Lethal white syndrome

Lethal white syndrome (LWS), also called overo lethal white syndrome (OLWS), lethal white overo (LWO), and overo lethal white foal syndrome (OLWFS), is an autosomal genetic disorder most prevalent in the American Paint Horse. Affected foals are born after the full 11-month gestation and externally appear normal, though they have all-white or nearly all-white coats and blue eyes. However, internally, these foals have a nonfunctioning colon. Within a few hours, signs of colic appear and affected foals die within a few days. Because the death is often painful, such foals often are humanely euthanized once identified. The disease is particularly devastating for breeders because foals are born seemingly healthy after being carried to full term.

The disease has a similar etiology to Hirschsprung's disease in humans. A mutation in the middle of the endothelin receptor type B (EDNRB) gene causes lethal white syndrome when homozygous. Carriers, who are heterozygous, that is, have one copy of the mutated allele, but themselves are healthy, can now be reliably identified with a DNA test. Both parents must be carriers of one copy of the LWS allele in order for an affected foal to be born.

Horses that are heterozygous for the gene that causes lethal white syndrome often exhibit a spotted coat color pattern commonly known as "frame" or "frame overo." Coat color alone does not always indicate the presence of LWS or carrier status, however. The frame pattern may be minimally expressed or masked by other spotting patterns. There also are different genetic mechanisms which produce healthy white foals and have no connection to LWS; another reason for genetic testing of potential breeding stock. Though there is no treatment or cure for LWS foals, a white foal without LWS that appears ill may have a treatable condition.

Contents

Clinical signs

Unlike some coat color dilution lethals, which may result in premature births, stillborn, or weak foals, foals born with lethal white syndrome appear to be fully formed and normal.[1][2] The coat is entirely or almost entirely pure white with underlying unpigmented pink skin.[1][2][3] Pigmented regions may be any color, and if present are most common around the muzzle, underside of the barrel, and the hindquarters or tail.[1] The eyes are blue. A few lethal white foals have been shown to be deaf.[2][4]

Healthy foals pass meconium, the first stool, soon after birth, though some healthy foals may require an enema to assist this process. But the meconium of LWS foals is impacted high in the intestine, and never appears, even with the use of enemas.[2] Symptoms of colic begin to appear within the first day,[1] and all LWS-afflicted foals die within the first few days of life.[3] The painful and inevitable death that follows usually prompts veterinarians and owners to euthanize foals suspected of having lethal white syndrome.[5][6]

Death is caused by an underdeveloped part of the digestive system. The large intestine of the horse is a complex system where most digestion takes place, and comprises the cecum, the colon, and the rectum.[7] Necropsies on lethal white syndrome foals reveal a pale, underdeveloped colon[1] and intestinal obstruction (impaction).[3] Samples of affected tissue show a lack of nerves that allow the intestine to move material through the digestive system, a condition called intestinal agangliosis.[1][2][3]

Closer examination of the skin and hair shows both to be unpigmented, and most hair follicles are inactive and many are devoid of hair altogether.[2] All Lethal White Syndrome foals test homozygous for a genetic abnormality.[8]

Inheritance and expression

Genetic conditions which affect more than one physical trait - in the case of Lethal White Syndrome, both pigment cells and enteric nerve cells - are termed pleiotropic. The unusual instance of pleiotropy in LWS foals suggested early on that the syndrome was related to an important section of embryonic tissue called the neural crest.[3] As the name suggests, the stem cells of the neural crest are precursors to nerve cells. Another cell type that descends from neural crest cells are melanocytes, pigment-producing cells found in hair follicles and skin. The migration of nerve- and melanocyte-precursors from the top of the embryo to their eventual destinations is carefully controlled by regulatory genes.[9]

Such regulatory genes include endothelin receptor type B (EDNRB). A mutation in the middle of the EDNRB gene, Ile118Lys, causes lethal white syndrome.[4][10] In this mutation, a "typo" in the DNA mistakes isoleucine for lysine.[10] The resulting EDNRB protein is unable to fulfill its role in the development of the embryo, limiting the migration of the melanocyte and enteric neuron precursors.

