Journal

In Silico Analysis of BRCA1 and BRCA2 Sequence Variants of Unknown Clinical Significance

Petar Ozretić

03.12.2007.

• polymorphisms, mutations & unclassified variants (UVs)

• BIC (The Breast Cancer Information Core Database)– BRCA1 : 1643 different sequence variants– BRCA2 : 1856 --- | | ---

• 1/3 BRCA1 & ½ BRCA2 variants from BIC are UVs• difficult to make population based studies

– low frequency alleles– ethnic specificty

• genetic counseling - problem for risk assessment – noninformative test results

• prophylactic mastectomy and/or oophorectomy?!

BRCA1 sequence variants in 220 healthy Croatian women:

Nucleotide change AA change

Proportion of carriers

Allele frequency

Exon BIC accession No. or reference

Times reported in BIC

Clinically important

Missense Variants (11)

3667A>G K1183R 49.09% 27.50% 11 1099 33 No

4956A>G S1613G 46.82% 24.77% 16 1140 36 No

3232A>G E1038G 46.36% 24.55% 11 1087 37 No

2731C>T P871L 43.18% 22.05% 11 1067 26 No

1186A>G Q356R 11.36% 5.68% 11 2558 82 Unknown

3238G>A S1040N 4.09% 2.05% 11 1089 45 Unknown

5075G>A M1652I 2.73% 1.36% 16 1143 39 Unknown

4654G>T S1512I 1.36% 0.68% 15 1136 53 No

2196G>A D693N 0.91% 0.45% 11 1045 16 No

4158A>G R1347G 0.91% 0.45% 11 1351 154 Unknown

2121C>T L668F 0.45% 0.23% 11 2313 25 Unknown

Synonymous Variants (4)

4427T>C S1436S 42.27% 25.23% 13 1128 35 No

2201C>T S694S 44.55% 24.77% 11 1047 14 No

2430T>C L771L 40.00% 20.23% 11 1055 25 No

2030T>C T637T 0.45% 0.23% 11 - Nil

Intronic Variants (6)

IVS17+65G>A 45.91% 26.59% 17 [12]

IVS7-34C>T 43.18% 22.95% 8 1424 9 No

IVS17-53C>T 0.91% 0.45% 18 10455 2 Unknown

IVS20+60ins12 0.91% 0.45% 20 1292 26 Unknown

IVS6+7G>A 0.45% 0.23% 6 2238 12 Unknown

IVS19+49delTA 0.45% 0.23% 19 - Nil

BRCA2 sequence variants in 115 healthy Croatian women:

Nucleotide change

AA change

Proportion of carriers

Alelle frequency

Exon BIC accession No. or reference

Times reported in BIC

Clinically important

Missense Variants (13)

1342A>C N372H 35.65% 19.13% 10 1668 9 No

1093A>C N289H 8.70% 4.35% 10 1129 13 No

3199A>G N991D 8.70% 4.35% 11 1903 6 No

5972T>C M1915T 5.22% 2.61% 11 1108 7 Unknown

1206C>A S326R 3.48% 1.74% 10 1222 105 No

3743C>T S1172L 2.61% 1.30% 11 1908 43 Unknown

7772C>T T2515I 2.61% 1.30% 15 1547 71 No

353A>G Y42C 0.87% 0.43% 3 1004 141 Unknown

3546C>A S1106R 0.87% 0.43% 11 - Nil

4486G>T D1420Y 0.87% 0.43% 11 1274 191 No

5018C>A S1596Y 0.87% 0.43% 11 - Nil

6328C>T R2034C 0.87% 0.43% 11 1324 97 Unknown

7163A>T D2312V 0.87% 0.43% 12 5423 5 Unknown

Synonymous Variants (14)

3624A>G K1132K 46.09% 23.91% 11 1661 8 No

7470A>G S2414S 43.48% 23.04% 14 1125 10 No

4035T>C V1269V 33.04% 19.57% 11 1662 3 No

1593A>G S455S 8.70% 4.35% 10 1106 7 No

2457T>C H743H 7.83% 3.91% 11 13616 7 No

6492T>C T2088T 1.74% 0.87% 11 - Nil

426A>G Q66Q 0.87% 0.43% 3 [13]

687A>C P153P 0.87% 0.43% 5 - Nil

2166C>T S646S 0.87% 0.43% 11 2907 2 No

3744G>A S1172S 0.87% 0.43% 11 3501 4 No

4296G>A L1356L 0.87% 0.43% 11 3525 3 Unknown

5427C>T S1733S 0.87% 0.43% 11 1941 2 No

9315G>A A3029A 0.87% 0.43% 23 - Nil

10092A>G T3288T 0.87% 0.43% 27 - Nil

Intronic and Non Coding Variants (5)

IVS16-14C>T 73.04% 46.52% 17 1126 15 No

IVS11+80del4 61.74% 38.26% 11 2022 1 Unknown

203G>A 55.65% 35.65% 2 1666 12 No

IVS4+67A>C 8.70% 4.35% 4 [14]

IVS24-16T>C 0.87% 0.43% 25 1123 6 Unknown

Truncating Mutations (3)

5873C>A S1882X 0.87% 0.43% 11 1036 26 Yes

10204A>T K3326X 0.87% 0.43% 27 1179 293 No

10323delCins11 3369X 0.87% 0.43% 27 1788 11 No

Deletions (1)

4640delGAA R1471del 0.87% 0.43% 11 - Nil

9 novel BRCA1 & BRCA2 sequence variants by mutation type:

Intronic Variants:BRCA1 IVS19+49delTA

Synonymous Variants:BRCA1 2030T>C (T637T)BRCA2 6492T>C (T2088T)

687A>C (P153P)9315G>A (A3029A)10092A>G (T3288T)

Missense Variants:BRCA2 3546C>A (S1106R)

5018C>A (S1597Y)

Deletions:BRCA2 4640delGAA (R1471del)

• the type of mutation (missense, nonsense, synonymous, frameshift, deletion, insertion...)

