Trapped Neutrophil Syndrome

[Alfreda]

I came across this in Wikipedia and was curious if the statement that TNS is "widespread" and "old as the breed itself" is accurate? I thought I had read that is was mostly in "show" bred BCs?

 

 

“Trapped Neutrophil Syndrome (TNS) is a hereditary disease in which the bone marrow produces neutrophils (white cells) but is unable to effectively release them into the bloodstream. Affected puppies have an impaired immune system and will eventually die from infections they cannot fight. The mutation responsible for TNS has been found in Border Collies in English working dogs, in show dogs that had originated in Australia and New Zealand, and in unrelated Australian working dogs. This indicates that the gene is widespread and probably as old as the breed itself. TNS was identified by Jeremy Shearman in the laboratory of Dr. Alan Wilton of the School of Biotechnology and Biomolecular Sciences, University of New South Wales. There is no cure, but a DNA test is now available to detect carriers as well as affected dogs.^[17][18]”

 

https://en.wikipedia.org/wiki/Border_Collie

[juliepoudrier]

It is mostly a disease of show dogs because of the line/inbreeding (which of course concentrates both good and bad genes), but since those dogs came from working stockdogs at some point, the show folks have always said that the working population must carry it. And in fact, it has been found in working bred dogs, just at an extremely low rate of incidence. The best way to keep that incidence low is to make sure that inbreeding is also kept low. And if you're someone who tests for it, you wouldn't want to make a carrier x carrier mating.

 

If I had to pay for extra testing, I wouldn't automatically test for it in working stock, but because there are now combination DNA tests available in which the gene for TNS is included, it's certainly easy enough to test. If I were in the conformation dog world, I would definitely test.

 

J.

[mjk05]

It's not really known how widespread it is, but in every population where testing is performed, it seems to be more common than predicted, and it certainly isn't limited just to show dogs.

 

Testing is much cheaper now than it used to be, and I'd definitely be testing from any dogs I thought might be bred from. Getting either a clear or carrier result allows breeding without worrying about producing affected pups. There are some absolutely top shelf working lines with known carriers (Nat. and Int. Sup. Ch.s), so having information about your own lines can be really useful.

 

Of course, there is still the stigma hanging over from past years (TNS = show dogs/poor breeding), and some breeders believe they'd definitely have noticed such a disease in their lines, so there's still reluctance to test or to even consider the issue, and that does make me sad. I heard recently about someone who bought a working-bred pup from a pretty experienced breeder with top dogs and a good reputation on the other side of the country (Australia) and lost it to TNS at about 4 months of age. Apparently it was a very healthy happy looking pup at the time of sale.

 

A couple of interesting pedigrees to check out:

Llangwm Cap, doesn't really need any introduction- advertised semen on the Come-Bye website, TNS carrier

Dot, a Czech dog, all ISDS lines, TNS carrier- info here

Jimmy- info here, TNS carrier

Blaze and Bess - half siblings to Dot and to Jimmy's dam

 

As testing is now cheaper and easier and becomes more common in working lines, I'm sure we'll have more information about its prevalence.

[Mark Billadeau]

Based upon reported affected rates where there have been population studies, the incidence rate is <0.5% of that population. One could extrapolate to other weakly connected gene pools to assume incidence rate is no greater than that in the gene pool where the disease was recognized and determined to be genetic.

 

To put <0.5% in perspective, the incidence rate of CEA is <2.5%. Now that you have numbers you can see what the author of the Wikipedia post means by "widespread" (about 5x less widespread than CEA).

 

My hypothesis is that there was a very low carrier rate for TNS in the UK Border Collie gene pool. When dogs were imported to Australia, by chance, several of them were carriers. This artificially raised the carrier rate in the very small gene pool in Australia leading to a higher than average affected rate eventually leading to people recognizing the disease was genetically linked. This does not mean the carrier rate in the source gene pool equals the carrier rate in a relatively isolated gene pool. However, now that there is more worldwide transfer of genes we could see an increase in the carrier (and affected) rates in the source gene pool or what were geographically insulated gene pools if we are not careful.

[mjk05]

Which population studies are those, Mark?

It doesn't seem that there is widespread testing yet, although obviously increasing.

