Third Base Of Codon Research Articles

Identification and comparison of point mutations associated in classic and variant infectious bursal disease viruses

Restriction enzymes (RE) were used to identify point mutations in the nucleotide sequences of the infectious bursal disease virus (IBDV) strains Del-E, Del-A, STC, IN, Bursine 2, and Bio-Burs. The point mutations at amino acid sites 222, 254 and 323 were identified using REs BstNI, StyI and MboI, respectively. The nucleotide sequences of STC, IN, Bursine 2 and Bio-Burs were determined at each of these sites. Nucleotide sequence analysis using this data and previously reported data for Del-E and Del-A was used to confirm the point mutation and predict the resulting amino acids. Although there were some exceptions, the BstNI and StyI enzymes detected point mutations in the first base of the 222 and 254 codons, respectively and could be used to predict an amino acid change in the viruses. MboI was detecting a mutation in the third base of codon 323 and could not reliably predict an amino acid change at that position. The results indicated that amino acids 222 and 254 were consistantly mutated in the variant viruses examined and that amino acid 323 was not. Furthermore, point mutations resulting in amino acid changes at position 222 suggested that several groups of viruses may be defined by this site alone; proline for classic strains like STC, threonine for Del-E and GLS, serine for IN, Bursine 2, and Bio-Burs and glutamine for Del-A.

N-RASMutations in Radiation-Induced Murine Leukemic Cells

ABSTRACT:N-rasmutations were examined in DNA samples extracted from the spleen of CBA/Ca mice that developed myeloid leukemia (ML) following exposure to radiations of different qualities. A total of 17 ML cases, i.e. 5 cases of neutron-induced and 12 cases of photon-(3 γ-ray and 9 x-ray) induced ML were included in the study along with 12 DNA samples from the bone marrow cells of control mice. Polymerase chain reaction-single strand conformational polymorphisms (PCR-SSCP) and the direct sequencing of PCR products were used to analyze three regions of theN-rasgene: (i) a 128 base-pair (bp) long portion of exon I (codons 2-37); (ii) a 103 bp long portion of exon II (codons 48-82); and (iii) a 107 bp long portion of exon III (codons 118-150). PCR-SSCP mobility shifts indicated mutations within only exon II of theN-rasgene. Such mutations were more prevalent in samples from mice exposed to fast neutrons. The exact type and location of these mutations were then determined by direct DNA sequencing. Silent point mutations, i.e. base transitions at the third base of codons 57 (GAC_→GAT_), 62 (CAA_→CAC_), or 70 (CAG_→CAA_) were present only in mice that developed ML after exposure to fast neutrons. A base transversion at the third base of codon 61 (CAA_→CAC_) was also observed in some ML cases. DNA sequencing demonstrated that ML samples contained normal as well as mutated DNA sequences. The higher frequency ofN-rasmutations in neutron-induced ML suggested that fast neutrons are more effective in inducing genomic instability at theN-rasregion of the genome. More importantly,N-rasmutations are not the initiating event in radiation leukemogenesis. This conclusion was supported by the finding thatN-rasmutations were detected only in mice with an overt leukemic phenotype but not in mice with minimal tissue infiltration of leukemic cells, suggesting that the disease may be present prior to the presence ofN-rasmutations. Alternatively,N-rasmay be present in these mice but a large number of normal spleen cells in these mice interferes with the detection of mutation in a small population of leukemic cells.

A Frame Shift Mutation in the Fibrinogen Aα Chain Gene in a Kindred With Renal Amyloidosis

A new American kindred with amyloidosis was found by single-strand conformation polymorphism analysis to have a mutation in the fibrinogen A alpha chain gene. Affected members in this kindred have autosomal dominant amyloid nephropathy. DNA sequencing showed a single nucleotide deletion at the third base of codon 524 of the fibrinogen A alpha chain genes (4904delG) that resulted in a frame shift and premature termination of the protein at codon 548. Antiserum was produced to a portion of the abnormal peptide predicted by the DNA sequence and amyloid deposits were immuno-histologically proven to contain this abnormal peptide. Two of the propositus' 4 children were positive for the mutant fibrinogen A alpha chain gene by restriction fragment length polymorphism analysis based on polymerase chain reaction. These two mutant gene carriers now in the second decade of life show no clinical symptoms of amyloidosis as yet but have lower plasma fibrinogen concentrations when compared with their normal siblings. This the first description of a kindred with renal amyloidosis and low plasma fibrinogen and also the first report of amyloidosis caused by a frame shift mutation.

