Letters to the Editor, Book review, Notice 349 the Pitt-Rogers-Danks syndrome may be or (2) involvement in the activation loop of stature syndromes associated with mutations caused by a deletion in the same region.5 7 the kinase domain.'4 in this are explained by loss of gain of In a recent issue of the JYournal of Medical Thus, the mutations for short limb skeletal function. We believe the jury is still out and all , Partington et al' reported on a dysplasias on FGFR3 (, will be revealed in good time. number of patients with deletions or duplica- , and ) However, we must protest at Cohen and J Med Genet: first published as 10.1136/jmg.35.4.349 on 1 April 1998. Downloaded from tions of 4pl 6.3, adding new information are gain of function mutations that "put the Neri's attempt to extinguish our overgrowth potentially useful for characterising this brakes on even more" to various degrees. syndrome by the rather cavalier manipulation segment of the human genome. Fgfr3 +/- heterozygous mice have been shown of the growth data presented in table 2 of our However, in reading their article "Translo- to be normal.'" paper. They have chosen to ignore two cations involving 4p16.3 in three families: In conclusion, the idea put forth by explicit points stated in the text: first that deletion causing Pitt-Rogers-Danks syn- Partington et a?' that FGFR3 in single dose II.12 in family 1 is a special case because of drome and duplication resulting in a new leads to growth failure and in triple dose to severe disease (empyema), which probably ", we have serious con- physical overgrowth is not tenable in view of limited growth in childhood and, second, that cerns about (1) their definition of a new over- current clinical and experimental evidence. overgrowth "became more obvious in late growth syndrome and, more importantly, (2) M MICHAEL COHEN JR adolescence and early adult life". their idea that a triple dose of FGFR3 results Department of Oral and Maxillofacial Pathology, What Cohen and Neri have done is to take in physical overgrowth. Department ofPediatrics, Dalhousie University, our mixed longitudinal and cross sectional First, we have rearranged the data from Halifax, Nova Scotia, Canada growth data, translate it all into cross their table 2 to show that overgrowth is not GIOVANNI NERI sectional data and, neglecting age, treat these really a prominent manifestation of duplica- Istituto di Genetica Medica, Universiui Cattolica, measurements in the same way. This does tion involving FGFR3 (table 1). For both Rome, Italy dilute the overgrowth patterns observed. height and head circumference, fewer than However, if one omits the data on II.12 for > half of the patients have values 90th centile; 1 Estabrooks LI, Breg WR, Hayden MR, et al. the reasons stated and takes the measure- only with respect to weight do slightly more Summary ofthe 1993 ASHG ancillary meeting ments at the oldest age of the 10 remaining than half of the patients have values 90th "Recent research on chromosome 4p syn- subjects (all over the age of 15 years), then all centile. On the other hand, a good proportion dromes and ." Am Y Med Genet 1995;55:453-8. the heights are at or above the 75th centile of patients have values 650th centile and 2 John RM, Robbins CA, Myers RM. Identifica- and five ofthem are above the 90th centile. In even 25th centile. Combining all three tion of genes within CpG-enriched DNA from the same way, the head circumferences are all parameters (n=76), at the extremes, 28% of human chromosome 4pl6.3. Hum Mol Genet at 1994;3:161 1-16. or above the 50th centile with five above patients have values 97th centile and 8% 3 Thompson LM, Plummer S, Schalling M, et al. the 97th centile. This tells quite a different have values < 3rd centile. Although some val- A gene encoding a fibroblast growth factor story from table 1 presented in Cohen and ues for all three growth parameters are large receptor isolated from the Huntington disease Neri's letter. and the trend appears to be in that direction, gene region of human . Genom- ics 1991;11:1133-42. Some confirmation of this late growth pat- so many patients have middle and lower 4 Wright TJ, Ricke DO, Denison K, et al. A tran- tern emerges from the heights recorded on range values that overgrowth per se does not script map of the newly defined 165 kb II. 11 and Il. 12 from family 1. Thus, between seem to us to be a particularly prominent Wolf-Hirschhorn syndrome initial region. Hum Mol Genet 1987;6:317-24. the ages of 18 and 28 years 5 months, II.11 manifestation of dup(4p 16.3). In fact, when 5 Clemens M, Martsolf JT, Rogers JG, Mowery- grew 15 cm and between 17 years 5 months thinking of classical overgrowth syndromes, Rushton P, Surti U, McPherson E. Pitt- and 28 years 11.12 grew 26 cm. Such growth such as Beckwith-Wiedemann syndrome, Rogers-Danks syndrome: the result of a 4p increments are commonplace between the Simpson-Golabi-Behmel syndrome, or microdeletion. Am J Med Genet 1996;66:95- 100. ages of 12 and 16 years, but are quite abnor- Bannayan-Riley-Ruvalcaba syndrome with 6 Donnai D. Editorial comment. Pitt-Rogers- mal after the age of 17 years. overgrowth frequently present at birth,9 it is Danks syndrome and Wolf-Hirschhorn syn- Physical measurements do not convey the difficult to think of dup(4p16.3) as an drome. Am J Med Genet 1996;66:101-3. 