Bottom-Up Copper Filling of Millimeter Size Through Silicon Vias.

This work demonstrates void-free Cu filling of millimeter size Through Silicon Vias (mm-TSV) in an acid copper sulfate electrolyte using a combination of a poloxamine suppressor and chloride, analogous to previous work filling TSV that were an order of magnitude smaller in size. For high chloride concentration (i.e., 1 mmol/L) bottom-up deposition is demonstrated with the growth front being convex in shape. Instabilities in filling profile arise as the growth front approaches the free-surface due to coupling with electrolyte non-uniform hydrodynamics. The reentrant notches at the bottom of the TSVs caused by intentional over-etching during fabrication negatively impact the filling results. In contrast, deposition from low chloride electrolytes (i.e., 80 µmol/L) proceeds with a passive-active transition on the via sidewalls. For a given applied potential the location of the transition is fixed in time and the growth front is concave in nature reflecting the gradient in chloride surface coverage. Application of a suitable potential wave form enables the location of the sidewall transition to be systematically advanced thereby giving rise to void-free filling of the TSV.

Bottom-Up Copper Filling of Large Scale Through Silicon Vias for MEMS Technology.

An electrodeposition process for void-free bottom-up filling of sub-millimeter scale through silicon vias (TSVs) with Cu is detailed. The 600 µm deep and nominally 125 µm diameter metallized vias were filled with Cu in less than 7 hours under potentiostatic control. The electrolyte is comprised of 1.25 mol/L CuSO4 -0.25 mol/L CH3SO3H with polyether and halide additions that selectively suppress metal deposition on the free surface and side walls. A brief qualitative discussion of the procedures used to identify and optimize the bottom-up void-free feature filling is presented.

Design, microfabrication, and analysis of micrometer-sized cylindrical ion trap arrays.

A description of the design and microfabrication of arrays of micrometer-scale cylindrical ion traps is offered. Electrical characterization and initial ion trapping experiments with a massively parallel array of 5 microm internal radius (r(0)) sized cylindrical ion traps (CITs) are also described. The ion trap, materials, and design are presented and shown to be critical in achieving minimal trapping potential while maintaining minimal power consumption. The ion traps, fabricated with metal electrodes, have inner radii of 1, 2, 5, and 10 microm and range from 5 to 24 microm in height. The electrical characteristics of packaged ion trap arrays were measured with a vector network analyzer. The testing focused on trapping toluene (C(7)H(8)), mass 91, 92, or 93 amu, in the 5 microm sized CITs. Ions were formed via electron impact ionization and were ejected by turning off the rf voltage applied to the ring electrode; a current signal was collected at this time. Optimum ionization and trapping conditions, such as a sufficient pseudopotential well and high ionization to ion loss rate ratio (as determined by simulation), proved to be difficult to establish due to the high device capacitance and the presence of exposed dielectric material in the trapping region. However, evidence was obtained suggesting the trapping of ions in 1%-15% of the traps in the array. These first tests on micrometer-scale CITs indicated the necessary materials and device design modifications for realizing ultrasmall and low power ion traps.

Interferon-gamma modulates human oligodendrocyte susceptibility to Fas-mediated apoptosis.

Interferon gamma (IFN-gamma) has been shown to be produced within multiple sclerosis (MS) lesions by infiltrating lymphocytes; systemic administration of this cytokine induces exacerbation of the disease. The aim of the current study was to establish the contribution of IFN-gamma to oligodendrocyte (OL) injury. Our studies utilized cultured human OLs, obtained by dissociation of surgically derived non-MS adult brain tissue. Neither cell survival nor myelin basic protein (MBP) gene expression were affected after 96 hours of treatment with IFN-gamma (100 U/ml), as assessed by LDH release, nucleosome enrichment assay, and RT-PCR. Expression of the death receptor Fas (CD95, APO-1) was, however, significantly increased. Furthermore, IFN-gamma-treated OLs became susceptible to Fas-mediated apoptosis when compared with untreated cells, and were protected by pretreatment with the caspase inhibitor ZVAD. TNF-alpha augmented the IFN-gamma-induced effect. Our results thus indicate that IFN-gamma is not directly cytotoxic for human OLs in culture, but could indirectly modulate functional injury-related responses by upregulating Fas on the cell surface.

Regulation of Th1 and Th2 lymphocyte migration by human adult brain endothelial cells.