HLA-G gene polymorphism in a Japanese population

Polymorphism of the HLA-G gene in a Japanese population was investigated employing polymerase chain reaction (PCR)-single-strand conformation polymorphism (SSCP) analysis, PCR sequence-specific oligonucleotide (SSO) analysis, and DNA direct sequencing. Nucleotide sequence variations in exons 2, 3, and 4 of the HLA-G gene in 54 healthy Japanese individuals were examined. In addition, seven Japanese samples carrying common HLA haplotypes were analyzed. In total, nine single-base substitutions compared with the sequence of G*01011 were identified: one in intron 1 (nucleotide position 970), one in exon 2 (the third base of codon 57: G --> A), three in intron 2 (1264, 1276, and 1292), three in exon 3 (the third base of codon 93: C --> T, the third base of codon 107: A --> T, and the first base of codon 110: C --> A), and one in intron 3 (2334). The substitution at codon 110 was non-synonymous and led to an amino acid substitution from leucine to isoleucine. The other three nucleotide substitutions in exons were synonymous. Through analysis of combinations of the exon 2, 3, and 4 nucleotide sequences we identified four alleles, which we provisionally designated GJ1, GJ2, GJ3, and GJ4. The allele frequencies were estimated to be 0.33, 0.16, 0.45, and 0.06, respectively. Nucleotide sequences of GJ1, GJ2, and GJ4 were identical to G*01011, the clone 7.0E, and G*01013, respectively. GJ3 was a newly observed allele and was officially designated G*0104 by the WHO Nomenclature Committee in January 1996. Strong positive associations were observed between HLA-G alleles and HLA-A, -B, or -DRB1 alleles.

A human pancreatic islet inwardly rectifying potassium channel: cDNA cloning, determination of the genomic structure and genetic variations in Japanese NIDDM patients.

Ligand gated potassium channels, such as the ATP-regulated potassium channel, play crucial roles in coupling of stimuli to insulin secretion in pancreatic beta cells. Mutations in the genes might lead to the insulin secretory defects observed in patients with non-insulin-dependent diabetes mellitus (NIDDM). We isolated a cDNA encoding a putative subunit of a ligand gated potassium channel from a human islet cDNA library. The channel, which we designated hiGIRK2, appeared to be an alternative spliced variant and a human homologue of recently reported mbGIRK2, KATP-2/BIR1. Transcripts were detected in human brain and pancreas, but not in other tissues including cardiac muscle. The sizes of transcripts in the pancreas differed from those in the brain, suggesting tissue-specific alternative splicing and possible isoforms. We then isolated human genomic clones, determined the complete genomic structure and localized the gene to chromosome 21 (21q22). The gene was comprised of four exons and the protein was encoded by three exons. The entire coding region of the hiGIRK2 gene was scanned by polymerase chain reaction-single strand conformation polymorphism analysis in 80 Japanese NIDDM patients. We found five nucleotide substitutions; three were silent mutations of the third base of codons, one in the first intron, 9 bases upstream of exon 2, and one in the 3'-untranslated region. We conclude that mutations in the gene encoding hiGIRK2, a (subunit of) ligand gated potassium channel, is not a major determinant of the susceptibility to NIDDM in Japanese.

Carcinogenicity, DNA adduct formation and K-ras activation by 7H-dibenzo[c,g]carbazole in strain A/J mouse lung.

N-Heterocyclic polynuclear aromatic hydrocarbons (NHA) are environmental pollutants formed during the combustion of organic materials. 7-H-Dibenzo[c,g]carbazole (DBC) is a potent carcinogen in lung, liver and skin. We undertook these studies to determine whether tissue specificity for DBC lung carcinogenicity in the strain A/J mouse is mirrored by formation of DBC-DNA adducts in lung tissue and whether these adducts are consistent with mutation patterns in the K-ras gene. Strain A/J mice were given a single i.p. injection of DBC at doses of 0, 5, 10, 20 or 40 mg/kg and levels of DNA adducts in the lung were monitored by 32P-postlabeling on days 1, 3, 5, 7, 14 and 21. The remaining animals were sacrificed 8 months after DBC treatment and lung tumor multiplicity and K-ras mutation patterns in the tumors were determined. The lung tumor response to DBC was dose related, with an average of 4.7 +/- 1.2 tumors/mouse at 5 mg/kg and 48.1 +/- 5.5 tumors/mouse at 40 mg/kg. As many as seven DBC-DNA adducts were observed in the lung. DNA binding levels in the lung were highest at 40 mg/kg, with maximum binding at 5-7 days. At lower dose levels the maximum binding to DNA decreased and shifted to earlier time points. The DBC-DNA adduct in the lung with the highest level of binding at all dose levels was DBC-DNA adduct 3. The majority of DBC-induced mutations in the K-ras gene in the lung were A-->T (80%) transversions in the third base of codon 61, a mutation that has not been previously observed in chemically induced lung tumors in strain A/J mice.

