Cystinuria is a disease involving a defect in the movement of certain amino acids, the building blocks of proteins, into and out of cells, particularly in the kidney. Normally in the kidney, amino acids and some other substances are filtered out of the blood, and then reclaimed from the urine. In cystinuria, some amino acids, including cystine, and the dibasic amino acids ornithine, lysine, and arginine (referred to collectively as COLA), are not reabsorbed by the kidney, and reach high concentrations in the urine. One of the amino acids involved, cystine, has a tendency to precipitate in acidic urine and lead to the formation of kidney and bladder stones (also called uroliths or calculi) that can cause severe illness as they irritate the urinary track system and sometimes cause blockage.

Cystinuria has been described in over 60 dog breeds, cats, maned wolves, and humans. As in many people with cystinuria, cystinuria in Newfoundland dogs and in some Labrador Retrievers is an autosomal recessive trait caused by a mutation in the renal Basic Amino acid Transporter rBAT (Henthorn et al., 2000). DNA-based testing in the Newfoundland is available and allows the detection of affected, carrier and normal dogs. Before DNA-testing was available, a simple, inexpensive qualitative urine test to detect high levels of cystine in urine, called the cyanide-nitroprusside test (NP test) was used to detect affected animals with and without clinical signs. In Newfoundland dogs, the NP test consistently gives a positive result in affected animals and a negative result in unaffected (both carrier and normal) animals, which has been confirmed by comparing NP test and DNA test results. Urine and DNA studies have clearly helped this breed to deal with the problem and informed breedings can now assure healthy offspring.

A potentially confusing issue about cystinuria in all dogs is better understood from examining the clinical situation in Newfoundland dogs. Since the disease in the Newfoundland is inherited as an autosomal recessive trait, we expect that disease incidence will be the same in male and female dogs. This is indeed true in Newfoundland dogs. Examination of both NP test results and DNA test results shows that equal proportions of male and female dogs are affected, as expected. However, when one looks at the data from the stone analysis laboratories (at the Universities of California and Minnesota), it is clear that male Newfoundland dogs form cystine stones that result in a clinical problem (and thus end up getting analyzed) much more frequently than female dogs. This finding is almost certainly due to anatomical differences between the male and female urinary tracts in dogs. So while male and female Newfoundland dogs demonstrate the underlying metabolic defect of cystinuria at equal frequencies (cystinuria without uroliths), affected males are more likely to exhibit clinical signs of the disease that result from stone formation. This situation is almost certainly true for other dog breeds in which cystinuria is found, as supported by the fact that the vast majority of cystine stones (98%) identified in stone analysis laboratories come from male dogs (Osborne et al., 1999; Ling et al., 1998).

Our ongoing studies of cystinuric dogs of many other dog breeds indicate that cystinuria is a more complicated disease than in the Newfoundland and in people. The first indication that there are at least two kinds of cystinuria in dogs again comes from data from stone analysis laboratories (Osborne et al., 1999). In Newfoundland dogs, cystine stones are removed between 6 months and 1 year of age, while the average age of cystine stone removal in all dogs is 4.8 ± 2.5 years. While the lower urinary tract is the most common site from which cystine stones are removed in all dogs combined (97%, Osborne et al., 1999), Newfoundland dogs are more likely to form stones in the kidney than other breeds. We have examined the rBAT gene in several non-Newfoundland dogs with cystinuria (with and without uroliths), and have found no mutations (Henthorn et al., 2000), except in a few cases of cystinuria in the Labrador Retriever.

Past metabolic studies of the urine of cystinuric dogs have in essence been retrospective in that only dogs that have already been diagnosed with cystine stones have been available for study, and few of these dogs have actually had urine studies performed. Unfortunately, such studies cannot answer important questions, such as: what is the mode of inheritance, when can elevated urine cystine first be detected in a dog that will eventually form cystine stones, what conditions are necessary for a dog with cystinuria to exhibit clinical signs, including stone formation, etc. Fortunately, with the cooperation and support of many dedicated breeders, owners, and veterinarians, we are beginning to better document cystinuria in several dog breeds, in particular Mastiffs, English Bulldogs, and Scottish Deerhounds. It is only through these types of studies, in which both affected and unaffected dogs and their relatives are studied over their lifetimes, that the data to answer many important questions can be gathered.

