Zoledronate-activated Vγ9γδ T cell-based immunotherapy is feasible and restores the impairment of γδ T cells in patients with solid tumors.

Gamma/delta (γδ) T cells play a role in innate immunity and exhibit cytotoxicity toward a large range of tumor types. Recent studies have shown that aminobisphosphonates may be applied to a culture in which a large number of γδ T cells are proliferated ex vivo. We carried out a clinical study of 25 patients with various solid tumors to determine further the safety, immunologic effect and feasibility of zoledronate-activated Vγ9γδ T cell-based immunotherapy. No severe toxicity was observed. In the cells used for the first treatment, the total cell number, frequency and number of CD3(+) Vγ9(+) γδ T cells were 409 ± 284 × 10(7) cells, 56 ± 33% and 255 ± 242 × 10(7) cells, respectively. Aminobisphosphonate therapy or chemotherapy resulted in the suppression of CD3(+) Vγ9(+) γδ T-cell proliferation. The numbers of CD3(+) T cells, CD3(+) Vγ9(+) γδ T cells and CD27(-) CD45RA(-) Vγ9(+) subsets in peripheral blood were significantly lower in patients than in healthy subjects (P < 0.05). From such an impaired immunologic condition, the numbers and frequencies of CD3(+) Vγ9(+) γδ T cells and CD27(-) CD45RA(-) subsets significantly increased in patients treated with this immunotherapy. Zoledronate-activated Vγ9γδ T cell-based immunotherapy that restores the number of Vγ9γδ T cells in cancer patients may provide another mode of adoptive immunotherapy.

High-pressure-induced hemolysis in papain-digested human erythrocytes is suppressed by cross-linking of band 3 via anti-band 3 antibodies.

Upon exposure of human erythrocytes to a high pressure of 200 mPa, both hemolysis and vesiculation occur. The hemolysis of erythrocytes at 200 mPa was enhanced by removal of sialic acids from the membrane surface with papain. However, such enhancement was suppressed by cross-linking of band 3 via an anti-band 3 antibody (AB3A), which recognizes the exofacial domain of band 3, or by clustering of band 3 via Zn2+. On the other hand, the size of high-pressure-induced vesicles increased from 423 to 525 nm in diameter upon exposure to papain of erythrocytes, but decreased to 444 nm with following treatment with AB3A. In these vesicles, the content of spectrin relative to band 3 was almost the same. Furthermore, the band 3-cytoskeleton interactions in erythrocyte membranes remained unaltered upon treatment with papain and AB3A. Flow cytometric analysis demonstrated that papain-pretreated erythrocytes mainly produce open ghosts at 200 mPa and that the production of such open ghosts is suppressed by AB3A. Thus, upon removal of negative charges from the membrane surface, open ghosts are readily produced due to the release of larger vesicles under pressure. Upon cross-linking of band 3 via AB3A, however, the release of smaller vesicles at 200 mPa is facilitated so that high-pressure-induced hemolysis is suppressed.

Topogenesis of NHE1: direct insertion of the membrane loop and sequestration of cryptic glycosylation and processing sites just after TM9.

Multispanning membrane proteins are synthesized by membrane-bound ribosomes and integrated into the endoplasmic reticulum membrane cotranslationally. To uncover the topogenic process of membrane loop, of which both ends are in the same side of the membrane, we examined topogenesis of a relatively hydrophobic lumenal loop segment (H10 segment) between TM9 and TM10 of human Na(+)/H(+) exchanger isoform 1 using an in vitro expression system. The H10 segment was translocated through the membrane. Any potential sites created within the H10 segment were not glycosylated. Just after TM9, there are potential glycosylation and signal peptidase processing sites. When the reporter domain of prolactin was fused at the position preceding the H10 segment, these sites were modified by the enzymes, while they were not modified in the original molecule. Thus, we concluded that the H10 segment was translocated through the membrane and directly inserted into the membrane and that its membrane insertion caused sequestration of the preceding processing and glycosylation sites from the lumenal modifying enzymes. This topogenic process shows clear contrast to that of pore loops of K(+) channels, which are once exposed in the lumen and accessible to glycosylation enzyme.

Membrane topogenesis of the three amino-terminal transmembrane segments of glucose-6-phosphatase on endoplasmic reticulum.

We investigated the membrane topogenesis of glucose-6-phosphatase (G6Pase), a multispanning membrane protein, on the endoplasmic reticulum. In COS-7 cells, the first transmembrane segment (TM1) with weak hydrophobicity is inserted into the membrane in the N-terminus-out/C-terminus-cytoplasm orientation. The following TM2 is inserted depending on TM3. TM3 has the same orientation as TM1. In contrast to data from living cells, the full-length molecule and N-terminal fusion constructs were not inserted into the membrane in a cell-free system. Addition of a signal recognition particle did not improve G6Pase insertion. When the 37-residue N-terminal segment was deleted, however, TM2 and TM3 were correctly inserted. We concluded that the three N-terminal TM segments are inserted into the membrane dependent on the two signal-anchor sequences of TM1 and TM3. TM1 is likely to be an unconventional signal sequence that barely functions in vitro. The 37-residue N-terminal segment inhibits the signal function of the following TM3 in cell-free systems.