Helper T (Th) cell differentiation is highly regulated by cytokines but initiated by mitogens. By examining gene expression in discrete generations of dividing cells, we have delineated the relationship between proliferation and differentiation. Initial expression of IL-2 is cell cycle-independent, whereas effector cytokine expression is cell cycle-dependent. IFNgamma expression increases in frequency with successive cell cycles, while IL-4 expression requires three cell divisions. Cell cycle progression and cytokine signaling act in concert to relieve epigenetic repression and can be supplanted by agents that hyperacetylate histones and demethylate DNA. Terminally differentiated cells exhibit stable epigenetic modification and cell cycle-independent gene expression. These data reveal a novel mechanism governing Th cell fate that initially integrates proliferative and differentiative signals and subsequently maintains stability of the differentiated state.

SummaryA number of investigations have established the critical role ofinterleukin 4 (IL-4) in mediating the development of T helper (Th)2 effector cells in vitro and in vivo. Despite intensive study, the origin of the IL-4 required for Th2 priming and differentiation remains unclear. Natural killer (NK)1.1 + ot/[3 T cell receptor + T (NT) cells, a unique lineage of cells capable of producing large amounts of IL-4 after activation in vivo, are important candidates for directing Th2 priming. These cells are selected by the nonpolymorphic major histocompatibility complex (MHC) class I molecule, CD1, and are deficient in [~2-microglobulin ([32m)-null mice. We used [32m-deficient mice on both BALB/c and C57BL/6 backgrounds to examine their capacity to mount Th2 immune responses after challenge with a number of well-characterized antigens administered by a variety of routes. As assessed by immunization with protein antigen, infection with Leishmania major, embolization with eggs of Schistosoma mansoni, intestinal infection with Nippostrongylus brasiliensis, or induction of airway hyperreactivity to aerosolized antigen, [32m-deficient mice developed functional type 2 immune responses that were not substantially different than those in wild-type mice. Production of IL-4 and the generation ofimmunoglobulin E (IgE) and eosinophil responses were preserved as assessed by a variety of assays. Collectively, these results present a comprehensive analysis of type 2 immune responses in [32m-deficient mice, and indicate that [32m-dependent NT cells are not required for Th2 development in vivo.

SummaryThe role of CD28-mediated signals in T helper cell maturation is not fully understood. We tested the requirement for costimulation through CD28 in several systems of CD4 + T cell differentiation. In vivo priming of mice with genetic disruption of CD28 (CD28 -/-) yielded normal levels of antigen-specific interferon ~/production but markedly diminished levels of interleukin 4 (IL-4) after in vitro restimulation. In response to the pathogenic microbe, Leishmania major, C57BL6 CD28 -/-mice were fully capable of controlling infection and exhibited a normal T helper 1 response. BALB/c CD28 -/-mice unexpectedly exhibited normal susceptibility to L. major. BALB/c CD28 -/-mice developed high levels oflL-4 mRNA and protein induction in the draining lymph nodes. In addition, susceptibility of BALB/c CD28 -/-mice was reversed by neutralization of IL-4 in vivo. We also activated transgenic CD28-bearing T cells from the BALB and C57BL background in vitro in the presence of CTLA4Ig. BALB cells had greater IL-4-producing capacity than C57BL cells in the absence of costimulation. Diverse factors including costimulatory signals, genetic polymorphism, and the nature of the immunogen all influence T helper phenotype commitment, but these results provide evidence that CD28 is not an absolute requirement for generating either Thl or Th2 responses.U 'nder most circumstances, successful T cell activation requires a signal delivered through the T cell antigen receptor and a second signal referred to as costimulation (1). T cells activated in vitro in the presence of CD28-B7 blockade display characteristics of anergy (2), and blockade in vivo has attenuated many T cell-mediated immune responses, including graft rejection (3, 4) and autoimnmnity (5-7). Analysis of CD28-deficient mice revealed relatively normal T cell development, preserved cytolytic T cell responses, but impaired B cell help (8). The role of costimulation in the generation and maintenance of antigen-specific CD4 + T helper (Th) subsets remains controversial. Data supports models in which both Thl and Th2 responses require CD28-mediated signals (2, 9-11), but some evidence suggests that only certain Th subsets are dependent on costimulation (12-20). Other findings suggest discrete stages of maturation selectively require CD28-mediated costimulation (12), and some current models propose that specific B7 ligands mediate biased Th maturation (5,6,21,22).Daniel R. Brown and Jonathan M. Green contributed equally to this work.In murine infection with Leishmania major, pathogenspecific CD4 + Th effectors mediate the polarized potential outcomes of infection (23). Thl responses that arise in C57BL/6 mice result in self-limited heating disease through successful macrophage activation by T cell-derived IFN-y. Th2 responses that arise in mice from the BALB background mediate inexorable susceptibility characterized by uncontrolled parasite dissemination. The unique localization of the parasite to MHC class II-rich intracellular compartments leads to a strong bias in CD4 +, clas...

Saccharomyces cerevisiae contains an irregular telomere sequence (TG 1-3 ) n , which differs from the regular repeat (TTAGGG) n found at the telomeres of higher organisms including humans. We have modified the entire 16-nt template region of the S. cerevisiae telomerase RNA gene (TLC1) to produce (TTAGGG) n repeats, the human telomere sequence. Haploid yeast strains with the tlc1-human allele are viable with no growth retardation and express the humanized gene at a level comparable to wild type. Southern hybridization demonstrates that (TTAGGG) n repeats are added onto the yeast chromosome ends in haploid strains with the tlc1-human allele, and sequencing of rescued yeast artificial chromosome ends has verified the addition of human telomeric repeats at the molecular level. These data suggest that the irregularity of the yeast telomere sequence is because of the template sequence of the yeast telomerase RNA. Haploid strains with the tlc1-human allele will provide an important tool for studying the function of telomerase and its regulation by telomere-binding proteins, and these strains will serve as good hosts for human artificial chromosome assembly and propagation.