I definitely agree care needs to be taken, especially when carriers feature among such prominent dogs as recent IntSup Chs (which deserve to have a significant influence on the gene pool). Now the test is so cheap and easy and DNA testing for other conditions (like CEA) is being done by so many breeders worldwide, it seems irrelevant how widespread or not it is, why wouldn't we test?

[Mark Billadeau]

This was one of the first, published in 2011 by AN Wilton. In order to estimate the carrier and affected rates you'll have to plug in the measured allele frequencies into the Hardy-Weinberg equation.

 

Something to keep in mind is how these samples were collected. Who would offer up DNA samples for a study about TNS? Would that sample set be truly random or would it be skewed towards owners who are concerned about TNS in their lines? This is the same issue ALL sample collection efforts (or other health studies) have when trying to use the sample data to assess breed wide frequencies. Based upon this sample bias I typically look at the reported frequencies as upper limits not true representations of the breed wide frequency (for any study with voluntary sample submission).

 

 

A canine model of Cohen syndrome: Trapped Neutrophil Syndrome
Jeremy R Shearman12 and Alan N Wilton13*
BMC Genomics 2011, 12:258

We applied the TNS carrier test to a sample set of 260 Australian Border collies collected before the TNS research began. These samples can be considered randomly sampled with respect to TNS. We also tested a set of 80 samples that were collected in Norway to represent the local Border collie population (phenotype data was not collected, Table 2). The sample sets gave allele proportions of 0.07 and 0.08, respectively, for the TNS mutation. We have tested 5000 Border collies from Europe, USA, UK, Australia, New Zealand and Japan, 1100 were carriers and 30 were affected. This sample set of 5000 has a strong ascertainment bias towards dogs with familial links to known carriers. To correct for this ascertainment bias, we removed any samples where either of the parents had been tested as carriers, leaving 2100 dogs. Allele frequencies ranged from 0.04 to 0.08 per country with an average of 0.064 from the sample set of 2100 dogs (Additional file 1, Table S5).

[mjk05]

This was one of the first, published in 2011 by AN Wilton. In order to estimate the carrier and affected rates you'll have to plug in the measured allele frequencies into the Hardy-Weinberg equation.

So, just to check because it's not my area, an allele frequency of 0.06 (which is what Wilton and Shearman reported in that study) would give a carrier frequency of about 12%, which is about what they've been quoted in other articles (Mason, Jepson, Maltman 2013). And a carrier frequency of 12% would be about 1 in 8.5 dogs?

(edited to remove my erroneous comment about CEA- which apparently has an affected/incidence rate of 2.5%, so a carrier frequency of about 25%, but of course is not fatal, and many genetically affected dogs will have normal vision)

 

Something to keep in mind is how these samples were collected. Who would offer up DNA samples for a study about TNS?

 

Oh, I totally agree with you. They did try to reduce sample bias by eliminating dogs from known carrier parents, but I still really wonder where they got their samples. We did send in samples in 2006 (from dogs related to our TNS pups), and later on our sheepdog organisation did collections for (paid) testing at trials, and most people agreed to use of those samples for research, but I don't know if they were used for this study. And there were a significant number of dog owners who elected not to test, so it was somewhat skewed towards those with lines related to our affected pups. But I'm pretty sure there was no widespread sampling of Australian working dogs outside our isolated state.

 

And anecdotally, I find it hard to believe that there is a TNS affected rate of 1 in 250, which I believe is what you'd expect from the allele frequency you've quoted above (correct me if I'm wrong, not my area). It is obviously poorly understood even now, and almost certainly frequently missed (interesting case from Israel Journal of Vet Medicine here, Gans 2015, where HOD was also diagnosed- that was the working diagnosis for one of our TNS pups also- "TNS is a relatively newly recognized disease entity, which has been identified in Britain and Australia since the 1990’s but relatively little has been published about it and its existence is not widely known in the practicing veterinary community. "). But 1 in 250 AFFECTED? Unless there is massively unrecognised in utero loss as a TNS presentation, I think that's got to be a massive overestimate.

 

But it is almost certainly more common than anyone believed 10 years ago (when I was happily telling people it was a show dog disease, not found in working lines), or even 5 years ago. And when at least a couple of the ISDS International Supreme Champions in the past decade are either tested carriers, likely carriers or close relatives of carriers, I'm surprised that there isn't more widespread testing (or at least publicity about testing) in working dog circles.