So… what do we know today? Despite extensive and ongoing work at the clinical, biochemical and molecular level, we do not know what gene or genes are responsible for cystinuria in most dog breeds. However, we have observed the following:
– Dogs that have formed cystine stones leading to clinical signs (blockage) before the age of one year suffer from the type of cystinuria caused by a mutation in both copies of the rBAT gene. Conversely, for breeds in which cystine stone formation does not occur in the first year of life, mutations in the rBAT gene have not been found.
– Cystinuria in Mastiffs and English Bulldogs (two breeds in which cystinuria is more common than other breeds) does not appear to be caused by mutations in either rBAT or in an additional gene known to cause cystinuria in humans.
– Not only is it predominantly male dogs that experience cystine bladder stone formation, but the vast majority of Mastiff and Scottish Deerhound dogs that have had a positive NP test are males (>10:1 male:female). This is in contrast to the Newfoundland breed, in which males and females have equal proportions of NP test positive dogs among dogs tested.
– The patterns of excess (above normal) COLA amino acids in the urine of Mastiffs and Scottish Deerhounds (and probably other breeds with late onset of clinical signs) vary widely and are generally less extreme than seen in the Newfoundland dogs. This variability is seen in several ways. For instance, urine cystine levels vary over time in individual dogs, not only within a single day, but from day-to-day, and also with age. Absolute levels of excess amino acids in the urine are also variable, ranging from clearly above normal levels that are only detectable using amino acid quantitation to high levels that are easily detected by the NP test. This variability is NOT seen in affected Newfoundland dogs.
– Several dogs shown to have above normal levels of the COLA amino acids in their urine (by urine amino acid quantitation, an expensive but accurate quantitative analysis) have not formed cystine stones. On the other end of the spectrum are dogs that have formed life-threatening and recurring cystine stones, but their urine sample did not show excess COLA amino acids by quantitative amino acid analysis.
– Examination of Mastiff, English Bulldog, and Scottish Deerhound pedigrees does not reveal any pattern that is completely consistent with a simple mode of inheritance (autosomal recessive, X-linked recessive, or autosomal dominant). However, our current data, including the two preceding points, can be explained by X-linked recessive inheritance with extremely variable expression of the disease (some might refer to this as incomplete penetrance). The expression of other genes or environmental factors may influence the expression of cystinuria. Whatever the terminology and causes, in practical terms it means that not every dog having the mutated version of the gene that is needed to cause cystinuria will be diagnosed using current tests for the disease. While we have not yet proven to our satisfaction that the disease is inherited as an X-linked recessive trait, our current data is best explained by this hypothesis, and thus forms our working hypothesis for the time being.

These observations raise many issues. Perhaps one of the most pressing is what are the implications of X-linked inheritance of cystinuria for breeding choices? In simple X-linked recessive inheritance, the gene involved is located on the X-chromosome, the X-chromosome being present in two copies (XX) in females and only one copy (XY) in males. In order for the disease to be exhibited in a female, both copies of the gene must be mutated. If a female has one mutant version and one normal version, she will NOT be affected, but on average will pass the mutant version of the gene to 1/2 of her. Such a female is referred to as a carrier (she appears normal clinically and by urine testing, but carries a “hidden” mutant version of the gene). A female with two normal alleles is normal, and cannot pass mutant alleles to her offspring. The terms affected, carrier, and normal female therefore refer to females with 2, 1 and 0 copies of the mutant allele, respectively. More importantly for this discussion, carrier females are indistinguishable from normal females. For X-linked genes, males come in two varieties, affected (with one mutant allele on their single X chromosome), and normal (with a normal allele on their single X chromosome).

Now let us consider the possible outcomes of different types of matings, assuming simple X-linked inheritance of the disease.
• Normal female to normal male – This type of mating will produce only normal offspring.
• Carrier female to normal male - When a carrier female is bred to a normal male, the sons who receive their mother’s mutant version of the gene, also referred to as the mutant allele, will have the disease, since their only copy of the gene is a mutant version. The sons who get her normal allele will be normal. Therefore, on average, 1/2 of the sons will be normal, and 1/2 will be affected. The daughters of the carrier female by normal male breeding will all get a normal allele from their normal father. They will get either a normal allele or a mutant allele from their mother. Therefore, while all of the daughters will be clinically normal, approximately 1/2 of them will be carrier females. There are two important issues to consider here.

o The proportion 1/2 stated above is an average. Each offspring has a 50:50 chance of getting the mutant allele from the mother. Consequently, it is possible, by chance, that a carrier mother will not give her mutant allele to ANY of her offspring, particularly, if she does not produce many pups. (Think of tossing a coin for example three times and getting three heads in a row. You wouldn’t be too surprised.) Consider a litter of six, containing three females. If, by chance, none of the three male puppies received a mutant allele from their carrier dam, this carrier female would not be recognized as a carrier because she did not produce any affected offspring, and her daughters would not be recognized as being at risk of being carriers themselves.
o Since a carrier female is indistinguishable from a normal female, without any additional information, there is no way to know if the normal-appearing female is normal or a carrier. In the case of cystinuria, the only way to identify a carrier female is from knowledge of her parents or offspring. If a female had an affected parent, she is an obligate carrier, having only the possibility of getting a mutant allele from that parent. Or, if the female produced an affected offspring, that offspring had to get a mutant allele from the mother, making her a carrier. This emphasizes why family information is so important.

• Normal female to affected male – The normal female can only contribute normal alleles to her offspring, while the affected male contributes his only X chromosome, which carries the mutant allele, to all of his daughters, and contributes his Y chromosome to his sons. Therefore, all males produced from this type of mating will be normal and ALL females produced will be CARRIERS. Important issues here are:

o Carrier and normal females are indistinguishable, so without additional information, it would not be possible to know if the normal-appearing female used for mating was normal or a carrier.
o The daughters of an affected male are obligate carriers, because all daughters get their father’s only X chromosome, which has a mutant allele in an affected male. This is a case where additional information allows us to differentiate between normal and carrier females.

