Evaluation of performance of the Gen-Probe human immunodeficiency virus type 1 viral load assay using primary subtype A, C, and D isolates from Kenya.

Accurate and sensitive quantification of human immunodeficiency virus type 1 (HIV-1) RNA has been invaluable as a marker for disease prognosis and for clinical monitoring of HIV-1 disease. The first generation of commercially available HIV-1 RNA tests were optimized to detect the predominant HIV-1 subtype found in North America and Europe, subtype B. However, these tests are frequently suboptimal in detecting HIV-1 genetic forms or subtypes found in other parts of the world. The goal of the present study was to evaluate the performance of a new viral load assay with non-subtype B viruses. A transcription-mediated amplification method for detection and quantitation of diverse HIV-1 subtypes, called the Gen-Probe HIV-1 viral load assay, is under development. In this study we examined the performance of the Gen-Probe HIV-1 viral load assay relative to that of the commonly used commercial HIV-1 RNA assays using a panel of primary isolates from Kenya. For comparison, we included several subtype B cloned viruses, and we quantified each virus using an in-house quantitative-competitive reverse transcriptase PCR (QC-RT-PCR) method and gag(p24) antigen capture. The Gen-Probe HIV-1 viral load assay and a version of the Roche AMPLICOR HIV-1 MONITOR test (version 1.5) that was designed to detect a broader range of subtypes were both sensitive for the quantification of Kenyan primary isolates, which represented subtype A, C, and D viruses. The Gen-Probe HIV-1 viral load assay was more sensitive for the majority of viruses than the Roche AMPLICOR HIV-1 MONITOR test version 1.0, the Bayer Quantiplex HIV RNA 3.0 assay, or a QC-RT-PCR method in use in our laboratory, suggesting that it provides a useful method for quantifying HIV-1 RNAs from diverse parts of the world, including Africa.

Use of calibrated viral load standards for group M subtypes of human immunodeficiency virus type 1 to assess the performance of viral RNA quantitation tests.

Optimal management of human immunodeficiency virus type 1 (HIV-1) disease requires accurate quantitation of viral RNA concentrations in plasma. Evidence for increasing geographic intermixing of HIV-1 subtypes makes equivalent quantitation of all subtypes essential. The performances of six quantitative viral RNA tests are described, for the first time, with calibrated viral isolates of diverse subtypes.

Subtypes of human immunodeficiency virus type 1 and disease stage among women in Nairobi, Kenya.

In sub-Saharan Africa, where the effects of human immunodeficiency virus type 1 (HIV-1) have been most devastating, there are multiple subtypes of this virus. The distribution of different subtypes within African populations is generally not linked to particular risk behaviors. Thus, Africa is an ideal setting in which to examine the diversity and mixing of viruses from different subtypes on a population basis. In this setting, it is also possible to address whether infection with a particular subtype is associated with differences in disease stage. To address these questions, we analyzed the HIV-1 subtype, plasma viral loads, and CD4 lymphocyte levels in 320 women from Nairobi, Kenya. Subtype was determined by a combination of heteroduplex mobility assays and sequence analyses of envelope genes, using geographically diverse subtype reference sequences as well as envelope sequences of known subtype from Kenya. The distribution of subtypes in this population was as follows: subtype A, 225 (70.3%); subtype D, 65 (20.5%); subtype C, 22 (6.9%); and subtype G, 1 (0.3%). Intersubtype recombinant envelope genes were detected in 2.2% of the sequences analyzed. Given that the sequences analyzed represented only a small fraction of the proviral genome, this suggests that intersubtype recombinant viral genomes may be very common in Kenya and in other parts of Africa where there are multiple subtypes. The plasma viral RNA levels were highest in women infected with subtype C virus, and women infected with subtype C virus had significantly lower CD4 lymphocyte levels than women infected with the other subtypes. Together, these data suggest that women in Kenya who are infected with subtype C viruses are at more advanced stages of immunosuppression than women infected with subtype A or D. There are at least two models to explain the data from this cross-sectional study; one is that infection with subtype C is associated with a more rapid disease progression, and the second is that subtype C represents an older epidemic in Kenya. Discriminating between these possibilities in a longitudinal study will be important for increasing our understanding of the role of specific subtypes in the transmission and pathogenesis of HIV-1.

Evaluation of Bcl-2/B cell transgenic mice (B6) for hybridoma production.