Aflatoxin and p53 abnormality in duck hepatocellular carcinoma.

Recent studies have revealed that a point mutation at codon 249 in the p53 gene predominates in hepatocellular carcinoma (HCC) cases from Southern Africa and China, where infection with hepatitis B virus (HBV) and contamination of aflatoxin B1 in food are risk factors for HCC. This unique mutation from G to T at the third base in codon 249 observed in human HCC cases is suggested to be linked to aflatoxin exposure. Six ducks with HCC, five of which were fed a diet containing aflatoxin B1 for 1-2 years, were analysed for the presence of point mutations at this codon of the p53 gene by polymerase chain reaction and direct nucleotide sequencing. None of the six ducks with HCC showed the change at this codon regardless of duck hepatitis B virus infection. This suggests that aflatoxin B1 itself might not be involved in the unique mutation at codon 249 in hepatocarcinogenesis, or that other factors coincident with aflatoxin may be responsible for this unique mutation.

Recent aflatoxin exposure and mutation at codon 249 of the human p53 gene: lack of association.

Experiments were done to show whether a G to T mis-sense mutation at the third base of codon 249 of the p53 tumour suppressor gene is a 'hot spot' of aflatoxin attack as suggested by the results of epidemiological studies. Liver tissue from liver cancer patients in Taiwan and Japan was analysed for the presence of aflatoxin-DNA adducts (ADA) as a marker for aflatoxin exposure and an AGG to AGT transversion at codon 249 of the p53 gene. Ten per cent of samples containing ADA, indicating definite exposure of the subjects to aflatoxin, was found to harbour the codon 249 mutation, whereas 18% of the samples with no detectable adducts also contained the mutation. Our data do not support the hypothesis that codon 249 of the p53 gene DNA is a hot spot for aflatoxin mutagenesis as a 'late stage event' in human hepatocellular carcinogenesis.

Induction of activating mutations in the human c-Ha-ras-1 proto-oncogene by oxygen free radicals.

The mutagenicity of oxygen free radicals was studied in a forward mutation system. pEC plasmid containing the human c-Ha-ras-1 proto-oncogene was reacted with oxygen free radicals generated by Cu2+ and H2O2 and was then transfected into NIH/3T3 cells. Transformed foci were observed with oxygen free radical-modified DNA but not with unmodified DNA. The mutations responsible for the Ha-ras-1 gene activation in 11 transformed foci were characterized. G-->T mutations at the second base of codon 12 were found in two transformed foci, A-->T transversions at the second base of codon 61 in five foci, and G-->T mutations at the third position of codon 61 in four transformed foci. These observed mutations are identical to those commonly found in human skin carcinomas, suggesting that reactive oxygen species may play an important role in the carcinogenesis of these tumors. Interestingly, a significant proportion of mutations was found at the second and third base of codon 61 (CAG). In a previous study, the same oxygen free radical-generating system was found to cause an intrastrand cross-link between adjacent purine nucleotides at AG sites in DNA (Carmichael et al., Carcinogenesis 13:1127-1135, 1992). These data demonstrate that oxygen radicals can induce DNA damage that can result in a specific activation of a human proto-oncogene.

Alternative splicing of the p53 tumor suppressor gene in the Molt-4 T-lymphoblastic leukemia cell line