Endothelial cells of the blood-brain barrier (BBB) have the ability to regulate and restrict the passage of cells and molecules from the periphery to the CNS. We have used an in vitro assay of lymphocyte migration across monolayers of human adult brain endothelial cells (HBEC) as a model of lymphocyte migration across the BBB. We found that human allogeneic or MBP-reactive Th2-polarized lymphocytes migrate more avidly than Th1-polarized lymphocytes. Migration of Th2 but not Th1 cells across brain endothelium was inhibited by antibodies directed at MCP-1, a chemokine produced by HBECs. We could detect CCR2, a chemokine receptor that recognizes MCP-1 on Th2 but not Th1 lymphocytes. ICAM-1 and VCAM-1 molecules were expressed on the surface of HBECs under basal conditions and were upregulated by Th1 but not Th2 cell-derived supernatants. Migration of both lymphocyte subsets was dependent on LFA-1/ICAM-1 interactions. Blocking VLA-4/VCAM-1 binding did not influence actual trans-endothelial migration. These results suggest that HBECs composing the BBB favor the migration of Th2 cells. We postulate that this selectivity may help prevent activated Th1 lymphocytes, the putative CNS autoimmune disease initiating cells, from reaching the CNS parenchyma and favor entry of Th2 cells, a putative means to induce bystander suppression in the CNS.

Benzodiazepine receptor binding is not altered in human epileptogenic cortical foci.

We measured benzodiazepine binding in cortical tissues from 13 patients with partial seizures. No differences in receptor density (Bmax) or affinity (Kd) were observed when epileptic foci were compared with nonspiking cortex.

CD40 engagement stimulates IL-12 p70 production by human microglial cells: basis for Th1 polarization in the CNS.

The APC-derived cytokine interleukin (IL)-12 polarizes CD4 T cells towards the pro-inflammatory Th1 phenotype and has been shown to be crucial for the development of EAE in rodents. In this study we demonstrate that production of IL-12 by human adult CNS-derived microglial cells can be triggered by cell contact with activated T cells. Microglial activation and IL-12 production can be blocked by anti-CD154 mAbs. IL-12 production could also be induced by direct engagement of CD40 on microglia using a CD40 agonist. IL-12 secretion by microglia is significantly reduced by TNF and IFN-gamma antagonists showing that the IL-12 production is subject to regulation by auto- and paracrine stimuli.

Activated suppressor cell function in multiple sclerosis--clinical correlations.

Activated suppressor cell function mediated by either freshly isolated peripheral blood mononuclear cells (MNCs), freshly isolated CD8+ lymphocytes or by CD8+ cell lines, has previously been found to be reduced compared to controls in multiple sclerosis (MS) patients with progressive disease (MS-P). In this study, we found that suppressor activity mediated by CD8+ cell lines, derived from MS patients with stable disease (MS-S) patients and maintained in culture for 14 days, was significantly greater (45 +/- 6%) compared to that mediated by MS-P patients' CD8+ cells (11 +/- 4%, P less than 0.005). The MS-S suppressor values were, however, suggestively reduced compared to controls (60 +/- 6%, P less than 0.05). MNC-mediated suppressor values for the MS-S group (61 +/- 5%) did not differ from the control group (67 +/- 6%). Values for the MS-P group (7 +/- 6%) were significantly reduced compared to MS-S and control groups. Cytotoxic activity mediated by CD8+ cell lines showing defective suppressor function did not differ from control values. The cell lines in MS and control did not differ with respect to their rate of proliferation in the presence of IL-2 and OKT3. Suppressor function in this assay was ablated if exogenous IL-2 was removed from the culture media. These data suggest that defective activated suppressor function is characteristic of the progressive form of MS, although a suppressor defect is also partially expressed in stable MS patients when CD8+ cell lines are studied.

Cytochemical analysis of the reconstitution of endoplasmic reticulum after microinjection of rat liver microsomes into Xenopus oocytes.