Bcl-2 is an oncogene associated with prevention of apoptosis in a variety of cell types. Bcl-2 expression in B lymphoid cells prolongs antibody production, in vitro and in vivo. A line of transgenic mice (B6) has been developed that expresses human Bcl-2 in the B cells of SWR/SJL mice. B6 transgenic, nontransgenic littermates, and BALB/c mice were immunized with beta-galactosidase (B-gal) or sheep red blood cells (SRBC). The number of spleen cells recovered from immunized B6 mice was 3-4 times greater than syngeneic, nontransgenic littermates or BALB/c mice. Spleen cells from B-gal or SRBC immune B6, SWR/SJL, and BALB/c mice were fused with P3 myeloma cells to produce hybridomas. Forty-eight percent of the wells plated with fused B6 spleen cells produced B-gal-specific antibodies compared to 14% from BALB/c and 12% from SWR/SJL. Antibody-specific wells were subcloned, resulting in enhanced recovery of antigen-specific subclones with B6-derived fusions compared to controls. In the SRBC fusions, 17% of the wells plated with fused B6 spleen cells produced SRBC-specific antibodies compared to 6% for BALB/c and SWR/SJL spleens. After subcloning, B6-derived clones produced 8% positive subclones compared to 9.5% from SWR/SJL and 3.5% from BALB/c. Comparison of the isotype distribution of subclones showed a higher ratio of IgG antibodies compared to IgM from B6 mice in the B-gal fusions. IgA antibodies were recovered only from B6 mice. These data indicate that B6 transgenic mice that overexpress Bcl-2 in their B cells may be superior to other mouse strains for production of antigen-specific hybridomas.

BCL-2 family proteins: regulators of cell death involved in the pathogenesis of cancer and resistance to therapy.

The BCL-2 gene was first discovered because of its involvement in the t(14;18) chromosomal translocations commonly found in lymphomas, which result in deregulation of BCL-2 gene expression and cause inappropriately high levels of Bcl-2 protein production. Expression of the BCL-2 gene can also become altered in human cancers through other mechanisms, including loss of the p53 tumor suppressor which normally functions as a repressor of BCL-2 gene expression in some tissues. Bcl-2 is a blocker of programmed cell death and apoptosis that contributes to neoplastic cell expansion by preventing cell turnover caused by physiological cell death mechanisms, as opposed to accelerating rates of cell division. Overproduction of the Bcl-2 protein also prevents cell death induced by nearly all cytotoxic anticancer drugs and radiation, thus contributing to treatment failures in patients with some types of cancer. Several homologs of Bcl-2 have recently been discovered, some of which function as inhibitors of cell death and others as promoters of apoptosis that oppose the actions of the Bcl-2 protein. Many of these Bcl-2 family proteins can interact through formation of homo- and heterotypic dimers. In addition, several nonhomologous proteins have been identified that bind to Bcl-2 and that can modulate apoptosis. These protein-protein interactions may eventual serve as targets for pharmacologically manipulating the physiological cell death pathway for treatment of cancer and several other diseases.

Biochemical and functional comparisons of Mcl-1 and Bcl-2 proteins: evidence for a novel mechanism of regulating Bcl-2 family protein function.

Mcl-1 is a recently described homologue of Bcl-2 whose function and biochemical characteristics remain poorly defined. Gene transfer experiments in lnterleukin-3 (IL-3)-dependent myeloid progenitor 32D.3 cells and pro-B-lymphoid FL5.12 cells demonstrated that enforced production of high levels of Mcl-1 protein failed to prolong the survival of cells when cultured in the absence of IL-3, whereas Bcl-2 did delay cell death. Mcl-1 also did not prolong the survival in vitro of 32D.3 cells that had been induced to differentiate into mature neutrophils using Granulocyte-Colony Stimulating Factor (G-CSF), whereas Bcl-2 did. 32D.3 and FL5.12 cells co-transfected with Mcl-1 and Bcl-2 displayed survival kinetics essentially identical to cells transfected with Bcl-2 alone, when cultured in the absence of IL-3, indicating that Mcl-1 neither enhances nor impairs Bcl-2 function. In contrast to the lack of effects of Mcl-1 in 32D.3 and FL5.12 cells, Mcl-1 (like Bcl-2) was able to neutralise Bax-induced cytotoxicity in yeast (S. cerevisiae). Moreover, the recombinant GST-Mcl-1 protein bound specifically to in vitro translated Bax protein, as well as to Bax protein present in detergent lysates prepared from 32D.3 and FL5.12 cells, based on in vitro binding assays. However, Mcl-1 and Bax proteins could not be co-immunoprecipitated from control and transfected 32D.3 and FL5.12 cells, whereas Bcl-2 and Bax were easily co-immunoprecipitated under the same conditions. The findings suggest that while Mcl-1 has the capacity to bind to and neutralise the cell death promoting activity of Bax, other factors such as perhaps additional proteins or undefined post-translational modifications may influence its ability to bind to Bax in vivo and thus affect its function as a cell death blocker.

