're encountering one of tho