Fragments of rough and smooth endoplasmic reticulum purified from rat liver were injected into Xenopus oocyte cytoplasm. Light and electron microscopy, cytochemistry, immunocytochemistry, and enzyme assay were employed to determine the fate of heterologous membranes in the host cytoplasm. The in vivo-incubated microsomes disappeared in a time-dependent manner. Within 3 hr, rough microsomes were replaced by flattened ER cisternae and smooth microsomes were replaced by a network of anastomosing tubules. Polyclonal antibodies against rat liver microsomes and protein A-gold complexes were applied to glycol methacrylate sections of microinjected oocytes. Specific labeling was observed over discrete rough and smooth ER cisternae 3 hr after microinjection. Endogenous ER was not labeled by this technique, and label was not observed when sections were treated with pre-immune antibodies. Diaminobenzidene cytochemistry of microinjected rat lacrimal gland microsomes revealed enzyme activity in heterologous microsomes after 3 hr of in vivo incubation. Control injected microsomes (inactivated by heat denaturation) became associated with autophagic vacuoles, coincident with changes in lysosomal activity. Freshly isolated un-denatured microsomes did not provoke changes in lysosomal activity, and glucose-6-phosphatase activity associated with microinjected membranes could be detected 21 hr after in vivo incubation. Since rat liver microsomes reconstitute after in vivo incubation into cytoplasmic structures resembling those from which they were derived, we conclude that the microinjected membrane fragments act as templates for their own three-dimensional organization.

Effect of tumor necrosis factor alpha and beta on human oligodendrocytes and neurons in culture.

Cytokines produced by infiltrating hematogenous cells or by glial cells activated during the course of central nervous system disease or trauma are implicated as mediators of tissue injury. In this study, we have assessed the extent and mechanism of injury of human-derived CNS oligodendrocytes and neurons in vitro mediated by the cytokines tumor necrosis factor alpha and beta and compared these with the tumor necrosis factor independent effects mediated by activated CD4+ T-cells. We found that activated CD4+ T-cells, but not tumor necrosis factor alpha or beta, could induce significant release of lactate dehydrogenase, a measure of cell membrane lysis, from oligodendrocytes within 24 hr. Neither induced DNA fragmentation as measured using a fluorescence nick-end labelling technique. After a more prolonged time period (96 hr), tumor necrosis factor alpha did induce nuclear fragmentation changes in a significant proportion of oligodendrocytes without increased lactate dehydrogenase release. The extent of DNA fragmentation was comparable to that induced by serum deprivation. Tumor necrosis factor beta effects were even more pronounced. In contrast to oligodendrocytes, the extent of DNA fragmentation, assessed by propidium iodide staining, induced in neurons by tumor necrosis factor alpha was less than that induced by serum deprivation. In-situ hybridization studies of human adult glial cells in culture indicated that astrocytes, as well as microglia, can express tumor necrosis factor alpha mRNA.

The serum IL-12:IL-6 ratio reliably distinguishes infectious from non-infectious causes of fever during autologous stem cell transplantation.

BACKGROUND: Fever during neutropenia and after neutrophil engraftment (post-engraftment fever) occurs commonly during autologous transplantation (ASCT), but infections are infrequently identified. Tests that reliably exclude infection may reduce the cost and toxicity of unnecessary diagnostic testing and empiric treatment. We assessed whether serum levels of inflammatory cytokines could distinguish infectious from non-infectious causes of fever in patients undergoing ASCT. METHODS: Serum levels of IL-1beta, IL-2, IL-6, IL-8, IL-10, IL-12(p70), TNF-alpha and IFN-gamma were measured by sandwich ELISA at multiple pre-determined times and at the onset of the first fever during neutropenia and after neutrophil engraftment in patients with hematologic malignancies undergoing ASCT. Standard clinical criteria were used to assess for the presence of infection. RESULTS: Seventy-two febrile episodes occurred in 54 of 65 enrolled patients; 29 (40%) of the episodes occurred after neutrophil engraftment. Infections were identified as the cause of 28% and 24% of the neutropenic and post-engraftment febrile episodes, respectively. The level of IL-12 decreased and that of IL-6 increased significantly during fever because of infection, such that the IL-12:IL-6 ratio accurately excluded infection. The area under the ROC curve for the IL-12:IL-6 ratio was 0.88 (95% CI 0.79-0.97). The sensitivity, specificity, positive predictive and negative predictive values associated with a cut-off ratio of 4.1 were 95%, 75%, 60%, and 97%, respectively. DISCUSSION: The IL-12:IL-6 ratio effectively discriminates infectious from non-infectious causes of fever during ASCT. It may be useful in assessing the probability of infection in patients with post-engraftment fever.

Expression of a homologue of rat NG2 on human microglia.