 

I'm massively saddened to still be hearing, just in this past year, of puppies bred by conscientious working dog breeders, taken home by experienced dog owners, no issues recognised, and later dying of TNS. It just shouldn't be happening with the available test. But, probably because of the lack of knowledge of the disease or the test among working dog breeders and their vets, and possibly because of that stigma and the persisting belief that this is mostly a disease of show dogs, it will keep happening.

[mjk05]

Massive apologies for my massive overuse of the word "massive" in that last post.

 

For those who are interested in the sorts of dogs that might have carriers in their lines (I know previously there's been some pedigree analysis to try to identify common ancestors), here's another pedigree of a really well bred dog tested as a carrier. And please do check out the pedigrees I linked above. I'm just so glad we have a simple, relatively cheap test that can prevent affected puppies being produced when such quality dogs are used extensively (as they should be).

 

 

[Mark Billadeau]

Using this calculator I get an carrier rate of 11.3% and affected rate of 0.5%

 

http://scienceprimer.com/hardy-weinberg-equilibrium-calculator

[mjk05]

So that's fairly significant for an inevitably fatal disease, no? Much higher than rates of cystic fibrosis in the Caucasian population.

 

If we compare to CEA, the Optigen website lists the rates (from eye testing in the 90s, presumably even lower that) of the severe manifestations of CEA/CH as 0.57% for coloboma, and 0.06% for retinal detachment. So similar to or lower than the quoted incidence of TNS (if we believe that).

Obviously for breeders, CEA is more of a threat because it isn't fatal and genetically affected dogs can breed, significantly raising the allele frequency in the population beyond what is possible with TNS. But from a puppy buyers POV, there may be as much risk (or more) of getting a pup that dies of TNS as one that goes blind from CEA, if parents are untested.

 

That's why it really surprises me that it isn't more publicised for working dog breeders. It isn't even mentioned on the ABCA website page on inherited disease. And so easily tested for!

 

Check out the pedigrees listed above- various prominent, top working lines (honestly, look at them!), not heavily inbred. With international AI breeding so popular now, and top sires producing pups all over the world, the potential for TNS to crop up without awareness and testing must be increasing. Why is this still considered not really worth testing for?

 

As far as the original question, re the word "widespread", I think its intended meaning was "spread across the population"- not necessarily common, but present in border collies around the world.

[Mark Billadeau]

It may not be more publicized because of the sampling bias we discussed previously; the incidence rates in the US and UK working dog populations are likely lower than the populations sampled in the published studies. I suspect that the incidence rate of cystic fibrosis in the Caucasian population than TNS in the various subpopulations of Border Collies to really make that an appropriate analogy (much less sampling bias).

 

The ABCA, to my knowledge, has not heard of a working dog with TNS or a working bred litter that was affected.

 

IGS has been reported (published studies) with a similar incidence rate as TNS; but unlike TNS the ABCA has received reports of IGS in working Border Collie litters.

[mjk05]

Did you look at those pedigrees linked to above, Mark?

[Mark Billadeau]

Beyond there being carriers (at an ill-defined low level) in working bred border collies, what else should we "see" in these pedigrees?

 

There are a few other recognized genetic diseases that are at an ill-defined low level; and I'm sure there are many other not yet recognized genetic diseases carried at a similar low level (such is genetics). The affected rate (and by extrapolation the carrier rate) of TNS (and these other diseases) are not high enough to be of concern as long as breeders are working to keep the COI as low as possible.

 

The focus should NOT be on anyone of these low incidence rate diseases (except for those one suspects of being present in their lines) but on keeping genetic diversity; by breeding for genetic diversity one best avoids crossing carriers of these low incident rate diseases.

[Mark Billadeau]

So let me turn the tables on you. We hear about EOD from our breeders frequently. Based upon the pedigrees for affected (not just carriers) it must be present in UK lines and therefore by extension in lines exported to Australia.

 

Shouldn't you be monitoring for EOD? Are you?

 

What about IGS? Is there a big push in your breeders to test for it?

