Elevated and sustained expression of the transcription factors Egr1 and Egr2 controls NKT lineage differentiation in response to TCR signaling.

Interactions driven by the T cell antigen receptor (TCR) determine the lineage fate of CD4(+)CD8(+) thymocytes, but the molecular mechanisms that induce the lineage-determining transcription factors are unknown. Here we found that TCR-induced transcription factors Egr2 and Egr1 had higher and more-prolonged expression in precursors of the natural killer T (NKT) than in cells of conventional lineages. Chromatin immunoprecipitation followed by deep sequencing showed that Egr2 directly bound and activated the promoter of Zbtb16, which encodes the NKT lineage-specific transcription factor PLZF. Egr2 also bound the promoter of Il2rb, which encodes the interleukin 2 (IL-2) receptor ß-chain, and controlled the responsiveness to IL-15, which signals the terminal differentiation of the NKT lineage. Thus, we propose that persistent higher expression of Egr2 specifies the early and late stages of NKT lineage differentiation, providing a discriminating mechanism that enables TCR signaling to 'instruct' a thymic lineage.

BTB-ZF factors recruit the E3 ligase cullin 3 to regulate lymphoid effector programs.

The differentiation of several T- and B-cell effector programs in the immune system is directed by signature transcription factors that induce rapid epigenetic remodelling. Here we report that promyelocytic leukaemia zinc finger (PLZF), the BTB-zinc finger (BTB-ZF) transcription factor directing the innate-like effector program of natural killer T-cell thymocytes, is prominently associated with cullin 3 (CUL3), an E3 ubiquitin ligase previously shown to use BTB domain-containing proteins as adaptors for substrate binding. PLZF transports CUL3 to the nucleus, where the two proteins are associated within a chromatin-modifying complex. Furthermore, PLZF expression results in selective ubiquitination changes of several components of this complex. CUL3 was also found associated with the BTB-ZF transcription factor BCL6, which directs the germinal-centre B cell and follicular T-helper cell programs. Conditional CUL3 deletion in mice demonstrated an essential role for CUL3 in the development of PLZF- and BCL6-dependent lineages. We conclude that distinct lineage-specific BTB-ZF transcription factors recruit CUL3 to alter the ubiquitination pattern of their associated chromatin-modifying complex. We propose that this new function is essential to direct the differentiation of several T- and B-cell effector programs, and may also be involved in the oncogenic role of PLZF and BCL6 in leukaemias and lymphomas.

Immune response to helper dependent adenoviral mediated liver gene therapy: challenges and prospects.

Adenovirus-mediated gene therapy holds significant potential especially for applications requiring high levels of target tissue transduction. While significant advances in clinical adenoviral gene therapy applications have been made in cancer, the clinical translation of adenoviral gene replacement therapy for genetic disease has lagged. Encouragingly, advances in vector production have led to the development of Helper-Dependent ("gutted" or "high capacity") adenoviral vectors (HDV) deleted of all viral coding genes. HDV significantly reduces the chronic toxicity associated with early generation adenoviral vectors that has been most significant after systemic administration in both small and large animal models. However, the field remains confounded by innate immune responses inherent to adenovirus, and more generally, to the adaptive immune response to transgene. Together they decrease the effective therapeutic index for any particular treatment. This review summarizes the current advances toward understanding the decisive cell and molecular mechanisms underlying the acute toxicity to systemic HDV administration. We focus on the complex immune response and consequences of systemic vector delivery in the context of liver-directed monogenic disease therapy. Future development of interventions to avoid the innate immune response, including vector and pharmacologic manipulations, should further contribute to minimizing vector toxicity while maximizing the efficacy of systemic HDV gene transfer.

Adeno-associated virus types 5 and 6 use distinct receptors for cell entry.