The expression of NG2 chondroitin sulfate has been widely associated with oligodendrocyte precursors in rodents. We used a monoclonal antibody (9.2.27) against the human homologue of the rat NG2 to determine whether expression of this molecule was associated with a specific glial cell population present in dissociated cell preparations derived from adult and fetal human brain tissue. Our data, derived using FACS and immunocytochemical analyses of immediately ex vivo or cultured glial cells, indicate that the large majority of NG2 expressing cells belonged to the microglial lineage (CD68, CD11c) rather than to the oligodendrocyte lineage (O4, A2B5, GalC). In situ immunohistochemistry performed on non-fixed normal spinal cord tissue confirmed the observation that NG2 is expressed by mononuclear phagocytes of the CNS. In contrast, peripheral blood-derived monocytes were NG2(-). Cells from fetal brain tissue showed only small numbers of NG2(+) cells, which was consistent with the number of microglial cells in this preparation. In absence of additional markers, we cannot exclude that this anti-NG2 mAb might also recognize human oligodendrocyte progenitor cells.

Differential responses of CD4+CD45RA+ and CD4+CD29+ subsets to activated CD8+ cells: enhanced stimulation of the CD4+CD45RA+ subset by cells from patients with multiple sclerosis.

To examine whether functionally different CD4+ cells respond uniformly to the immunoregulatory influences of allogeneic activated CD8+ cells (*CD8+), we subfractionated the CD4+ population into two subsets, based on the high expression of either CD45RA or CD29. We confirmed that the CD45RA+ cells proliferated poorly in response to soluble anti-CD3 mAb, compared to the vigorous response obtained with the CD29+ subset; the CD45RA+ cells were more responsive to stimulation with Con A. Using normal healthy controls, we found that whereas *CD8+ had a significant suppressive effect on the proliferation of the CD29+ subset, they augmented the mitogen-induced proliferative response of the CD45RA+ cells. We further demonstrated that *CD8+ derived from MS patients augmented the response of the CD45RA+ subset to a significantly higher degree compared to healthy age- and sex-matched controls. There were no significant differences between the degree of suppression exerted by the *CD8+ of either the MS or the control group on the CD29+ cells. These results demonstrate that helper/memory CD4+CD29+ cells are more sensitive to the suppressive influences of *CD8+ compared to the CD4+CD45RA+ subset. In addition, in MS, *CD8+ may contribute to a more pronounced "on" signal for virgin CD4+CD45RA+ cells, which might serve as a means to perpetuate the autoimmune disease process.

Oligodendrocyte lysis by CD4+ T cells independent of tumor necrosis factor.

The capacity of human CD4+ T cells to lyse heterologous human oligodendrocytes in an 18-hour chromium 51-release assay was compared to that of systemic blood-derived macrophages and central nervous system-derived microglia. CD4+ T cells, activated with either phytohemagglutinin, anti-CD3 antibody, or antigen (myelin basic protein), could induce lysis of the oligodendrocytes whereas macrophages and microglia, activated with interferon-gamma and lipopolysaccharide, could not. The CD4+ T-cell effect was not inhibited with an anti-tumor necrosis factor-alpha-neutralizing antibody. Both the CD4+ T cells and the macrophages could induce lysis of tumor necrosis factor-sensitive rodent cell lines, Wehi 164, and L929; these effects were inhibited with anti-tumor necrosis factor antibody. Pretreatment of the CD4+ T cells with cyclosporine or mitomycin C did not inhibit oligodendrocyte lysis. These results indicate that at least in vitro, CD4+ T cells can induce a form of oligodendrocyte injury that is not reproduced by macrophages or microglia or by tumor necrosis factor. The non-major histocompatibility complex (MHC)-restricted injury of oligodendrocytes induced by both myelin antigen-reactive and mitogen-stimulated T cells may provide a basis whereby cytotoxic CD4+ T cells could interact with a target cell that does not express MHC class II molecules. Our results suggest that immune-mediated oligodendrocyte/myelin injury, as is postulated to occur in the disease multiple sclerosis, may involve multiple effector mechanisms.

Induction of primary T cell responses by human glial cells.

Glial cells of the central nervous system (CNS) are postulated to function as immune accessory cells which may regulate immune reactivity occurring within the CNS, activating or alternatively inhibiting T cell responses. We have utilized surgically resected cerebral tissue derived from young adult humans to prepare dissociated cultures of glial cells (mixed astrocyte-microglia-oligodendrocyte cultures) and demonstrate that such cells are capable of acting as stimulators of primary T cell responses, using proliferation of T cells to allogeneic determinants on the glial cells as the test system. Studies of resected adult cerebral tissue indicated major histocompatibility complex (MHC) class II antigen expression on microglia in situ. Using a mixed lymphocyte reaction (MLR), we observed that enriched microglial cultures alone were capable of stimulating primary responses of freshly isolated T cells or the CD4+ T cell subset, a response which could be inhibited with an anti-MHC class II blocking antibody. In agreement with previous studies using rodent-derived astrocytes, we found that human astrocytes (fetal), could not initiate a primary T cell response even after up-regulation of MHC class II antigen expression with interferon gamma (IFN gamma) and tumor necrosis factor alpha (TNF alpha). Our results indicate that a primary T cell response, as well as a secondary response to a recall antigen, can occur within the CNS; our data implicate microglia as the central cell involved in the former.