[mjk05]

Yes, we're aware of EOD and keep it in mind when looking at dogs. If there was a simple cheap genetic test for EOD, then yes, absolutely we would test for it. Yes, we have tested for IGS since a cheap, simple test (like that for TNS) became available. Why wouldn't we? I have never met or heard of a local dog with IGS (or confirmed CEA in our state either, FWTW) but if the test is easily performed and affordable, it seems like a total no-brainer.

 

If a test (such as gonioscopy or screening echo or BAER) was very expensive and logistically difficult (given we are a fair distance from the only city in our state with such facilities) AND the likely yield was not really high (ie those tests for conditions where mode/degree of inheritance is murky) then we would think hard and possibly elect not to do them. But a test like IGS or TNS these days? Of course.

 

Because then (if I had the opportunity) I could confidently use a top dog like Llangwm Cap or one of those listed above (the Dewi Tweed offspring or Nij Vyas' young dog) to outcross to one of our dogs (lovely low COI!) and know that the pups would absolutely not be TNS affected. Wouldn't you?

[mjk05]

Beyond there being carriers (at an ill-defined low level) in working bred border collies, what else should we "see" in these pedigrees?

Some of the more successful and popular working border collies of recent years. Dogs that (rightfully) have a lot of first degree relatives being produced and producing around the working border collie world.

 

The focus should NOT be on anyone of these low incidence rate diseases (except for those one suspects of being present in their lines) but on keeping genetic diversity; by breeding for genetic diversity one best avoids crossing carriers of these low incident rate diseases.

Those don't have to be mutually exclusive, though, do they? We can breed with concern for maintaining genetic diversity AND perform cheap and simple genetic tests for known diseases.

[mjk05]

The ABCA, to my knowledge, has not heard of a working dog with TNS or a working bred litter that was affected.

.

Of course I assume you mean "a working dog... Or a working bred litter that was affected in the U.S."

 

TNS is almost certainly underdiagnosed and poorly understood. The Israeli article I linked to earlier was very interesting for its mention of HOD as a concurrent diagnosis- that's what our later presenting TNS pup was provisionally diagnosed with (by a specialist vet centre with past experience with TNS). Osteomyelitis, gastroenteritis, respiratory tract infections, fading puppy syndrome- all potentially TNS-related. If they're fatal before extensive investigations are conducted, TNS could easily be missed.

[Mark Billadeau]

I see known carriers from well known working dogs without any information to know if the well known working dog was a carrier. This is a flaw in may people holding up pedigrees of carriers; they say see, its in working lines but there were other dogs in the carrier's lines of unknown working ability which may have been the carrier. I'm not saying it is not in working lines; I'm saying it occurs at a low frequency.

 

"the test is so cheap"

Based upon this logic you'd likely want breeders to run the full panel (why should one low frequency disease get preference) of available DNA test on breeding pairs: 2x$288(current discount price for a $480 panel). This is not really that cheap considering puppies run about $600.

 

Perhaps you are sensitized to TNS and based upon your experience "see" TNS more prevalent than it really is. The same thing occurred recently for IGS; people starting reporting IGS in dogs they know about creating the impression of a higher prevalence because people didn't report all their normal dogs. IGS, like TNS, is often misdiagnosed and it can lead to death.

[mjk05]

Orivet do an individual test for AU$75 but a full breed panel including CL, IGS for $150 (AU, so US$107). Compare that to CEA, which has a similar incidence of severe disease- AU$200 for a single test. Even in the past, TNS done individually by UniNSW was AU$55, and our small working sheepdog community negotiated a discounted price for group testing. I imagine a group like the ABCA would have significantly more bargaining power than 30 rural West Australian sheepdog triallers. Once many local breeders identified the majority of dogs that are clear, they rarely test unless using untested dogs.

I don't think I see TNS where it isn't- I am open to the possibility that it's presenting without being identified. I still haven't seen any affected pups other than ours (although I have heard of them, in other working litters).

Those pedigrees have well known working dogs on both sides, and it came from somewhere. The pedigrees don't have one easily identifiable common line. We don't know where it came from. So without testing, we don't know where it is going.

If I were expending significant time and effort in breeding a litter, I would want to take the relatively simple step of DNA testing to make sure I knew where I stood on these easily avoidable diseases.