The expression of the p53 tumor suppressor gene in ten human cell lines (nine cancers and one normal) was studied using reverse transcription, polymerase chain reaction (PCR) and direct sequencing. Using P53U and P53D primers for amplifying a 371-base pair (bp) target fragment spanning exons 7–10 of p53 cDNA, normal-sized PCR products were amplified from 9 cell lines but not from the Hep3B hepatocellular carcinoma (HCC) cell line. An additional larger band (504 bp) was observed for the Molt-4 T-lymphoblastic leukemia cell line. Employing P531 and P53D primers which flank a 76-bp p53 cDNA fragment, 76 bp as well as 209 bp products were generated by PCR of Molt-4 cDNA. Direct sequencing of the 504 bp and 209 bp bands confirmed the presence of a 133 bp insertion between exons 9 and 10 in the aberrant transcript. This insertion was homologous to a 130-bp sequence within the wild-type p53 intron 9, except for 2 point mutations and 3 base insertions. Sequencing of P53U P53D PCR products of Molt-4 genomic DNA revealed an 8 bp deletion just downstream to the 133 bp insertion, creating a novel donor splicing site within intron 9. This site, coupled with an inherent acceptor splicing site just upstream to the 133 bp insertion, suggests that the 133 bp stretch represents an alternative exon. The occurrence of a termination signal within this alternative transcript is predicted to culminate in a truncated p53 translational product. The sequences of the 371 bp PCR products of Molt-4, HT-1080, SiHa, CaSki, HeLa and MRC-5 cell lines corresponded with the wild-type p53 cDNA. G → T transversions at the third base of codon 249 of p53 were detected in Mahlavu and PLC/PRF/5 HCC lines, while a TAC to CAC mutation at codon 234 was observed in an allele of the Raji Burkitt lymphoma line.

Three new HLA-G alleles and their linkage disequilibria with HLA-A.

Three new allelic forms of the HLA-G DNA sequence (HLA-G*II, HLA-G*III, and HLA-G*IV) have been identified. With the HLA-G*I sequence (previously designated HLA 6.0) as a reference, HLA-G*II shows a silent (G-->A) mutation at the third base of codon 57, HLA-G*III bears a non-synonymous (A-->T), but conservative, (Thr-->Ser) substitution at the first base of codon 31, and HLA-G*IV shows two silent substitutions: (A-->T) at the third base of codon 107 and (G-->A) at the third base of codon 57. A rapid method of singling out each allele on genomic DNA has been developed by using polymerase chain reaction amplification followed by restriction endonuclease treatment. Also, more or less strong linkage disequilibria has been found between most HLA-A alleles and either HLA-G*I or *II, both being the most prevalent alleles in the population, with a genotypic frequency of 0.55 and 0.38, respectively; HLA-G*III is very rare and HLA-G*IV has a genotypic frequency of 0.07. An evolutive classification of HLA-A alleles results according to their association with either HLA-G*I or HLA-G*II, which does not correlate with the classical serological cross-reacting groups classification. The finding of a strong and selective A/G linkage disequilibria with most HLA-A alleles, together with the existence of less frequent random A/G associations, may suggest that there exist in different haplotypes true and varied A/G genetic distances (and not a recombinational hotspot). It may be inferred from preliminary data that in primates HLA-A/G haplotypes bearing G*II may have appeared later than those bearing G*I.

Aberrations of p53 gene in human hepatocellular carcinoma from China.

Allele losses and mutations have been examined in 38 cases of primary hepatocellular carcinomas (HCC) from different geographic areas of China by Southern, single-strand conformational polymorphism (SSCP) and direct DNA sequencing analyses. Two of 12 samples from Qi-Dong and six of 18 HCCs from Shanghai showed loss of heterozygosity (LOH) at the loci on chromosome 17p13.3. All of the nine mutations in the p53 gene detected in HCC from Qi-Dong were clustered at the third base of codon 249, i.e. G:C to T:A, leading to an arginine to serine change. In contrast, 18 HCC samples from Shanghai contained three mutations at codons 249, 255 and 279. These results suggested a relationship between the spectrum of p53 aberration and environmental risk factors in these two geographic areas. Since no correlation between the state of HBV DNA and p53 aberration was observed, other factors such as dietary exposure to aflatoxin B1 (AFB1) might be responsible for the mutational hotspot at codon 249 in HCCs from Qi-Dong area.

A new polymorphism in exon 11 of the LDL receptor gene in healthy people and in familial hypercholesterolemia subjects.

We have screened exon 11 of the low density lipoprotein receptor (LDLR) gene from familial hypercholesterolemia (FH) heterozygotes for point mutations by using analysis of single strand conformation polymorphisms (SSCP). A variant pattern was observed in three out of 39 subjects. By DNA sequencing, this variant pattern was found to be due to a C-->T transition at nucleotide 1617 that affects the third base of codon 518. A PCR method was developed to screen FH heterozygotes and normal subjects for this mutation. The gene frequencies in FH heterozygotes and normal subjects were 4% and 4.5%, respectively. Thus, the mutation cannot be in linkage disequilibrium with a mutation that causes FH. Rather, the mutation may be a useful genetic marker at the LDLR locus. Haplotype analysis at the LDLR locus in two FH families where the proband possessed the mutation revealed that the mutation was on two different haplotypes. This finding is consistent with the mutation occurring at a mutational hot spot.