In the case of LWS, a single copy of the EDNRB mutation, the heterozygous state, produces an identifiable trait, but with a very different outcome from the homozygous state.[11]

To produce a foal with lethal white syndrome, both parents must be heterozygotes or carriers of the mutated gene.[6][12][13] Without genetic testing, some carriers are misidentified as having white markings due to another gene, while some are even classified as solids.[2]

The presence of this gene in a variety of horse populations in North America suggests that the mutation occurred in early American history, perhaps in a Spanish-type horse.[4][5][6]

Heterozygotes

Horses heterozygous for the Ile118Lys mutation on the equine endothelin receptor type B gene - carriers of lethal white syndrome - usually exhibit a white-spotting pattern called "frame," or "frame overo."[4][6][8][12] Frame is characterized by jagged, sharply defined, horizontally oriented white patches that run horizontally along the horse's neck, shoulder, flank, and hindquarters. The frame pattern by itself does not produce white markings that cross the back, or affect the legs or tail. It does, however, often produce bald faces and blue eyes.[4][6][8] The term "frame" describes the effect of viewing a frame-patterned horse from the side: the white markings appear to be "framed" by a dark-colored border.[6] To date, animals who are heterozygous carriers do not exhibit health concerns associated with carrier-only status.[14] Natural processes of DNA repair are likely responsible for weeding out EDNRB proteins made from the mutated instructions.

Not all horses with the heterozygous mutation exactly fit the standard visual description. A horse with the Ile118Lys mutation on EDNRB that is not readily identified as frame-patterned is called a cryptic frame. In addition to cryptic frames, a significant proportion of horses with the frame phenotype are visually misidentified, even in clinical settings.[8] One study found from a group of visually inspected registered Paints, 18% of breeding stock solids and 35% of bald-faced horses were actually frames.[8] On the other hand, over one quarter of Paints registered in the "overo" category were not frames, and conversely, 10% of horses registered as tobiano also carried frame genetics.[8] The difficulty in accurately identifying frames has contributed to the accidental breeding of Lethal White Syndrome foals.

Minimally marked horses heterozygous for the Ile118Lys mutation are not uncommon: one DNA-tested Thoroughbred has white markings limited to a bottom-heavy blaze and two socks below the knee.[15] A Quarter Horse mare tested positive for the gene after she and a frame Paint stallion produced a Lethal White Syndrome foal; the mare's markings were a thin blaze with a disconnected white spot in the right nostril, with no other white markings.[2] One major study identified two Miniature horses that were completely unmarked, but were positive for the Ile118Lys gene.[8]

There are multiple theories for this. Variability in the percentage of individuals with a specific genotype that express an associated phenotype is called penetrance, and this may simply be evidence of variable penetrance.[16][17] Several research groups have suggested that other, "suppressor" genes may limit the expression of frame-pattern white spotting.[12][18]

On the other end of the spectrum, some white-spotted horses are so extensively marked that the character of the frame pattern is masked. In particular, the tobiano pattern, a dominant gene, is epistatic to overo.[16] Other white-spotting genes include splashed white or "splash," sabino, and "calico."[8] Any combination, or all, of these white-spotting genes can act together to produce horses with so much white that the presence of frame cannot be determined without a DNA test.[19]

Ambiguous terminology has also contributed to the confusion surrounding this disease. Currently, the American Paint Horse Association (APHA) categorizes horses as tobiano, solid, "overo," and tovero.[20] The association breaks down "overo" into three categories: Frame, Splash and Sabino.[20] In the past, "overo" was used even more loosely, to refer to spotted animals that were "Paint, but not tobiano."[20] However, no fewer than four—and likely many more--genetically distinct patterns are included under the term "overo."[8][18][21] To be categorized as "overo" by the APHA, a horse must fit a written description: white spotting does not cross the back, at least one solid-colored leg, solid tail, face markings, and irregular, scattered, or splashy white patches.[22] To further complicate matters, various Sabino patterns also appear in some horse breeds that do not carry genetics for frame or any other spotting pattern.[23]

Likewise, official classification of a horse as an unspotted solid is based not on genetic testing, but on a visual description. Horses carrying genetics for frame and other white-spotting patterns may be so minimally marked as to lack the registry's minimum requirements for white.[24] This helps to account for allegedly solid horses producing spotted offspring, called cropouts.[6]