Immunohistochemical analysis of Mcl-1 protein in human tissues. Differential regulation of Mcl-1 and Bcl-2 protein production suggests a unique role for Mcl-1 in control of programmed cell death in vivo.

The mcl-1 gene encodes an approximately 37-kd protein that has significant homology with Bcl-2, an inhibitor of programmed cell death that is expressed in many types of long-lived cells. In this study we determined the in vivo patterns of Mcl-1 protein production in normal human tissues by immunohistochemical means, using specific polyclonal antisera, and made comparisons with Bcl-2. Like Bcl-2, Mcl-1 immunostaining was observed in epithelial cells in a variety of tissues, including prostate, breast, endometrium, epidermis, stomach, intestine, colon, and respiratory tract. However, often the expression of mcl-1 and bcl-2 in complex epithelia occurred in gradients with opposing directions, such that Bcl-2 immunostaining tended to be higher in the less differentiated cells lining the basement membrane, whereas Mcl-1 immunostaining was more intense in the differentiated cells located in the upper layers of these epithelia. The in vivo patterns of mcl-1 and bcl-2 expression were also strikingly different in several other tissues as well. Within the secondary follicles of lymph nodes and tonsils, for example, germinal center lymphocytes were Mcl-1 positive but mostly lacked Bcl-2; whereas mantle zone lymphocytes expressed bcl-2 but not mcl-1. Intense Mcl-1 immunoreactivity was also detected in several types of neuroendocrine cells, including the adrenal cortical cells that are Bcl-2 negative, sympathetic neurons that also contain Bcl-2, a subpopulation of cells in the pancreatic islets, Leydig cells of the testis, and granulosa lutein cells of the ovarian corpus luteum but not in thyroid epithelium, which is strongly Bcl-2 positive. Little or no Mcl-1 was detected in neurons in the brain and spinal cord, in contrast to Bcl-2, which is present in several types of central nervous system neurons. Conversely, strong Mcl-1 immunostaining was found in cardiac and skeletal muscle, which contain comparatively less Bcl-2. Additional types of cells that are Bcl-2-negative but that expressed mcl-1 include chondrocytes and hepatocytes. These findings demonstrate that mcl-1 expression is widespread in vivo and imply that the Mcl-1 and Bcl-2 proteins fulfill different roles in the overall physiology of cell death regulation.

Immunohistochemical analysis of Mcl-1 and Bcl-2 proteins in normal and neoplastic lymph nodes.

The Bcl-2 protein blocks programmed cell death and becomes overproduced in many follicular non-Hodgkin's lymphomas as the result of t(14; 18) translocations involving the Bcl-2 gene. Mcl-1 is a recently discovered gene whose encoded protein has significant homology with Bcl-2 but whose function remains unknown. In this study, we compared the in vivo patterns of Bcl-2 and Mcl-1 protein production in normal and neoplastic lymph node biopsies by immunohistochemical means using specific polyclonal antisera. Intracellular Mcl-1 immunoreactivity was located primarily in the cytosol in a punctate pattern and was also seen in association with the nuclear envelope in many cases, similar to the results obtained for Bcl-2, which resides in the outer mitochondrial membrane, nuclear envelope, and endoplasmic reticulum. In 4 of 4 reactive tonsils and 28 of 28 nodes with reactive follicular hyperplasia, reciprocal patterns of Bcl-2 and Mcl-1 protein expression were observed. Bcl-2 immunostaining was highest in mantle zone lymphocytes and absent from most germinal center cells, whereas Mcl-1 immunoreactivity was highest in germinal center lymphocytes and absent from mantle zone lymphocytes. Mcl-1 was also expressed in some interfollicular lymphocytes, particularly those that had the appearance of activated lymphocytes. Similar to the patterns of Bcl-2 and mcl-1 expression seen in reactive nodes, Mcl-1 protein was largely absent from the malignant cells in 2 of 2 mantle cell lymphomas, whereas strong Bcl-2 immunostaining was found in these cells. In contrast to normal nodes, however, the neoplastic follicles of t(14;18) containing follicular non-Hodgkin's lymphomas immunostained positively for both Bcl-2 and Mcl-1 in 24 of 27 cases. Intense immunostaining for Mcl-1 was also observed in Reed-Sternberg cells in 2 of 2 cases of Hodgkin's disease but Bcl-2 immunoreactivity was present at much lower levels. These findings demonstrate that the levels of Mcl-1 and Bcl-2 proteins are differentially regulated in normal and neoplastic cells in lymph nodes and thus suggest different roles for these proteins in the control of cell life and death in these tissues.

