This study was approved by the Ethics Committee of the Division of Obstetrics and Gynecology, National Taiwan University Hospital. Eighty-four unrelated patients with confirmed clinical diagnoses of TSC and their family members were tested for mutations in TSC1 and TSC2 genes.

The general clinical features of TSC patients were determined by clinicians in accordance with the TSC diagnosis criteria set forth by the Tuberous Sclerosis Consensus Conference 3 . All patients' symptoms were investigated by a person blind to mutational status. High-resolution brain magnetic-resonance imaging (MRI) or computed tomography (CT) was performed on most patients.

The extent of facial angiofibroma or forehead plaques, non-traumatic ungal or periungal fibromas, hypomelanotic macules, shagreen patches, multiple retinal nodular hamartomas, cortical tubers, subependymal nodules, subependymal giant cell astrocytomas, cardiac rhabdomyomas, lymphangiomyomatoses, renal angiomyolipomas and confetti-like lesions were all assessed. Moreover, most patients' medical histories of mental development were assessed by a certified psychologist.

After genetic counseling and obtaining informed consent, 5–10 mL of peripheral blood were collected from the participants. Genomic DNA was isolated from peripheral whole blood using the Puregene DNA Isolation Kit (Gentra Systems, Inc., Minneapolis, MN, USA).

PCR primers and running conditions for each exon were available from previous studies 11 12 13 . The PCR reaction was run on each exon with a total sample volume of 25 μL containing 100 ng of genomic DNA, 0.12 μM of each respective primer, 100 μM dNTPs, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 2 mM MgCl₂, and 0.5 units of AmpliTaq Gold enzyme (PE Applied Biosystems, Foster City, CA, USA). Amplification was performed in a multiblock system thermocycler (ThermoHybaid, Ashford, UK). The PCR amplification started with a denaturing step at 95°C for 5 minutes, followed by 35 cycles of denaturing at 94°C for 30 seconds, annealing at melting temperature (Tm) for 30 seconds, extension at 72°C for 45 seconds, and ends with a final extension step at 72°C for 10 minutes.

The screening of mutations was performed using the Transgenomic Wave Nucleic Acid Fragment Analysis System (Transgenomic Inc, San Jose, CA) with a C₁₈ reversed-phase column containing 2-μm nonporous poly (styrene/divinylbenzene) particles (DNASep Column, Transgenomic Inc). PCR products were analyzed using linear acetonitrile gradients and triethylammonium acetate acting as mobile phases with the provision of buffer A (0.1 M TEAA) and buffer B (0.1 M TEAA with 25% acetonitrile) (WAVE Optimized, Transgenomic Inc). Heteroduplex analyses were performed according to the manufacturer's protocol and of previous studies 14 15 .

The χ² and Fisher exact tests were used to examine the differences in clinical manifestations, phenotypes, and mutation distributions in independent Taiwanese probands between patients with TSC1 and TSC2 genes.

PCR products were purified by solid-phase extraction and bidirectionally sequenced using Applied Biosystems' Taq DyeDeoxy terminator cycle sequencing kit (Applied Biosystems). Sequencing reactions were separated on a PE Biosystems 373A/3100 sequencer.

In the current study, we performed mutational analysis on the coding exons and the exon/intron junctions of both TSC1 and TSC2 in a total of 84 individuals with TSC and their family members. The determination of mutation vs. polymorphism was done by: 1) checking the mutation tables at the Chromium site (http://chromium.liacs.nl/); 2) comparison of findings to those of 100 healthy Taiwanese controls; and 3) checking the families similarly.

Nine mutations were identified in the TSC1 gene while 55 were identified in the TSC2 gene. Mutations in the TSC1 gene included five nonsense mutations with early termination codons and four insertions/deletions which caused frameshifts and resulted in premature truncation of the protein. Three of these mutations were novel, while six were previously reported (Table 1).