• the location of the mutation within the gene, e.g. functional domains, splice sites

• the presence or absence of the variant in a control population

• co-segregation, or lack of co-segregation, of the variant and disease within families

• co-occurrence with a deleterious mutation• the type of amino acid change• the conservation of the amino acid across species• biochemical/functional analysis

Informations needed to assess clinical significance of a variant:

Intronic Splicing

Splice site consensus sequences (from Blencowe, 2000)

• human - 8.8 introns per gene on average, mean intron length 4400 nt, mean exon length 165 nt (5’ donor, 3’ acceptor)

Splicing in outline. Cleavage of the 5′ splice site is promoted by the hydroxyl (OH) attached to the 2′-carbon of an adenosine nucleotide within the intron sequence. This results in the lariat structure and is followed by the 3′-OH group of the upstream exon inducing cleavage of the 3′ splice site. This enables the two exons to be ligated, with the released intron being debranched and degraded. (from Brown, T.A.: Genomes. New York and London: Garland Science; c2002)

Models of SR protein action in exonic-splicing-enhancer-dependent splicing (from Cartegni et al., 2003)

Exonic Splicing Enhancers• ESEs (exonic splicing enhancers) - motifs located in exon regions

– 4-18 nucleotides, degenerative, overlaping, purine-rich– required for regulation of constitutive and alternative splicing (35%-60% of

human genes undergo alternative splicing)– binds splicing regulatory proteins:

• SR-family proteins (serine/arginine-rich)– N-terminal RRMs – RNA recognition motifs – RNA-proteins

interactions– C-terminal RS domain – protein-protein interactions

· ESSs (exonic splicing silencers) - heterogeneous nuclear

ribonucleoproteins (hnRNP A1)

Splicing mutations• up to 50% of all point mutations responsible for genetic diseases

cause aberrant splicing• such mutations can disrupt splicing by directly inactivating or

creating a splice site, by activating a cryptic splice site or by interfering with splicing regulatory elements

• ECEs can be disrupted by single nonsense, missense and translationaly silent point mutations

Two aberrant forms of splicing. (A) In exon skipping the aberrant splicing results in an exon being lost from the mRNA. (B) When a cryptic splice site is selected, part of an exon might be lost from the mRNA, as shown here, or if the cryptic site lies within an intron then a segment of that intron will be retained in the mRNA. (from Brown, T.A.: Genomes. New York and London: Garland Science; c2002)

BDGP: Splice Site Prediction by Neural Network (NNSPLICE)

http://www.fruitfly.org/seq_tools/splice.html

ACTUAL SPLICE SITES DON’T ALWAYS GET THE HIGHEST SCORES!!!

SpliceSiteFinderhttp://violin.genet.sickkids.on.ca/~ali/splicesitefinder.html

Score probability for the recognition of actual intron donor and acceptor sites

NNSPLICE SpliceSiteFinder donor acceptor donor

acceptorBRCA1 IVS19 w.t. 1.0 0.73 100.0 91.3BRCA1 IVS19+49delTA 1.0 0.73 100.0 91.3

Slovenian founder mutation BRCA2 IVS16-2A>G

SpliceSiteFinder

start donor sequence score BRCA2 IVS16 w.t. 10 AG|GTACTC 68.3BRCA2 IVS16-2A>G 10 AG|GTACTC 68.3

start acceptor sequence score start dist. branch seq. score score of D.+B.BRCA2 IVS16 w.t. 4591 CTTTATTTGTTCAG|G 96.1 4576 -26 ATGAT 76.0 172.1BRCA2 IVS16-2A>G 4611 GTGTGACACTCCAG|G 81.7 4579 -43 ATAAT 75.1 156.8

BRCA2 IVS16-2A>G

ESEfinder 3.0 http://rulai.cshl.edu/cgi-bin/tools/ESE3/esefinder.cgi

ESE analysis of novel synonymous sequence variants:

• BRCA1 – 2030T>C (T637T)> + SC35

• BRCA2– 687A>C (P153P) > - SRp40, + SC35, +

SF2/ASF– 6492T>C (T2088T) > - SF2/ASF, + SC35

Poly-Phenhttp://genetics.bwh.harvard.edu/pph/

• PolyPhen (=Polymorphism Phenotyping) predicts possible impact of an amino acid substitution on the structure and function of a human protein, using straightforward physical and comparative considerations.

• position-specific independent count (PSIC) score for each variant – difference ≥1.1 considered to be damaging

• higher PSIC score difference – higher functional impact a particular aminoacid substitution has

• BRCA2 newly found missense variants:PSIC score difference prediction

S1106R (3546C>A) 1.773 possibly damagingS1597Y (5018C>A) 1.815 possibly damaging

Lee PH, Shatkay H.F-SNP: computationally predicted functional SNPs for disease association

studies.Nucleic Acids Res. 36(Database issue):D820-4.