Interactions among members of the Bcl-2 protein family analyzed with a yeast two-hybrid system.

Interactions of the Bcl-2 protein with itself and other members of the Bcl-2 family, including Bcl-X-L, Bcl-X-S, Mcl-1, and Bax, were explored with a yeast two-hybrid system. Fusion proteins were created by linking Bcl-2 family proteins to a LexA DNA-binding domain or a B42 trans-activation domain. Protein-protein interactions were examined by expression of these fusion proteins in Saccharomyces cerevisiae having a lacZ (beta-galactosidase) gene under control of a LexA-dependent operator. This approach gave evidence for Bcl-2 protein homodimerization. Bcl-2 also interacted with Bcl-X-L and Mcl-1 and with the dominant inhibitors Bax and Bcl-X-S. Bcl-X-L displayed the same pattern of combinatorial interactions with Bcl-2 family proteins as Bcl-2. Use of deletion mutants of Bcl-2 suggested that Bcl-2 homodimerization involves interactions between two distinct regions within the Bcl-2 protein, since a LexA protein containing Bcl-2 amino acids 83-218 mediated functional interactions with a B42 fusion protein containing Bcl-2 amino acids 1-81 but did not complement a B42 fusion protein containing Bcl-2 amino acids 83-218. In contrast to LexA/Bcl-2 fusion proteins, expression of a LexA/Bax protein was lethal to yeast. This cytotoxicity could be abrogated by B42 fusion proteins containing Bcl-2, Bcl-X-L, or Mcl-1 but not those containing Bcl-X-S (an alternatively spliced form of Bcl-X that lacks a well-conserved 63-amino acid region). The findings suggest a model whereby Bax and Bcl-X-S differentially regulate Bcl-2 function, and indicate that requirements for Bcl-2/Bax heterodimerization may be different from those for Bcl-2/Bcl-2 homodimerization.

Cyclin D1 transgene impedes lymphocyte maturation and collaborates in lymphomagenesis with the myc gene.

Cyclin D1 is the regulatory subunit of certain protein kinases thought to advance the G1 phase of the cell cycle. Deregulated cyclin D1 expression has been implicated in several human neoplasms, most consistently in centrocytic B lymphoma, where the cyclin D1 gene usually has been translocated to an immunoglobulin locus. To determine directly whether constitutive cyclin D1 expression is lymphomagenic, transgenic mice were generated having the cyclin D1 gene linked to an immunoglobulin enhancer. Despite abundant transgene expression, their lymphocytes were normal in cell cycle activity, size and mitogen responsiveness, but young transgenic animals contained fewer mature B- and T-cells. Although spontaneous tumours were infrequent, lymphomagenesis was much more rapid in mice that co-expressed the cyclin D1 transgene and a myc transgene than in mice expressing either transgene alone. Moreover, the spontaneous lymphomas of myc transgenic animals often ectopically expressed the endogenous cyclin D1 gene. These findings indicate that this G1 cyclin can modulate differentiation and collaborate with myc-like genes in oncogenesis.

Genetic linkage map of facioscapulohumeral muscular dystrophy and five polymorphic loci on chromosome 4q35-qter.

A genetic map of five polymorphic markers in the area of the facioscapulohumeral muscular dystrophy (FSHD) gene on chromosome 4q35-qter has been constructed. With these five markers, a number of recombinants have been identified that allow ordering of the marker and the disease loci. The most likely locus order and the relative position of the FSHD gene supported by the recombinants is centromere-D4S171-F11-D4S187-D4S163-D4S139-FS HD-telomere. However, at least one recombination event appears to be inconsistent with this order and suggests a location of FSHD proximal to D4S139.

