Eukaryotic pre-mRNA splicing is regulated by consensus sequences at the intron boundaries and branch site. Recently, Sirand-Pugnet et al. reported the importance of an additional intronic sequence, an (A/U)GGG repeat in chicken [beta]-tropomyosin that is a binding site for a protein required for spliceosome assembly. Interestingly, we have detected mutations in IVS3 of the human growth hormone (GH) gene that affect a putative, homologous consensus sequence and which also perturb splicing. In a series of dominant-negative GH mutations that cause exon skipping, we found two mutations that do not occur within the 5' and 3' splice sites,or branch consensus sites. The first mutation is a G -> A transition of the 28th base (+28G -> A) of and the second deletes 18 bp (del+28-45) of IVS3 of the human GH gene. These mutations segregated with autosomal dominant GH deficiency in both kindreds and no other allelic GH gene changes were detected. RT-PCR amplification of transcripts from expression vectors containing the +28G -> A or del+28-45 alleles yielded products showing a >10-fold preferred use of alternative splicing, similar to findings previously reported for IVS3 donor site mutations. Both mutations are located 28 bp downstream from the 5' splice site and examination of the sequences perturbed revealed an intronic XGGG repeat similar to the repeat found to regulate mRNA splicing in chicken [beta]-tropomyosin. Interestingly, the XGGG repeats involved in our mutations exhibit homologous spacing to those in a so-called `winner' RNA sequence. Binding of A1 heterogeneous nuclear ribonucleoprotein (hnRNP) by `winner' sequences in pre-mRNA transcripts is thought to play an important role in pre-mRNA packaging and transport as well as 5' splice site selection in pre-mRNAs that contain multiple 5' splice sites. Our findings suggest that (i) XGGG repeats may regulate alternative splicing in the human GH gene and (ii) mutations of these repeats cause GH deficiency by perturbing alternative splicing. Mutations of homologous intron sequences may underlie other human diseases.
Pre-mRNA processing to remove intervening sequences (introns) utilizes conserved sequences at the 5' and 3' splice sites, and branch consensus site of the introns (1 ). However, these conserved sequences alone are not sufficient to explain the splicing process because the presence of cryptic splice sites, which are either not utilized or are preferentially involved in tissue specific alternative splicing, suggests that other cis-acting elements must exist (2 ,3 ).
Growth hormone (GH) gene defects that cause some cases of autosomal dominant isolated GH deficiency (IGHD II) have been previously reported (4 -6 ). These mutations have all occurred within the 5' splice site of IVS3 resulting in aberrant splicing and deletion of exon 3. This exon 3 skipping deletes 120 nucleotides (nt) which correspond to amino acids 32-71 and results in a truncated GH protein that corresponds to the 17.5 kDa isoform of GH.
We report here two additional IGHD II IVS3 mutations, from two non-related kindreds, which also perturb splicing and cause exon 3 skipping, but do not occur within the IVS3 branch consensus, 5' or 3' splice sites. These mutations derange sequences homologous to XGGG repeats that regulate alternative mRNA splicing in the chicken [beta]-tropomyosin gene and sequences that form G-quartets which bind the A1 heterogeneous nuclear ribonucleoprotein (hnRNP) (16 -17 ).
To determine the segregation patterns of GH alleles, DNA from the members of two IGHD II families were genotyped for a dinucleotide repeat polymorphism within the GH gene cluster (8 ). The results were consistent with transmission of an abnormal GH1 allele from the affected parent to each affected child and co-segregation of this GH1 allele with the IGHD II phenotype (data not shown). Furthermore, the mutant alleles in the Thai family were in coupling with the same polymorphic dinucleotide repeat allele in all individuals genotyped.
The sequences of the GH1 alleles of an affected child from Family 1 were determined by direct sequencing of PCR products while those in the affected child from Family 2 were determined by sequencing the cloned PCR products (Fig. 1 ). Note that patient 1 (arrow) from Family 1 as well as both affected parents and sibs are heterozygous for a G -> A transition of the 28th base of IVS3 (IVS3 +28G -> A). Patient 2 (arrow) from Family 2, as well as his affected father and son, is heterozygous for an 18 bp deletion of bases 28-45 of IVS3 (IVS3 del+28-45).
Sequence analysis of RT-PCR products from mRNA transcripts generated in transient expression studies of these two mutations in IVS3 of the human GH gene showed that both mutations caused aberrant splicing, with skipping of exon 3 as reported previously with IVS3 donor splice site mutations. Since neither IVS3 +28G -> A nor del+28-45 are located within the 5' and 3' splice, or branch consensus sites, we concluded they must perturb a previously unidentified regulatory region (see Fig. 4 ).
We studied DNAs from two non-related kindreds. The patients were diagnosed as having familial IGHD II by the following criteria: multiple relatives were affected in an apparent autosomal dominant mode and probands had severe growth retardation, low basal and post stimulation GH levels, normal thyroid hormone levels, and positive responses to hGH replacement. Family 2 was originally described by Poskitt and Rayner in 1974 (7 ).
DNAs were isolated from peripheral blood samples from all family members and controls. The samples were genotyped for a dinucleotide repeat polymorphism within the GH gene cluster as described by Polymeropoulos et al. with some modifications (8 ). The forward and reverse primers corresponding to nt 25718-25736 (5'-ACTGCACTCCAGCCTCGGA-3') and the complement of 26049-26031 (5'-ATGGCACTCCAGCCTGGGC-3') of the GH gene cluster respectively (9 ). The PCR reaction mixture was denatured for 5 min at 94oC, cycled 27 times (94oC, 30 s; 59oC, 30 s; 72oC, 30 s), followed by a 10 min extension at 72oC. The PCR products were separated on a 6% denaturing polyacrylamide gel and visualized by autoradiography.
