Immunophenotypic transformation from acute undifferentiated leukemia to Burkitt's-like acute lymphoblastic leukemia.

Progression of more differentiated to less differentiated malignant phenotypes has been described infrequently during the natural evolution or at relapse of treated hematopoietic malignancies. This report describes an unusual instance of immunophenotypic transformation from an immunologically undifferentiated acute leukemia to a leukemia that at relapse possessed morphologic and immunologic markers characteristic of a Burkitt's-like acute lymphoblastic leukemia. A 26-year-old man initially presented with pancytopenia and a bone marrow diffusely replaced with blast cells morphologically most consistent with a French-American-British L2 subclassification. The surface immunophenotype of the blasts at diagnosis showed HLA-DR surface antigen but no myeloid, lymphoid, or immunoglobulin determinants. Despite successful induction and ongoing consolidation chemotherapy, the patient had a relapse five months after diagnosis; blast cells at relapse demonstrated marked cytoplasmic vacuolation consistent with a Burkitt's-like L3 acute lymphoblastic transformation. Immunophenotypic analysis revealed the presence of restricted immunoglobulin determinants (mu heavy chain and kappa light chain), as well as two separate B lineage surface determinants (BA-1 and B-1). Immunophenotypic transformations may reflect the presence of either a multiclonal or multipotent leukemic population; documentation of the frequency of such transformations and genomic analysis of the transformed subpopulations may be helpful in furthering the understanding of molecular mechanisms involved in leukemogenesis.

Selective inhibition of the growth of human erythroid bursts by monoclonal antibodies against transferrin or the transferrin receptor.

The relative requirements of colonies derived from erythroid (BFU-E) and myeloid (CFU-c) progenitors for transferrin were examined using monoclonal antibodies directed against the transferrin molecule (TF-6) or its cell surface receptor (TFR-A12, TFR1-2B). Growth of erythroid bursts was profoundly reduced at concentrations of all three antibodies that had no effect on CFU-c-derived colonies. When TFR1-2B was layered over cultures established one to seven days previously, further burst development was inhibited, and degeneration of early erythroid colonies was observed. Addition of erythropoietin augmented transferrin receptor expression on cells harvested after 1 to 2 weeks in culture and analyzed by flow cytometry. Recombinant human erythropoietin gave results comparable to those obtained in experiments using human urinary erythropoietin. Analysis of erythroblasts plucked directly from culture plates confirmed the presence of transferrin receptors on BFU-E-derived colonies. Thymidine incorporation was maximal early in the second week of culture and coincided with high transferrin receptor expression. These data demonstrate that transferrin must be available into the second week of culture to support the growth and differentiation of BFU-E-derived erythroid bursts, that the generation of erythroid colonies from BFU-E is more dependent on transferrin than myeloid colony formation from CFU-c, and that erythropoietin modulates the expression of transferrin receptors on growing bursts.

Markers of genetic variation among the Waorani Indians of the Ecuadorian Amazon headwaters.

Until recently, the Waorani Indians of Ecuador's Amazon headwaters maintained a fierce resistance to all intruders into their territory, and as a result of their actions and reputations a population of 600 people controlled a very large territory (about 8,000 square miles). The isolation of the Waorani has resulted in a large linguistic and genetic distance from their neighbors. Our survey of red cell enzymes, immunoglobulin allotypes, and dermatoglyphics demonstrates that the Waorani are a highly inbred and homogeneous population. Of 18 red cell enzymes studied, the Waorani have a limited polymorphism for only 6. Only two Gm haplotypes (Gm1,2,17,21, Gm1,17,21) were found and 60% of those tested were homozygous for the Gm1,17,21 haplotype. All individuals were A2m (1) and 95% of these were homozygous. The Waorani's dermatoglyphic traits fell within the wide range found among other South American Indians with close affinity to the Ecuadorian Jivaro group. Despite the limitations of these genetic systems, they demonstrate that the Waorani share limited genetic traits with the neighboring Jivaro Indians and are isolated from other tribal populations in South America.

The U3 portion of feline leukemia virus DNA identifies horizontally acquired proviruses in leukemic cats.