Tumours of the liver.

One of the major debates in hepatocellular carcinogenesis at present is whether the hepatitis-B and -C viruses are directly carcinogenic or exert their effect indirectly by causing chronic necro-inflammatory hepatic disease, which in turn is responsible for malignant transformation of hepatocytes. This debate has been fueled by the observation that hepatitis C virus is a single-stranded RNA virus with no precedent for inducing cancer but with a marked propensity to cause chronic necro-inflammatory hepatic disease and by the findings in Chisari's transgenic mouse model, which suggest that severe and prolonged hepatocellular injury per se induces a proliferative response that progresses to tumour formation. Recent reports of a guanine to thymine mutation of the third base of codon 249 of the tumour suppressor gene, p53, in 50% of patients with hepatocellular carcinoma in regions of high aflatoxin exposure, and mutagenic experiments showing that aflatoxin B1 binds particularly to guanine residues in G-C-rich domains and that codon 249 is a preferred target have suggested a mechanism whereby aflatoxin might induce malignant transformation.

The mutation causing the common apolipoprotein A-IV polymorphism is a glutamine to histidine substitution of amino acid 360.

Apolipoprotein (apo) A-IV is a protein involved in the metabolism of chylomicrons and high density lipoproteins. This protein displays genetic polymorphism due to two main codominant alleles, A-IV1 and A-IV2. We have identified the mutation that leads to this polymorphism. It is caused by a single-base substitution of guanine for thymine in the third base of codon 360. This substitution leads to a glutamine to histidine change. Direct sequencing of amplified DNA from eight subjects in a three-generation pedigree has demonstrated that the guanine to thymine substitution can explain the apo A-IV polymorphism. In 32 unrelated individuals, a correspondence between apo A-IV phenotype determined by isoelectric focusing and genotype determined with Fnu4HI digestion of amplified DNA could be demonstrated. The enzyme lecithin:cholesteryl acyltransferase (LCAT) is activated by apo A-IV. Under our in vitro conditions, the isoprotein apo A-IV 1-1 is a better LCAT activator than is the isoprotein apo A-IV 2-2. A knowledge of the molecular mechanism underlying the apo A-IV polymorphism will help to elucidate the mechanisms involved in LCAT activation.

Genotyping of apolipoprotein A-IV by digestion of amplified DNA with restriction endonuclease Fnu4HI: use of a tailored primer to abolish additional recognition sites during the gene amplification

Apolipoprotein A-IV (apoA-IV) is involved in the metabolism of chylomicrons and high density lipoproteins. It displays genetic polymorphism due to two co-dominant alleles apoA-IV1 and apoA-IV2. The mutation that causes the polymorphism is a G to T substitution in the third base of codon 360 in the apoA-IV2 allele which results in a glutamine (Gln) to histidine (His) change of amino acid 360. This substitution leads to the abolition of a recognition site for the restriction enzyme Fnu4HI. Part of the third exon of the apoA-IV gene is amplified by the polymerase chain reaction (PCR) using a tailored primer, which abolishes the downstream recognition sites during the DNA amplification. The PCR products are digested with the restriction enzyme Fnu4HI and electrophoresed on a polyacrylamide gel. The apoA-IV genotypes are determined after staining with either ethidium bromide or silver. To validate the method, we determined the inheritance of the apoA-IV alleles in a three-generation kindred of 8 subjects and analyzed amplified DNA of 32 subjects of different apoA-IV phenotypes with this method. The results were compared to those obtained from isoelectric focusing and immunoblotting. In all cases studied, the two methods gave concordant results.

Nucleotide sequence of the S RNA of Lassa virus (Nigerian strain) and comparative analysis of arenavirus gene products

The nucleotide sequence of the small (S) genomic RNA of Lassa virus (strain GA391, of Nigerian origin) has been determined. The RNA has features which conform to those seen in most other arenavirus S RNAs which have been characterised, including conserved terminal sequences, an ambisense arrangement of the coding regions for the precursor glycoprotein (GPC) and nucleocapsid (N) proteins and an intergenic region capable of forming a base-paired “hairpin” structure. Comparison of the nucleotide sequence with that of the Josiah strain of Lassa virus (from Sierra Leone) reveals considerable nucleotide divergence in the third base of codons in the reading frames of all three proteins, although the resulting protein sequences are highly conserved, with 92, 94 and 91% identical residues for the mature glycoproteins G1 and G2 and the N protein, respectively. Sequence alignments of the available arenavirus structural proteins and dendrograms summarising the relationships between the viral proteins are presented.