The long-standing practice of categorizing Paint horses in this manner contributed to the incorporation of the word "overo" into some of the titles used to describe the disease, such as Overo Lethal White Foal Syndrome.[3] However, "overo" is a term that applies to several genetically unrelated white-spotting patterns, and only the frame pattern is indicative of the syndrome.[6][8][12] The confusion about the nature of LWS is then furthered by statements such as "there are many overos that do not carry the lethal allele,"[20] which is technically correct, but only because the term "overo" also encompasses splash and sabino patterns as well as frame.[20]

Homozygotes

Homozygotes for the Ile118Lys mutation on the equine endothelin receptor type B gene have lethal white syndrome.[4][8] In any crossing of two carrier parents, there is a 25% statistical probability of producing a solid-colored, living foal; a 50% chance for a frame-patterned, living foal; and a 25% chance of a LWS foal.[13]

Producing frame color patterns without producing lethal white

Spotted coat colors, including frame, are popular and sought after by breeders.[25] While many lethal white syndrome foals are accidentally produced when breeders cross two untested cryptic frames, or a known frame and a cryptic frame, some are produced by the intentional breeding of two known frames, whether out of ignorance or indifference. Producing a foal with lethal white syndrome is now completely avoidable. This is because most major animal genetics labs now offer the DNA test for LWS. Whether a horse visually appears to have the frame pattern or not, testing horses of frame or "overo" lineage is highly recommended.

The statistic likelihood of producing a living, frame-patterned foal by crossing two frames is 1 in 2, or 50%; the same odds of producing a living, frame-patterned foal from a frame-to-non-frame breeding which carries no risk of producing a lethal white syndrome foal.[13] Therefore, breeding two frame overos conveys no benefit to breeders hoping to produce another frame overo.

Dominant or recessive?

Lethal white syndrome has been described by researchers as both dominantly-inherited and recessively inherited.[26][27][28] Lethal white syndrome can be described as recessive because heterozygotes (written Oo or N/O) are not affected by intestinal agangliosis. The term "carrier", by definition, applies only to recessive conditions. However, if the frame pattern trait is included in the lethal white syndrome, inheritance of the trait follows an incomplete dominant pattern. It should be noted that the concept of "recessive" and "dominant" predate molecular biology and technically apply only to traits, not to genes themselves. In pleiotropic conditions, such as lethal white syndrome, the application of "recessive" or "dominant" can be ambiguous.[29]

A separate issue is the nomenclature applied to the frame pattern itself. While it follows a dominant pattern of inheritance, deviances occur.[30] The majority of horses with the Ile118Lys mutation do exhibit the recognizable frame pattern, but a small percentage are too modestly marked to be classified as "spotted" by breed registries. Such "solid" horses, bred to a solid partner, can produce classically marked frames.[31] The "crop-out" phenomenon can make frame appear to follow a recessive mode of inheritance.

Prevalence

The gene for lethal white syndrome is most common in the American Paint Horse, but occurs in any breed that may carry frame genetics, including American Quarter Horses, Appaloosas, Thoroughbreds, Morgan horses, Miniature horses, Tennessee Walking Horses and Mustangs, as well as horses that are descended from these breeds. Only two Morgan horses have been identified as frame overos.[32] Breeds that do not carry genes for the frame pattern also do not carry LWS.[33]

Lethal white mimics

Foals affected with lethal white syndrome are not the only white, blue-eyed horses. There are other genes that produce healthy pink-skinned, blue-eyed horses with a white or very light cream-colored coat. For a time, some of these completely white horses were called "Living Lethals," but this is a misnomer. Before reliable information and the DNA test were available to breeders, perfectly healthy, white-coated, blue-eyed foals were sometimes euthanized for fear they were lethal whites,[13] an outcome which can be avoided today with testing and a better understanding of coat color genetics. The availability of testing also allows a breeder to determine if a white-coated, blue-eyed foal that becomes ill is an LWS foal that requires euthanasia or a non-LWS foal with a simple illness that may be successfully treated.