The presence and location of DNA sequences related to the U3 and U5 portions of the infectious exogenous feline leukemia virus (FeLV) long terminal repeat (LTR) in various cat DNAs have been determined by hybridization experiments. In uninfected cat DNAs, the U5 LTR segment from the Gardner-Arnstein strain B virus is present at approximately 150 copies per cell. This level is approximately 10-fold greater than that of endogenous internal FeLV sequences. The U5 sequences differ in copy number and, to some extent, in location from one animal to another. For any one animal, the sequence organization of the U5 segments is the same among different tissues, showing that the pattern is inherited through the germ line. Most importantly, the viral U3 LTR probe hybridizes only very weakly with uninfected cat DNAs. Both the U3 and the U5 regions of the LTR from the Gardner-Arnstein strain of virus cross-hybridize with DNA derived from four other infectious FeLVs representing A, B, and C subtypes. Thus, the C3 region may be used as a probe for studying the number and location of exogenously acquired FeLV proviruses in infected cat tissues. In some cases exogenously acquired proviruses are present in unique sites in the genome of virus-positive cat lymphosarcomas, indicating a monoclonal origin for the tumor. In other tumors, the proviral sequences are randomly distributed over many sites. Lymphosarcomas of virus-negative cats have no exogenous U3 sequences despite epidemiological evidence of an association of virus-negative leukemia with exposure to FeLV.

Sequence arrangement and biological activity of cloned feline leukemia virus proviruses from a virus-productive human cell line.

We examined 14 different feline leukemia virus proviruses from the productively infected human cell line RD(FeLV)-2 after cloning in the modified lambda vector Charon 4A. Each isolate was characterized by restriction digestion and Southern blot analysis. The DNA of each isolate was tested for competence to express virus after uptake by sensitive animal cells (transfection). All but one isolate contained an apparently complete provirus, but only four were infectious. Seven isolates (four noninfectious, three infectious) were studied by heteroduplexing followed by electron microscopy or by S1 nuclease treatment and gel electrophoresis. No regions of nonhomology between proviruses were detected by either criterion, and in no case did we observe homology between flanking sequences. Random shearing or removal of flanking sequences by S1 nuclease had no effect on the status of infectivity of the clones. Thus, we were unable to find molecular differences between infectious and noninfectious proviruses. Our data are consistent with either of the following hypotheses: (i) that there is a short host sequence which is essential as a promoter for virus expression; or (ii) that lack of infectivity is due to small mutations within the proviral genome.

Electron microscopic analysis of partially replicated bacteriophage T7 DNA.

Partially replicated bacteriophage T7 DNA was isolated from Escherichia coli infected with UV-irradiated T7 bacteriophage and was analyzed by electron microscopy. The analysis determined the distribution of eye forms and forks in the partially replicated molecules. Eye forms and forks in unit length molecules were aligned with respect to the left end of the T7 genome, and segments were scored for replication in each molecule. The resulting histogram showed that only the left 25 to 30% of the molecules was replicated. Several different origins of DNA replication were used to initiate replication in the UV-irradiated experiments in which 32P-labeled progeny DNA from UV-irradiated phage was annealed with ordered restriction fragments of T7 DNA (K. B. Burck and R. C. Miller, Jr., Proc. Natl. Acad. Sci. U.S.A. 75:6144--6148, 1978). Both analyses support partial-replica hypotheses (N. A. Barricelli and A. H. Doermann, Virology 13:460--476, 1961; Doermann et al., J. Cell. comp. Physiol. 45[Suppl.]:51--74, 1955) as an explanation for the distribution of marker rescue frequencies during cross-reactivation; i.e., replication proceeds in a bidirectional manner from an origin to a site of UV damage, and those regions of the genome which replicate most efficiently are rescued most efficiently by a coinfecting phage. In addition, photoreactivation studies support the hypothesis that thymine dimers are the major UV damage blocking cross-reactivation in the right end of the T7 genome.

Marker rescue and partial replication of bacteriophage T7 DNA.

Experiments reported here show that some UV-irradiated wild-type T7 phage markers can be rescued efficiently by coinfection with T7 amber mutant phage in a permissive host. Other results show that the segments of a UV-irradiated genome that replicate efficiently are those that also are rescued efficiently during a marker rescue experiment. At higher doses, fewer markers are rescued efficiently and fewer segments of the genome replicate efficiently. The results clearly indicate that the probability of marker rescue is correlated with the ability of the DNA containing the marker to replicate. Sucrose density gradient analysis shows that UV irradiation does not produce double-strand scissions in T7 DNA at doses used here. Therefore, the partial replication and rescue of markers from the left end of the genome is not due simply to injection of only the left end of the T7 DNA.