The GH1 genes of an affected child from each family were PCR amplified using forward and reverse primers corresponding to nt 4556-4575 (5'-CCAGCAATGCTCAGGGAAAG-3') and the complement of 7255-7226 (5'-TGTCCCACCGGTTGGGCATGGCAGGTAGCC-3') of the GH gene cluster, respectively (10 ). The PCR reaction mixture was denatured for 6 min at 94oC, cycled 32 times (94oC, 1 min; 65oC, 45 s; 72oC, 3 min), followed by a 10 min extension at 72oC. The resulting GH1 PCR products (2700 bp) were cleaned by filtration with a Centricon 100 microconcentrator (Amicon Corp., Danvers, MA) and used as templates for a second PCR using forward and reverse primers corresponding to nt 4599-4615 (TTCTCTCTAGTGGTCAGTGT) and the complement of 7206-7187 (CTCCTGCTGGTATACTTATT) of the gene cluster, respectively (2 ). The PCR reaction mixture was denatured for 6 min at 94oC, cycled 32 times (94oC, 1 min; 54oC, 45 s; 72oC, 3 min), followed by a 10 min extension at 72oC. The resulting GH1 PCR products (2608 bp) were cleaned by filtration with a Centricon 100 microconcentrator (Amicon Corp., Danvers, MA) and then either sequenced directly (patient 1) or cloned into a pUC18 vector using the SureClone Ligation Kit (Pharmacia Biotech) (patient 2) and then sequenced by the dideoxy method (11 ) using the ThermoSequenase cycle sequencing system (Amersham Life Science, Cleveland, OH) (see Fig. 1 ).
The 2608 bp GH1 PCR amplification products were used as templates for an internal PCR containing the mutated region under standard PCR conditions with forward and reverse primers corresponding to nt 5810-5833 (GTAACAATGGGAGCTGGTCTCCAG) and the complement of 6068-6046 (GAAGGACGGGCATTGGCTGTGCT) of the gene cluster, respectively (9 ). PCR reaction mixtures were denatured for 6 min at 94oC, cycled 32 times (94oC, 45 s; 67oC, 45 s; 72oC, 1 min), followed by a 10 min extension at 72oC. For Family 1 aliquots were then digested with FokI and separated on a 5% Metaphor agarose gel and visualized by ethidium bromide staining. In Family 2, aliquots were analyzed directly by visualization in an ethidium bromide stained 3% Metaphor agarose gel.
Human GH expression vectors containing the IVS3 +28G -> A and the del+28-45 mutations were constructed using the Muta-Gene T7 M13 In Vitro Mutagenesis Kit (BioRad Laboratories, Hercules, CA). The entire GH1 gene, with introns, was excised from the GH expression vector pXGH5 (Nichols Diagnostics Institute, San Juan Capistrano, CA) by digestion with EcoRI (New England Biolabs, Inc., Beverly, MA) and subcloned into the M13mp19 vector which had also been digested with EcoRI. Site directed mutagenesis (12 ,13 ) was carried out using an oligonucleotide containing the desired mutation (5'-CTCCCCCATCTCCGCCTGGGG-3' for Family 1 and 5'-CCGGGGGCTCTGACCCGCCTGGGGAGAA-3' for Family 2) according to the manufacturer's specifications.After confirming that the desired mutation was present, a GH1 gene fragment containing the mutation was excised from the M13mp19/hGH clone by digestion with BglII and AvrII and ligated to the corresponding BglII/AvrII sites of the pXGH5 expression vector.
DNA transfections were carried out in HeLa cells using the Gibco calcium phosphate transfection system (Gibco-Bethesda Research Laboratories, Gaithersburg, MD). Cells were grown in 110 mm2 dishes to 50% confluency and then transfected with 10 [mu]g GH plasmid DNA per dish in accordance with the manufacturer's instructions. The cells were harvested 48 h after transfection, flash frozen and stored at -70oC (4 ).
Messenger RNA was isolated from transfected cells and used as template for cDNA synthesis using RT-PCR (4 ). First strand cDNA synthesis was done using the Superscript pre-amplification system (Gibco-Bethesda Research Laboratories, Gaithersburg, MD) and a reverse primer corresponding to the complement of nt 6775-6753 (5'-ACAAGGCTGGTGGGCACTGGAGT-3') of the GH gene cluster (9 ). First strand cDNA products were then used as templates for PCR amplification under standard conditions. The forward and reverse primers corresponded to nt 5543-5565 (5'-GGCTTCAAGAGGGCAGTGCCTTC-3') and the complement of 6775-6753. The PCR reaction mixture was denatured for 6 min at 94oC, cycled 55 times (94oC, 1 min; 69oC, 30 s; 72oC, 1 min), followed by a 10 min extension at 72oC. The resulting GH1 cDNA products were cleaned by filtration with a Microcon-100 microconcentrator, separated on an 8% polyacrylamide gel and the fragments sequenced (Fig. 2 ) as described above.
This work was supported by NIH grant DK35592.
GH, Growth hormone; RT-PCR, reverse transcriptase polymerase chain reaction; IVS, intervening sequence; IGHD II, isolated growth hormone deficiency type II; PCR, polymerase chain reaction.
Human Molecular Genetics
Pages
Introduction
Results
Discussion
Materials And Methods
Subjects
Segregation
PCR/restriction endonuclease detection of mutations
Site directed mutagenesis and in vitro expression studies
Synthesis and amplification of GH complementary DNA (cDNA)
ACKNOWLEDGEMENT
Abbreviations
References
REFERENCES
This page is maintained by OUP admin. Last updated Mon May 12 18:10:04 BST 1997. Part of the OUP Journals World Wide Web service. Copyright Oxford University Press, 1996