Genetic and physical mapping on chromosome 4 narrows the localization of the gene for facioscapulohumeral muscular dystrophy (FSHD).

We have used a combination of classical RFLPs and PCR-based polymorphisms including CA repeats and single-strand conformation polymorphisms to generate a fine-structure genetic map of the distal long arm of chromosome 4q. This map is now genetically linked to the pre-existing anchor map of 4pter-4q31 and generates, for the first time, a complete linkage map of this chromosome. The map consists of 32 anchor loci placed with odds of greater than 1,000:1. The high-resolution map in the cytogenetic region surrounding 4q35 provides the order 4cen-D4S171-F11-D4S187-D4S163-D4S139-4q ter. When we used somatic cell hybrids from a t(X;4)(p21;q35) translocation, these five markers fell into three groups consistent with the genetic map-D4S171 and F11 in 4pter-4q35, D4S163 and D4S139 in 4q35-4qter, and D4S187 as a junction fragment between these two regions. These markers are in tight linkage to the gene for facioscapulo-humeral muscular dystrophy (FSHD) mapped to this region by several collaborating investigators and provide a framework for further detailed analysis of this region.

Suppression of growth factor-induced CYL1 cyclin gene expression by antiproliferative agents.

A recently identified novel mammalian cyclin (CYL1), induced by growth factors and apparently functional during the G1 phase of the cell cycle, is of potential significance, given that cell division is primarily controlled in G1. We have measured CYL1 gene expression in murine bone marrow-derived macrophages (BMM), a normal cell type dependent upon colony-stimulating factors (CSFs) for survival and proliferation. The induction of CYL1 mRNA levels correlated strongly with stimulation of DNA synthesis, since elevated CYL1 mRNA levels occurred in response to the mitogenic stimuli, CSF-1, and granulocyte/macrophage CSF, but not to nonmitogenic macrophage-activating agents. BMM are subject to cell cycle arrest by numerous agents, including tumor necrosis factor alpha, interferon gamma, bacterial lipopolysaccharide, and agents that increase cAMP. These antiproliferative agents suppressed CSF-1-stimulated CYL1 gene expression, even when added late in G1. This pattern of CYL1 gene expression was remarkably consistent with the ability of these agents to inhibit progression into S phase. The mechanisms of negative growth regulation are largely unknown, and given the likely importance of G1 cyclins in the control of cell division, we propose that antiproliferative agents may exert their effects by suppressing G1 cyclin gene expression.

Molecular analysis of X-autosome translocations in females with Duchenne muscular dystrophy.

To further an understanding of the mechanism of constitutional chromosomal rearrangement, the translocation breakpoints of two X-autosome translocations carried by females with Duchenne or Becker muscular dystrophy have been mapped, cloned and sequenced. Breakpoints were mapped to specific introns within the dystrophin gene and intron sequences spanning the two breakpoints were cloned and used as probes to identify DNA fragments containing the translocation junctions. The junction-containing fragments were cloned after amplification by inverse PCR or single-specific-primer PCR. Sequence through the junctions and the autosomal regions spanning the breakpoints identified the mechanism of rearrangement as non-homologous exchange with minor additions or deletions (0-8 nucleotides) at the breakpoints. Paternal origin of these X-autosome translocations, coupled with evidence for non-transmission of X-autosome translocations through male meiosis suggested that the translocations were the result of a post-meiotic rearrangement in spermiogenesis.

Prenatal identification of a girl with a t(X;4)(p21;q35) translocation: molecular characterisation, paternal origin, and association with muscular dystrophy.

There are 23 females known with Duchenne or Becker muscular dystrophy (DMD or BMD) who have X;autosome translocations that disrupt the X chromosome within band p21. A female with a t(X;4)(p21;q35) translocation was identified prenatally at routine amniocentesis. At birth, she was found to have a raised CK level, consistent with a diagnosis of Duchenne muscular dystrophy. Her cells were fused with mouse RAG cells and the translocated chromosomes were separated from one another and from the normal X chromosome by segregation in the resulting somatic cell hybrids. Southern blot analysis of the hybrids indicated that the translocation occurred on the X chromosome between genomic probes GMGX11 and J-66, both of which lie within the DMD gene. Further localisation with a subfragment of the DMD cDNA clone placed the translocation breakpoint in an intron towards the middle of the gene, confirming that the de novo translocation disrupted the DMD gene. RFLP analysis of the patient, her parents, and the hybrid cell lines showed that the translocation originated in the paternal genome. This brings to six out of six the number of DMD gene translocations of paternal origin, a fact that may be an important clue in future studies of the mechanism by which X;autosome translocations arise.