The transduction efficiency of adeno-associated virus (AAV) vectors in various somatic tissues is determined primarily by the viral capsid proteins. In contrast to vectors made with AAV type 2 capsids, those having type 5 or 6 capsids show high transduction rates in airway epithelial cells, in a range that should be sufficient for treating lung disease. Here we have compared the properties of vectors made with AAV5 or AAV6 capsid proteins to determine whether their receptor usage is similar, and found several differences between the viruses. First, an AAV6 vector did not hemagglutinate red blood cells, whereas an AAV5 vector did, and this property was sialic acid dependent. Second, AAV5 vector transduction required sialic acid in all cells tested, whereas AAV6 vector transduction was sialic acid dependent or independent, depending on the target cells tested. Third, levels of an AAV6 vector that interfered with entry of another AAV6 vector only poorly inhibited AAV5 vector transduction and vice versa. These results indicate that AAV5 and AAV6 vectors use distinct cellular receptors for cell entry. Although both AAV5 and AAV6 vectors exhibited high transduction rates in well-differentiated human airway epithelial cultures, they exhibited distinct cell-type transduction profiles in mouse lung that may reflect differences in receptor usage.

Correction of murine hemophilia A and immunological differences of factor VIII variants delivered by helper-dependent adenoviral vectors.

Bioengineering of the factor VIII (FVIII) molecule has led to the production of variants that overcome poor secretion and/or rapid inactivation. We tested six modified FVIII variants for in vivo efficacy by expressing them from helper-dependent adenoviral (HD-Ad) vectors. We constructed a wild-type (WT) variant, a B-domain-deleted (BDD) variant, a point mutant for improved secretion (F309S), a variant with a partial B-domain deletion for improved secretion (N6), a combination of the point mutant and partial BDD variant (F309N6), and an inactivation-resistant (IR8) FVIII variant. All the constructs expressed functional protein after injection of high-dose HD-Ad. Activity ranged from 20 to 50% with WT, to approximately 100% with the N6 and F309N6 variants. Interestingly, mice treated with N6 showed long-term FVIII activity and phenotypic correction for up to 74 weeks, with low anti-FVIII antibody titer. Importantly, the N6 variant was therapeutically efficacious even after a 50% reduction of viral dose, thereby indicating that transgene modification itself can improve the dose efficacy of HD-Ad. This finding is significant, because dose efficacy is a key factor in clinical application. In summary, bioengineering of the FVIII molecule may be an effective approach to improving the safety, immunogenicity, and efficacy of HD-Ad gene therapy in hemophilia A (HA).

Toll-like receptor 9 triggers an innate immune response to helper-dependent adenoviral vectors.

A major obstacle to the clinical application of systemic adenoviral gene replacement therapy is the host innate immune response. Although recent studies have attempted to characterize the cellular basis for this response to systemically administered helper-dependent adenoviral vector (HD-Ad), the underlying molecular components of the innate immune repertoire required to recognize the viral vector have yet to be identified. Here, we show that primary macrophages can sense HD-Ad vectors via the Toll-like Receptor 9 (TLR9) and respond by increasing pro-inflammatory cytokine secretion. Moreover, TLR9 sensing is involved in the rapid innate immune response to HD-Ad in vivo. TLR9 deficiency attenuates the innate immune response to HD-Ad, whereas TLR9 blockade reduces the acute inflammatory response after intravenous injection of the vector. Moreover, HD-Ad upregulates TLR9 gene expression independent of TLR9 function, suggesting that additional innate signaling pathways work cooperatively with TLR9. The identification of the components of the innate immune response to adenovirus will facilitate the development of combinatorial therapy directed at increasing the maximal tolerated dose of systemically delivered adenoviral vectors.

Dendritic cell function after gene transfer with adenovirus-calcium phosphate co-precipitates.