Analogous conditions

From very early genetic research,[3] Lethal white foal syndrome has been compared to Hirschsprung's disease (HSCHR) in humans, which is also caused by mutations on the EDNRB gene. Various polymorphisms on this gene result in intestinal agangliosis, in some cases attended by unusual pigmentation of the skin and eyes, and deafness. The occasionally attendant pigmentation condition in humans is called Waardenburg-Shah syndrome.[4]

The terms "piebald-lethal" and "spotting lethal" apply to similar conditions in mice and rats respectively, both caused by mutations on the EDNRB gene.[34][35] Only lethal in the homozygous state, the mutations are associated with white-spotted coats, deafness, and megacolon caused by intestinal agangliosis.[4]

References

  1. ^ a b c d e f Vonderfecht, SL; Bowling AT, Cohen M (January 1983). "Congenital intestinal megacolon in white foals". Veterinary Pathology (The American College of Veterinary Pathologists) 20 (1): 65–70. PMID 6849219. 
  2. ^ a b c d e f g h Lightbody T (September 2002). "Foal with Overo lethal white syndrome born to a registered quarter horse mare". Can. Vet. J. 43 (9): 715–7. PMC 339559. PMID 12240532. http://www.pubmedcentral.nih.gov/articlerender.fcgi?tool=pmcentrez&artid=339559. "The mare was registered as solid chestnut with a star, strip and snip, a dark spot above her nostrils, a white spot in the right nostril, and no other white markings." 
  3. ^ a b c d e f g McCabe L, Griffin LD, Kinzer A, Chandler M, Beckwith JB, McCabe ER (July 1990). "Overo lethal white foal syndrome: equine model of aganglionic megacolon (Hirschsprung disease)". Am. J. Med. Genet. 36 (3): 336–40. doi:10.1002/ajmg.1320360319. PMID 2363434. 
  4. ^ a b c d e f g h Metallinos, DL; Bowling AT, Rine J (June 1998). "A missense mutation in the endothelin-B receptor gene is associated with Lethal White Foal Syndrome: an equine version of Hirschsprung Disease". Mammalian Genome (New York: Springer New York) 9 (6): 426–31. doi:10.1007/s003359900790. PMID 9585428. http://www.springerlink.com/content/xehe4p28y8p7cw98/. Retrieved 2008-09-04. 
  5. ^ a b Santschi, Elizabeth M. "Overo Lethal White Syndrome". NetPets. http://www.netpets.com/horses/healthspa/lethal.html. Retrieved 2008-05-08. 
  6. ^ a b c d e f g h "Genetic Equation". American Paint Horse Association. http://www.apha.com/breed/geneticeq3.html. Retrieved 2008-09-04. "Recent characterization of the gene involved in the lethal white foals has confirmed that the foals with two doses of the gene are white, and die soon after birth from gut innervation abnormalities. Horses with only one dose are frame overos, and survive." 
  7. ^ "Horse Nutrition - The Horse's Digestive System. Bulletin 762-00". Ohio State University. http://ohioline.osu.edu/b762/b762_5.html. Retrieved 2007-02-09. 
  8. ^ a b c d e f g h i j k Vrotsos, PD; Santschi EM, Mickelson JR (2001). "The Impact of the Mutation Causing Overo Lethal White Syndrome on White Patterning in Horses" (PDF). Proceedings of the Annual Convention of the AAEP 2001 (American Association of Equine Practitioners) 47: 381–91. http://www.ivis.org/proceedings/aaep/2001/91010100385.pdf. Retrieved 2008-09-05. 
  9. ^ Thiruvenkadan, AK; Kandasamy N, Panneerselvam S (2008). "Coat colour inheritance in horses". Livestock Science (Elsevier) 117 (2–3): 109–129. doi:10.1016/j.livsci.2008.05.008. "There are many mutations that may affect the migration of the cells from the neural crest...Very many different mutants that affect migration of the cells from neural crest region cause lethal white-spotting condition. These mutations affect normal melanocytic and enteric neural crest cell differentiation, proliferation and migration during development (Searle, 1968). This leads to depigmentation and also lack of neural connections to the colon." 