Molecular and phenotypic analysis of patients with deletions within the deletion-rich region of the Duchenne muscular dystrophy (DMD) gene.

Eighty unrelated individuals with Duchenne muscular dystrophy (DMD) or Becker muscular dystrophy (BMD) were found to have deletions in the major deletion-rich region of the DMD locus. This region includes the last five exons detected by cDNA5b-7, all exons detected by cDNA8, and the first two exons detected by cDNA9. These 80 individuals account for approximately 75% of 109 deletions of the gene, detected among 181 patients analyzed with the entire dystrophin cDNA. Endpoints for many of these deletions were further characterized using two genomic probes, p20 (DXS269; Wapenaar et al.) and GMGX11 (DXS239; present paper). Clinical findings are presented for all 80 patients allowing a correlation of phenotypic severity with the genotype. Thirty-eight independent patients were old enough to be classified as DMD, BMD, or intermediate phenotype and had deletions of exons with sequenced intron/exon boundaries. Of these, eight BMD patients and one intermediate patient had gene deletions predicted to leave the reading frame intact, while 21 DMD patients, 7 intermediate patients, and 1 BMD patient had gene deletions predicted to disrupt the reading frame. Thus, with two exceptions, frameshift deletions of the gene resulted in more severe phenotype than did in-frame deletions. This is in agreement with recent findings by Baumbach et al. and Koenig et al. but is in contrast to findings, by Malhotra et al., at the 5' end of the gene.

Myogenic regulation of dystrophin gene expression.

Muscle-specific transcriptional regulation of DMD gene expression has been inferred from both the histopathology of the disease and, more recently, from the use of cDNA sequences to detect DMD gene transcripts by Northern blot, RNase protection, in situ hybridization, and polymerase chain reaction (PCR) analyses. Several muscle-specific genes have been shown to be transcriptionally activated early in myogenesis and a number of cis-acting promoter elements required for muscle-specific transcriptional induction have been described. In this report we review our recent progress on studies of the mechanisms underlying myogenic regulation of dystrophin gene expression. Indirect immunofluorescence has been used to demonstrate that dystrophin is present at the muscle cell surface very early in the myogenic program. The cloning and sequencing of the dystrophin gene promoter reveals the presence of pre-defined muscle-specific cis-acting promoter elements. Functional assays provide evidence that these upstream sequences are capable of regulating DMD gene expression in a cell-and developmental stage-specific manner.

Mapping of four translocation breakpoints within the Duchenne muscular dystrophy gene.

There are over 20 females with Duchenne or Becker muscular dystrophy (DMD or BMD) who have X-autosome translocations that break the X chromosome within band Xp21. Several of these translocations have been mapped with genomic probes to regions throughout the large (approximately 2000 kb) DMD gene. In this report, a cDNA clone from the 5' end of the gene was used to further map the breakpoints in four X-autosome translocations. A t(X;21) translocation in a patient with BMD and a t(X;1) translocation in a patient with DMD were found to break within a large 110-kb intron between exons 7 and 8. Two other DMD translocations, t(X;5) and t(X;11), were found to break between the first and the second exon of the gene within a presumably large intron (greater than 100 kb). These results demonstrate that all four translocations have disrupted the DMD gene and make it possible to clone and sequence the breakpoints. This will in turn determine whether these translocations occur by chance in these large introns or whether there are sequences that predispose to translocations.

Frame-shift deletions in patients with Duchenne and Becker muscular dystrophy.

Duchenne muscular dystrophy (DMD) and its less severe form Becker muscular dystrophy (BMD) are allelic disorders. It has been suggested that in the mutations involving BMD, the translational reading frame of messenger RNA is maintained and a smaller, though partially functional, protein is produced. In order to test this, the exon-intron boundaries of the first ten exons of the DMD gene were determined, and 29 patients were analyzed. In a number of BMD patients (mild and severe BMD), the reading frame of messenger RNA was not maintained. On the basis of these findings, a model for reinitiation from an internal start codon is suggested.