The 55 mutations in the TSC2 gene included 12 missense, 15 nonsense, 21 frameshifts due to insertions and deletions and 7 putative splice-site mutations. Twenty-seven of these mutations were previously reported while 28 were novel (Table 2). Of the familial TSC2 missense mutations, A1141T and R1793Q may be rare polymorphic variants co-segregating with TSC. There was no direct evidence that these familial TSC2 missense mutational changes were pathogenic.

For both genes, sequence variants that were possible mutations were tested in all other family members, including the parents and both the affected and the unaffected family members. In total, 31 of the 64 mutations (48%) had not been reported elsewhere. Moreover, no mutational hotspots were identified in either gene, with only four different mutations being found twice in TSC2.

Compared with those of European and American counterparts 9 10 16 , the distribution of the TSC1 and TSC2 mutations among Taiwanese population is similar. Therefore, the spectrum of mutations seen among the Taiwanese is no different in comparison to those already reported thus far for these two genes, based on the genetic analyses of European and American TSC patients using the Fisher exact test (P = 0.85, 0.46, and 0.14, respectively).

In order to identify whether the observed changes were mutations or polymorphisms, samples from 100 normal individuals serving as controls were analyzed. Changes that were not found in more than 200 control alleles were considered pathogenic. Therefore, unique or less frequent changes such as missense and splicing site mutations (Table 2) were considered likely pathogenic mutations. The nonpathogenic TSC1 and TSC2 mutations identified in the Taiwanese TSC patients are described in Table 3. We identified nine nonpathogenic polymorphisms in the TSC1 gene and 12 in the TSC2 gene. The nonpathogenic sequence variants were identified in both the TSC patients and the normal controls. Fourteen of these polymorphisms had not been reported previously (4 at the TSC1 locus and 10 at the TSC2 locus) that included one missense variant within the TSC1 coding region.

Mutations were identified and located in exons of both TSC1 and TSC2 genes (see Figure 1 and 2). Of the 64 mutations found, nine and 55 were associated with TSC1 (14%) and TSC2 (86%), respectively, as shown in Table 4. Of the 10 familial cases, 2 (20%) and 8 (80%) were TSC1and TSC2 mutations, respectively. Among the 54 sporadic cases, 7 TSC1 (13%) and 47 TSC2 (87%) mutations were found. Accordingly, there was no significant difference between sporadic and familial TSC cases with respect to the frequency of TSC1 vs TSC2 mutation (P = 0.62).

The clinical characteristics associated with each mutation in the proband are shown in Tables 5 (eight TSC1 mutations) and Table 6 (43 TSC2 mutations). Most patients with TSC1 and TSC2 mutations had seizures, brain lesions (subependymal nodules and/or cortical tubers detected by MRI), and dermal manifestations. Our criteria for intellectual disability included any degree of mental retardation and learning disorder. The incidence of intellectual disability appeared lower in patients with TSC1 mutations (3/8 = 38%) compared to that of patients with TSC2 mutations (27/43 = 63%). However, this difference was not statistically significant (P = 0.25), but this would be expected because of such small sample sizes. Similarly, the incidence of mental retardation in patients with TSC1 mutations (1/8 = 13%) appeared to be less than that of patients with TSC2 mutations (17/43 = 40%), but this difference was not statistically significant (P = 0.23). Similarly, the frequencies of renal findings, cortical tubers, subependymal giant cell astrocytomas, liver tumors, cardiac tumors, or skin manifestations, including hypomelanotic macules, facial angiofibromas, shagreen patches, and ungual fibromas did not significantly differ between the patients with TSC1 and TSC2 mutations. However, all of these comparisons are under-powered due to the relatively small number of patients with TSC1 mutations that were studied. For nearly all of the clinical features studied, the frequencies were less for those bearing TSC1 mutations than for those bearing TSC2 mutations. This is consistent with findings from other large studies, showing that TSC1 disease is less severe than TSC2 disease 9 10 16 .