  10. ^ a b Yang GC, Croaker D, Zhang AL, Manglick P, Cartmill T, Cass D (June 1998). "A dinucleotide mutation in the endothelin-B receptor gene is associated with lethal white foal syndrome (LWFS); a horse variant of Hirschsprung disease". Hum. Mol. Genet. 7 (6): 1047–52. doi:10.1093/hmg/7.6.1047. PMID 9580670. http://hmg.oxfordjournals.org/cgi/pmidlookup?view=long&pmid=9580670. "sequence analysis, together with allele-specific PCR and the amplification-created restriction site (ACRS) technique, revealed a dinucleotide TC-->AG mutation, which changed isoleucine to lysine in the predicted first transmembrane domain of the EDNRB protein. This was associated with LWFS when homozygous and with the overo phenotype when heterozygous." 
  11. ^ Metallinos DL, Bowling AT, Rine J (1998). "In three unrelated lethal white foals, the EDNRB gene contained a 2-bp nucleotide change leading to a missense mutation (I118K) in the first transmembrane domain of the receptor, a highly conserved region of this protein among different species. Seven additional unrelated lethal white foal samples were found to be homozygous for this mutation. No other homozygotes were identified in 138 samples analyzed, suggesting that homozygosity was restricted to lethal white foals. All (40/40) horses with the frame overo pattern (a distinct coat color pattern that is a subset of overo horses) that were tested were heterozygous for this allele, defining a heterozygous coat color phenotype for this mutation."
  12. ^ a b c d Australian Equine Genetics Research Center. "Overo-Lethal White Foal Syndrome (OLW)". DNA Screening Tests. University of Queensland. http://www.aegrc.uq.edu.au/index.html?page=30053. Retrieved 2008-09-05. "An individual contains two copies of a gene, one inherited from its father, the other from its mother. When both copies of the EDNRB gene are the mutated form, LWFS results. However, when an individual has one copy that is normal (also called the "wild-type" of the gene) and one that is the mutated form, then an overo colour pattern results." 
  13. ^ a b c d Overton, Rebecca (2004-03-01). "By A Hair" (PDF). Paint Horse Journal (American Paint Horse Association). http://www.painthorsejournal.com/archives/pdfs/ByAHair-Mar04.pdf. Retrieved 2008-09-06. 
  14. ^ "Lethal White Overo (LWO)". Animal Genetics, Inc.. http://www.horsetesting.com/LWO.htm. Retrieved May 21, 2010. 
  15. ^ "Orchard View Sporthorses". http://www.orchardviewsporthorses.net/. Retrieved 2009-05-01. 
  16. ^ a b Metallinos DL, Bowling AT, Rine J (June 1998). "A missense mutation in the endothelin-B receptor gene is associated with Lethal White Foal Syndrome: an equine version of Hirschsprung disease". Mamm. Genome 9 (6): 426–31. doi:10.1007/s003359900790. PMID 9585428. http://link.springer-ny.com/link/service/journals/00335/bibs/9n6p426.html. 
  17. ^ Santschi EM, Vrotsos PD, Purdy AK, Mickelson JR (January 2001). "Incidence of the endothelin receptor B mutation that causes lethal white foal syndrome in white-patterned horses". Am. J. Vet. Res. 62 (1): 97–103. doi:10.2460/ajvr.2001.62.97. PMID 11197568. 
  18. ^ a b Santschi, EM; Purdy AK, Valberg SJ, Vrotsos PD, Kaese H, Mickelson JR (April 1998). "Endothelin receptor B polymorphism associated with lethal white foal syndrome in horses.". Mammalian Genome (New York: Springer New York) 9 (4): 306–9. doi:10.1007/s003359900754. PMID 9530628. http://www.springerlink.com/content/k4ud8mlbp5qgwwtn/. Retrieved 2008-04-08. 
  19. ^ Santschi EM, Mickelson JR (2001). 80% of breeding stock white (all-white horses of Paint pedigree) were genotyped N/L for Endothelin Receptor B.
  20. ^ a b c d e "'American Paint Horse Association's Guide to Coat Color Genetics" (PDF). American Paint Horse Association. 2007. http://www.apha.com/forms/PDFFiles/guidebooks/07ColorGen.pdf. Retrieved 2008-08-19. 
  21. ^ Vrotsos, Paul D.; Elizabeth M. Santschi (1998-07-01). "Stalking the Lethal White Syndrome". Paint Horse Journal (American Paint Horse Association). http://www.apha.com/breed/lethalwhites03.html. Retrieved 2008-09-06. 
  22. ^ "Overo Pattern". American Paint Horse Association. http://www.apha.com/breed/overo.html. Retrieved 2008-09-03. 
  23. ^ a b "Horse coat color tests". Veterinary Genetics Laboratory, University of California, Davis. http://www.vgl.ucdavis.edu/services/coatcolorhorse.php. Retrieved 2008-09-06. 