• Affected female to normal male – The affected female has only mutant alleles to contribute to her offspring, and the normal male has only normal alleles. Consequently, ALL male puppies are AFFECTED (they get their X chromosome from their mother), and ALL female offspring are CARRIERS. Remember that:

o Due to anatomical differences, unlike affected males, affected females are much less likely to form stones that will cause clinical problems, and in the absence of urine testing, are very likely to be undetected.

• Carrier female to affected male – Here, the mother contributes either a mutant allele, or a normal allele, while the father has only a mutant allele to contribute. On average, 1/2 of the daughters will be affected and 1/2 will be carriers. Of the sons, on average, 1/2 will be affected and 1/2 will be normal. Therefore, on average, 1/2 of any litter will be affected. The important issues are:

o As explained above, the 1/2 ratios are the average that would be seen from this type of mating. Within a single litter, the actual ratios will vary. Litters from this type of mating could range from containing no affected dogs to containing all affected dogs.
o Affected females can be produced, even when a disease shows X-linked inheritance.

• Affected female to affected male – All offspring will be affected. In the case of cystinuria, they may not all form stones or develop clinical signs of the disease.

o This is a good place to reiterate the additional complications present for cystinuria in particular. In the case of cystinuria, we know that there can be cystinuric dogs without uroliths, and that uroliths often form for the first time after a dog has been bred. For both of these cases, mutant alleles that are not detected in the parents are passed from parent to offspring.

The bottom line is that the breeding of affected dogs of either gender or of carrier females will transmit mutant alleles to offspring and may produce cystinuric animals. Complicating factors in making breeding decisions are the facts that we do not yet have any estimate of the rate at which dogs with excess cystine in the urine develop clinical manifestations of the disease. We do know that dogs that have had only negative NP test results have formed cystine stones. What this tells us is that, presumbably due to the variability of cystine levels in the urine, a single or even a few NP test results do not necessarily give a complete understanding of renal cystine transport in a dog.

This raises the question of how the NP test should be used. The NP test as an inexpensive screening test for detecting affected animals that are at risk of developing clinical signs due to cystine stones. At the present time, we do not believe that it detects every dog at risk for the two reasons mentioned above: 1) Not all dogs with excess cystine in the urine have high enough levels to give the positive test result and 2) because excess cystine in the urine appears to be highly variable over time. While we have quantitative amino acid analysis performed on some dogs as part of the research program, the cost of this test prohibits its routine use.

What are our cystinuria research priorities? Our primary goal is to understand the genetic basis of cystinuria in all dogs and to develop DNA-based genetic tests. This will allow breeders to preserve valuable dogs without producing affected offspring and to eliminate the mutant cystinuria alleles from all breeds. Molecular genetic, biochemical, and clinical studies of affected dogs and their relatives are necessary and ongoing. To facilitate these studies, we are undertaking the development of urine tests that are more quantitative and more sensitive, but still affordable. We are also collecting clinical and biochemical information from dogs that have formed cystine stones or have had a past positive NP test and entering some of these dogs and their relatives into our research study. Updates and requests for information from NP positive dogs will appear on our website, at http://www.vet.upenn.edu/penngen/research/. A more complete understanding of cystinuria may also lead to the development of new treatments.

Thus, after our initial optimism for the development of DNA-based genetic tests for all dogs, based on the progress made in the Newfoundland, we now have data indicating that the situation in many breeds is much more complicated and differs significantly from the situation in Newfoundland dogs and humans. We are poised to continue the research to further define the clinical course, the mode of inheritance, and the molecular basis of the cystine defect other canine breeds. Our goal is to develop a comprehensive understanding of the disease that will allow us to develop DNA-based tests that can be used to make rational breeding decisions that will allow breeders to eliminate the risk of cystinuria while preserving desirable traits in their breed. Much more work is needed and the participation of owners and breeders in this process is essential and greatly appreciated.

Henthorn, P.S., Liu, J., Gidalevitz, T., Fang, J., Casal, M.L., Patterson, D.F. and Giger, U. (2000) Polymorphism in the canine SLC3A1 gene and identification of a nonsense mutation in cystinuric Newfoundland dogs. Human Genetics 107:295-303.
Ling, G.V., Franti, C.E., Ruby, A.L., and Johnson, D.L. (1998) Urolithiasis in dogs II: Breed prevalence, and interrelations of breed, sex, age, and mineral composition. Am. J. Vet. Research 59:630-642.
Osborne, C.A., Sanderson, S.L., Lulich, J.P., Bartges, J.W., Ulrich, L.K., Koehler, L.A., Bird, K.A. and Swanson, L.L. (1999) Canine cystine urolithiasis. Vet. Clinics of North America: Small Animal Practice 29:193-211.