7 Zollino M, Bova R, Neri G. From Pitt-Rogers- whole picture. Clinically, those with over- overgrowth syndrome at all. Danks syndrome to Wolf-Hirschhorn syn- growth appear to be big people with large Secondly, the function of FGFR3 can be drome. And back? Am J Med Genet 1996;66: body frames, prominent supraorbital ridges, http://jmg.bmj.com/ deduced from the Fgfr3 -/- knockout mouse,'" 113-15. heavy facial features, and big hands and feet. which is overgrown with excessively long 8 Partington MW, Fagan K, Soubjaki V, Turner G. Translocations involving 4pi6.3 in three Some, but not all, of these features are shown femora and elongated vertebrae, resulting in a families: deletion causing the Pitt-Rogers- in the illustrations in our article. long tail. Thus, the normal function of Danks syndrome and duplication resulting in a Lastly, your readers may be interested to FGFR3 is to regulate endochondral ossifica- new overgrowth syndrome. J Med Genet 1997; know that since our paper was published we on 34:719-28. tion by "putting the brakes growth". 9 Cohen MM Jr. A comprehensive and critical have managed to meet III.8 in family 1 again Evidence is accumulating that the known assessment of overgrowth and overgrowth syn- and, on this occasion, were able to make some mutations on FGFRs are of the gain of func- dromes. In: Harris H, Hirschhorn K, eds. measurements and take a blood sample. At on September 25, 2021 by guest. Protected copyright. tion type. For example, Neilson and Friesel" Advances in human genetics. Vol 18. Chap 4. New York: Plenum Press, 1989:181-303. Ad- the age of 33, III.8 is a big woman with a made mutations in mRNA and expressed dendum, pp 373-6. heavy body frame and rather coarse or heavy them in Xenopus that corresponded to known 10 Deng C, Wynshaw-Boris A, Zhou F, Kuo A, facial features of which she is self-conscious human mutations on FGFR1 and FGFR2. Leder P. Fibroblast 3 is and will not allow herselfto be a negative regul4tor of bone growth. CeU 1996; photographed. Analysis of mutant receptor proteins ex- 84:911-21. Her height is 180 cm, head circumference 60 pressed in Xenopus oocytes indicated that all 11 Neilson KM, Fiiesel R. Iigand-independent cm, and hand length is 20.5 cm. She has mild but one had increased tyrosine kinase activity activation of fibrbblast growth factor receptors intellectual handicap and a duplication of compared to their wild type counterparts. by point mutatiqns in the extracellular, trans- 4pl6.3. FGFR3 mutations for achondroplasia and membrane, and cinase domains. J Biol Chem 1996;271 :25049-57. MICHAEL thanatophoric dysplasia have also been 12 Naski MC, Wang Q Xu J, Ornitz DM. Graded PARTINGTON shown to have greatly increased levels of activation offibro+last growth factor receptor 3 GILLIAN TURNER 1' by mutations causing achondroplasia and Hunter Genetics, PO Box 84, Waratah, New South phosphotyrosine.'2 Ligand independent thanatophoric dysplasia. Nat Genet 1996;13: Wales 2298, Australia constitutive signalling produced by these 233-7. mutations results in premature maturation of 13 Webster MK, Donoghue DJ. Constitutive acti- bones of the skeleton and cranium. This type vation of fibroblast growth factor receptor 3 by the transmembrane domain point mutation Robinow syndrome of activation depends on the particular muta- found in achondroplasia. EMBO Y 1996;15: tion and its location on the receptor and 520-7. appears to result from (1) aberrant disulphide 14 Webster MK, Donoghue D. FGFR activation in I would like to comment on the paper by bonded or hydrogen bonded FGFR dimers skeletal disorders: too much of a good thing. Sabry et al' in the September issue of the Trends Genet 1997;13:178-82. Journal ofMedical Genetics. In it, they describe Table 1 Rearranged data three patients they diagnosed as having Rob- This letter was shown to Dr Partington et al, inow syndrome in conjunction with a number Centiles (%) who reply as follows. of unusual abnormalities. Thank you for the opportunity of seeing When looking at the photographs of their No ยข 90 50 S25 the comments of Cohen and Neri on our patients, I doubt whether the diagnosis of recent paper. We were mildly surprised at Height 31 32 29 26 Robinow syndrome is correct, especially in Weight 23 52 35 22 their confidence not only in dismissing the the first patient. She has a number of facial Head circumference 22 41 32 9 possibility of a dosage effect of FGFR3 on characteristics that are indeed seen in Rob- growth, but also in asserting that all the short inow syndrome, such as hypertelorism and