Characterization of the human cell line TE671.

The cell line TE671 has been widely used as a model of human medulloblastoma. In the present study we have demonstrated that transfection of DNA from this cell line into NIH 3T3 cells reveals the presence of an activated N-ras gene. Using oligonucleotide probes we have shown that the N-ras gene is activated by a point mutation at the third base of codon 61 resulting in the substitution of histidine for glutamine in the p21 ras gene product. We noted that this relatively uncommon activating mutation is also present in the human rhabdomyosarcoma cell line RD. Based on this finding and on the observation that several of the phenotypic characteristics of TE671, such as the presence of muscle-type nicotinic acetylcholine receptors and the intermediate filament protein desmin, are suggestive of myoid origin we investigated the possible identity of these two cell lines. Cytogenetic analysis revealed the presence of marker chromosomes common to both TE671 and RD. DNA fingerprinting using both locus specific and multilocus core probes showed indistinguishable band patterns in the two cell lines. Taken together our data show that TE671 and RD are derivatives of the same cell line and we conclude that the properties of the TE671 line should be ascribed to rhabdomyosarcoma rather than medulloblastoma cells.

Occurrence of unmodified adenine and uracil at the first position of anticodon in threonine tRNAs in Mycoplasma capricolum.

Codon usage pattern in the threonine four-codon (ACN) box in Mycoplasma capricolum is strongly biased towards adenine and uracil for the third base of codons. Codons ending in uracil or adenine, especially ACU, predominate over ACC and ACG. This bacterium contains two isoacceptor threonine tRNAs having anticodon sequences AGU and UGU, both with unmodified first nucleotides. It would thus appear that ACN codons are translated in an unusual way; tRNA(Thr)(AGU) would translate the most abundantly used codon ACU exclusively, because adenine at the first anticodon position can, according to the wobble rule, pair only with uracil of the third codon position. The tRNA(Thr)(UGU) would mainly be responsible for translation of three other codons, ACA, ACG, and ACC. Anticodon UGU would also be used for reading codon ACU as a redundancy of tRNA(Thr)-(AGU), as deduced from the mitochondrial code where unmodified uracil at the first anticodon position can pair with adenine, cytosine, guanine, and uracil by four-way wobble. The tRNA(Thr)(AGU) has much higher sequence homology to tRNA(Thr)(UGU) from M. capricolum (88%), Bacillus subtilis (77%) and Escherichia coli (86%) than to tRNA(Thr)(GGU) from B. subtilis (66%) and E. coli (63%), suggesting that tRNA(Thr)-(AGU) has been derived from tRNA(Thr)(UGU), but not from tRNA(Thr)(GGU).

Mutations in the human adenosine deaminase gene that affect protein structure and RNA splicing.

Adenosine deaminase (ADA; adenosine aminohydrolase, EC 3.5.4.4) deficiency is one cause of the genetic disease severe combined immunodeficiency. To identify mutations responsible for ADA deficiency, we synthesized cDNAs to ADA mRNAs from two cell lines, GM2756 and GM2825A, derived from ADA-deficient immunodeficient patients. Sequence analysis of GM2756 cDNA clones revealed a different point mutation in each allele that causes amino acid changes of alanine to valine and arginine to histidine. One allele of GM2825A also has a point mutation that causes an alanine to valine substitution. The other allele of GM2825A was found to produce an mRNA in which exon 4 had been spliced out but had no other detrimental mutations. S1 nuclease mapping of GM2825A mRNAs showed equal abundance of the full-length ADA mRNA and the ADA mRNA that was missing exon 4. Several of the ADA cDNA clones extended 5' of the major initiation start site, indicating multiple start sites for ADA transcription. The point mutations in GM2756 and GM2825A and the absence of exon 4 in GM2825A appear to be directly responsible for the ADA deficiency. Comparison of a number of normal and mutant ADA cDNA sequences showed a number of changes in the third base of codons. These changes do not affect the amino acid sequence. Analyses of ADA cDNAs from different cell lines detected aberrant RNA species that either included intron 7 or excluded exon 7. Their presence is a result of aberrant splicing of pre-mRNAs and is not related to mutations that cause ADA deficiency.