Dendritic cells (DCs) are essential for initiating and directing antigen-specific T-cell responses. Genetic modification of DC is under study for cancer immunotherapy, vaccine development, and antigen-targeted immunosuppression. Adenovirus (Ad) type 5 (Ad5)-mediated gene transfer to mouse bone marrow DCs and human monocyte-derived DCs is inefficient because neither express the cognate high-affinity Ads receptor. We show that co-precipitating adenoviral vectors with calcium phosphate (CaPi) increased gene expression (2000-fold) and transduction efficiency (50-fold) in mouse DC, primarily owing to receptor-independent viral uptake. Moreover, Ad5:CaPi-treated DCs were activated to express the maturation surface molecules CD40 and CD86, and to secrete proinflammatory cytokines tumor necrosis factor-alpha and interleukin 6. However, neither DC transduction nor maturation was dependent on viral protein interactions with cell surface integrin. Ad5:CaPi also transduced human DC more efficiently than Ad5 alone, similar to a genetically modified vector (Ad5f35) targeted to the CD46 receptor. As such, this approach combines the efficiency of adenoviral-mediated endosomal escape and nuclear trafficking with the receptor independence of nonviral gene delivery. Importantly, CaPi co-precipitation could be used to functionally modify DC to activate and expand cytomegalovirus-specific memory cytotoxic T lymphocytes. This study identifies a simple technique to improve the efficacy of current Ad5 gene transfer, in support of clinical adoptive immunotherapy.

Thixotropic solutions enhance viral-mediated gene transfer to airway epithelia.

Adenovirus-mediated gene transfer to airway epithelia is inefficient in part because its receptor is absent on the apical surface of the airways. Targeting adenovirus to other receptors, increasing the viral concentration, and even prolonging the incubation time with adenovirus vectors can partially overcome the lack of receptors and facilitate gene transfer. Unfortunately, mucociliary clearance would prevent prolonged incubation time in vivo. Thixotropic solutions (TS) are gels that upon a vigorous shearing force reversibly become liquid. We hypothesized that formulating recombinant adenoviruses in TS would decrease virus clearance and thus enhance gene transfer to the airway epithelia. We found that clearance of virus-sized fluorescent beads by human airway epithelia in vitro and by monkey trachea in vivo were markedly decreased when the beads were formulated in TS compared with phosphate-buffered saline (PBS). Adenovirus formulated in TS significantly increased adenovirus-mediated gene transfer of a reporter gene in human airway epithelia in vitro and in murine airway epithelia in vivo. Furthermore, an adenovirus encoding the cystic fibrosis transmembrane regulator (CFTR) gene (AdCFTR) formulated in TS was more efficient in correcting the chloride transport defect in cystic fibrosis airway epithelia than AdCFTR formulated in PBS. These data indicate a novel strategy to augment the efficiency of gene transfer to the airways that may be applicable to a number of different gene transfer vectors and could be of value in gene transfer to cystic fibrosis (CF) airway epithelia in vivo.

Foxj1 is required for apical localization of ezrin in airway epithelial cells.

Establishment and maintenance of epithelial cell polarity depend on cytoskeletal organization and protein trafficking to polarized cortical membranes. ERM (ezrin, radixin, moesin) family members link polarized proteins with cytoskeletal actin. Although ERMs are often considered to be functionally similar, we found that, in airway epithelial cells, apical localization of ERMs depend on cell differentiation and is independently regulated. Moesin was present in the apical membrane of all undifferentiated epithelial cells. However, in differentiated cells, ezrin and moesin were selectively localized to apical membranes of ciliated airway cells and were absent from secretory cells. To identify regulatory proteins required for selective ERM trafficking, we evaluated airway epithelial cells lacking Foxj1, an F-box factor that directs programs required for cilia formation at the apical membrane. Interestingly, Foxj1 expression was also required for localization of apical ezrin, but not moesin. Additionally, membrane-cytoskeletal and threonine-phosphorylated ezrin were decreased in Foxj1-null cells, consistent with absent apical ezrin. Although apical moesin expression was present in null cells, it could not compensate for ezrin because ERM-associated EBP50 and the beta2 adrenergic receptor failed to localize apically in the absence of Foxj1. These findings indicate that Foxj1 regulates ERM proteins differentially to selectively direct the apical localization of ezrin for the organization of multi-protein complexes in apical membranes of airway epithelial cells.