Wrapping up DiGeorge syndrome in a T-box?

Abstract

(DGS) is charac-terized by congenital heart diseasewith parathyroid and thymic hy-poplasia. Patients with related velocar-diofacial syndrome (VCFS) have similarconotruncal defects and abnormal facies.Most patients with DGS/VCFS havechromosome 22q deletions that com-monly include 24 contiguous genes (1).However, human genetic studies havefailed to identify specific gene mutationsto account for individual components ofthis multi-organ phenotype. Therefore,investigators have now used “knockout”mice to study genes deleted in DGS/VCFS. Scientists have focused on neuralcrest-related genes since the pharyngealand aortic arches which develop abnor-mally in DGS/VCFS are neuralcrest-derived.Jerome and Papaioannou (2) studiedthe mouse gene encoding the Tbx1transcription factor; this gene maps tothe murine equivalent of the DGS/VCFS critical region. Because Tbx1 isstrongly expressed in the third, fourthand sixth arches during embryogene-sis, they genetically engineered micehaploinsufficient ( Tbx1 / ) or null(Tbx1 / )inTbx1. Tbx1 / micedied in utero with abnormal facies andthymus and parathyroid aplasia alongwith heart failure and malformed car-diac outflow tracts and aortic arches.Although Tbx1 / mice were viableand had no noncardiovascular abnor-malities, many had DGS/VCFS-likeaortic arch abnormalities. Aortic archabnormalities were also observed byLindsay et al. (3) and Merscher et al.(4) who independently created micehemizygous for large genomic seg-ments including Tbx1 and other genes.Both groups showed that specific re-placement of only the Tbx1 gene cor-rected aortic defects. Tbx1 / micecreated by these groups also exhibitedabnormal aortic arches.Taken together, these studies showthat Tbx1 haploinsufficiency is sufficientto produce DGS/VCFS-like cardiovas-cular phenotypes in mice. Investigatorshypothesize that human TBX1 haploin-sufficiency via chromosome 22q11 dele-tion plays a major role in human DGS/VCFS conotruncal defects. Mutations inthe TBX5 and TBX3 transcription fac-tors cause Holt-Oram and ulnar-mam-mary syndromes, respectively, and thesenew findings add to the body evidencethat human congenital disease can becaused by alterations in gene dose ofT-box transcription factors (5).Several issues remain to be addressed.First, cardiovascular gene expressionpatterns in mice and humans are notalways identical, and human TBX1 ex-pression patterns have yet to be eluci-dated. Second, what is the contributionof TBX1 haploinsufficiency to noncar-diovascular disease? Only mice that arecompletely lacking Tbx1, unlike haplo-insufficient DGS/VCFS patients, exhibitnoncardiovascular phenotypes. Thesediscrepancies may reflect different sensi-tivities of human and mouse embryos toTBX1 dosage or requirements for se-quences critical to regulation of Tbx1expression that are physically remotefrom Tbx1’s coding sequence. Knockoutmouse models of several other humanchromosome 22q genes (e.g., Ufd1l,HIRA, Gsc1) that are expressed in DGS/VCFS-affected tissues have failed todemonstrate phenotypes. However, theirhaploinsufficiency in humans may syn-ergize with TBX1 haploinsufficiency tocause DGS/VCFS. Homozygous knock-out mice for the murine homolog of thehuman chromosome 22q11 CRKL genehave cardiovascular, skeletal, thymus,and parathyroid phenotypes similar toDGS/VCFS patients (6). Thus, humandisease may reflect interactions betweenhaploinsufficiencies of multiple genes.Lastly, to definitively establish arole for TBX1 in human DGS/VCFScardiovascular disease, it is critical todemonstrate that mutation of this genealone causes a human phenotype.Lindsay et al. (3) note failure to detectTBX1 missense mutations in 100 DGSpatients without chromosome 22q de-letions. However, disease in these pa-tients may relate to mutations at otherloci, e.g. a chromosome 10p DGS lo-cus. Particularly important will be on-going TBX1 mutational analysis in pa-tients with chromosome 22q deletionsthat do not encompass TBX1; thesepatients may have subtle TBX1 se-quence abnormalities to account forcardiovascular phenotypes withoutTBX1 deletion. There is a high preva-lence of chromosome 22q deletions inpatients without DGS/VCFS but withother isolated conotruncal defects (7).Thus, patients with isolated conotrun-cal defects but without chromosome22q deletions will also be an importanttarget for TBX1 mutational analyses.In summary, the recent mouse studiesof Tbx1 haploinsufficiency have estab-lished Tbx1’s contribution to conotrun-cal development and provide a new scaf-folding for analysis of specific genecontributions to the complex abnormali-ties that comprise DGS/VCFS.