  24. ^ "Color Requirements". American Paint Horse Association. http://www.apha.com/breed/colorreq.html. Retrieved 2008-09-03. 
  25. ^ Brooks, Samantha (2006) (PDF). Studies of Genetic Variation at the KIT Locus and White Spotting Patterns in the Horse. University of Kentucky. http://archive.uky.edu/handle/10225/564. Retrieved 2010-12-29. "Aside from the scientific benefits of studying coat color in the horse there is a more immediate advantage. People like the look of unusual coat colors, and they are willing to pay for a way to improve their ability to get more of them. The value in a flashy colored horse isn’t as easily measured as that in a fast race horse, but it does translate in to real dollars at the point of sale. For example, the average price of an APHA registered yearling (young horses with undetermined athletic ability) without a spotting pattern on Equine.com was $1540. A yearling APHA registered horse with the tobiano pattern averaged $2803 (Appendix B). This price difference creates an enormous interest among breeders to increase their production of spotted foals." 
  26. ^ Finno, Carrie J.; Sharon J. Spier, Stephanie J. Valberg (March 2009). "Equine diseases caused by known genetic mutations". The Veterinary Journal (Elsevier) 179 (3): 336–47. doi:10.1016/j.tvjl.2008.03.016. PMID 18472287. "Ileocolonic aganglionosis, or overo lethal white foal syndrome (OLWS), is an autosomal recessive trait" 
  27. ^ Metallinos et al 1998. "Based on the strength of this association and its complete compatibility with simple mendelian recessive inheritance, we inferred that Lethal White Foal Syndrome was tightly linked to the mutation."
  28. ^ Thiruvenkadan et al 2008. "The overo lethal syndrome due to dominant homozygotes (OO) at the overo locus results in death of the foals a few days after birth."
  29. ^ den Dunnen, Johan T; Antonarakis (2000). "Mutation nomenclature extensions and suggestions to describe complex mutations: A discussion" (PDF). Human Mutation (Wiley-Liss Inc.) 15 (1): 7–12. doi:10.1002/(SICI)1098-1004(200001)15:1<7::AID-HUMU4>3.0.CO;2-N. PMID 10612815. http://mecp2.chw.edu.au/mecp2/info/nomenclature_for_complex_mutations.pdf. 
  30. ^ Bowling AT (1994 pmid=8014463). "Dominant inheritance of overo spotting in paint horses". J. Hered. 85 (3): 222–4. PMID 8014463. http://jhered.oxfordjournals.org/cgi/pmidlookup?view=long&pmid=8014463. "Analysis of selected studbook records of the American Paint Horse Association, consisting of 687 foals sired by 13 overo stallions from non-overo mares, supports the inheritance of overo spotting as an autosomal dominant gene. More than one gene may control patterns registered as overo." 
  31. ^ "Genetic Equation". American Paint Horse Association. http://www.apha.com/breed/geneticeq3.html. Retrieved 2008-09-04. "Frame overo behaves as a dominant gene. [T]here are records of frame overos being produced by two nonspotted parents. This is typical of a recessive gene, and it is not logical to have both a recessive and a dominant control over the same pattern." 
  32. ^ Laura Behning. "Other Colors". Morgan Colors. http://www.morgancolors.com/othercolors.htm. Retrieved 2009-05-03. 
  33. ^ Walker, Dawn (February 1997). "Lethal Whites: A Light at the End of the Tunnel.". Paint Horse Journal. http://www.apha.com/breed/lethalwhites01.html. Retrieved 2008-01-11. 
  34. ^ Gariepy, CE; Cass DT, Yanagisawa M (1996-01-23). "Null mutation of endothelin receptor type B gene in spotting lethal rats causes aganglionic megacolon and white coat color.". Proceedings of the National Academy of Sciences of the United States of America 93 (2): 867–72. doi:10.1073/pnas.93.2.867. PMC 40149. PMID 8570650. http://www.pnas.org/cgi/pmidlookup?view=long&pmid=8570650. 
  35. ^ Hosoda, K; Hammer RE, Richardson JA, Baynash AG, Cheung JC, Giaid A, Yanagisawa M (1994-12-30). "Targeted and natural (piebald-lethal) mutations of endothelin-B receptor gene produce megacolon associated with spotted coat color in mice.". Cell 79 (7): 1267–76. doi:10.1016/0092-8674(94)90017-5. PMID 8001159. http://linkinghub.elsevier.com/retrieve/pii/0092-8674(94)90017-5. 

See also