[Arterial hypertension due to abuse of sympathomimetic drugs. One case (author's transl)]. / Hypertension artérielle par abus de sympathicomimétiques. Une observation.

A 36-year-old woman developed severe arterial hypertension after taking for five consecutive years increasing dose (up to 10 mg per day) of phenoxazoline HCl in nasal spray. A relationship between the abuse of this sympathomimetic drug and the hypertension was suggested by the unusual appearance of renal arteries on arteriography (stenosis and dilatations resembling aneurisms), the increase in renin activity and the disappearance of hypertension after the drug was discontinued. On control examination, two years later, blood pressure, renin activity and renal arteries were normal. The possibility of sympathomimetic drug overdosage must be borne in mind in cases of suspected iatrogenic arterial hypertension.

Inhibition of Th1 polarization by soluble TNF receptor is dependent on antigen-presenting cell-derived IL-12.

Th1-polarized CD4+ T cells are considered central to the development of a number of target-directed autoimmune disorders including multiple sclerosis. The APC-derived cytokine IL-12 is a potent inducer of Th1 polarization in T cells. Inhibition of IL-12 in vivo blocks the development of experimental allergic encephalomyelitis, the animal model for multiple sclerosis. Based on previous work that suggests that the production of IL-12 by activated human central nervous system-derived microglia is regulated by autocrine TNF-alpha, we wanted to determine whether inhibition of TNF could induce a reduction of Th1 responses by its impact on systemic APCs. We found that soluble TNFR p75-IgG fusion protein (TNFR:Fc) inhibited production of IFN-gamma by allo-Ag-activated blood-derived human CD4 T cells. We documented reduced IL-12 p70 production by APCs in the MLR. By adding back recombinant IL-12, we could rescue IFN-gamma production, indicating that TNFR:Fc acts on APC-derived IL-12. Consistent with an inhibition of the Th1 polarization, we found a decreased expression of IL-12R-beta2 subunit on the T cells. Furthermore, the capacity of T cells to secrete IFN-gamma upon restimulation when previously treated with TNFR:Fc is impaired, whereas IL-2 secretion was not altered. Our results define a TNF-dependent cytokine network that favors development of Th1 immune responses.

Phenotypic and functional characteristics of activated CD8+ cells: a CD11b-CD28- subset mediates noncytolytic functional suppression.

Freshly isolated human CD8+ cells can be divided into mutually exclusive subsets bearing the phenotypes CD11b+(CD28-) or CD28+(CD11b-). We found that activation of CD8+ cells with anti-CD3 mAb and IL-2 preferentially expanded the CD11b-(CD28+) subset. This subset, when separated and activated independently, mediated both functional suppression and lectin-dependent cell cytotoxicity (LDCC). CD28- cells, prepared by elimination of the CD 28+ cells from expanded unfractionated CD8+ cell cultures, retained functional suppressor activity but demonstrated reduced LDCC compared to either the CD28+(CD11b-)-enriched fraction or the unfractionated CD8+ population. The majority of the CD28- cells were also CD11b-, reflecting the observation that initially CD11b+ cells lose CD11b expression following activation with anti-CD3 mAb and IL-2. Our results therefore indicate that CD8+ cells deriving from the CD11b+CD28- subset, but expressing neither CD11b nor CD28 after activation, represent the main noncytotoxic functional suppressor cell in the mitogen "activated" suppressor assay. The preferential expansion of CD8+CD28+ cells relative to CD8+CD28- cells, if occurring in vivo in the central nervous system (CNS) compartment, would be consistent with observed phenotypic analysis of cerebrospinal fluid-derived T cells and might contribute to the reduced functional suppressor activity previously found for CNS compared to peripheral blood-derived lymphocytes.

NG2 immunoreactivity on human brain endothelial cells.

In this study, we evaluated the expression of NG2 on human brain endothelial cells derived from temporal lobe tissue resected as a treatment for intractable epilepsy. Using dissociated cell cultures, we found expression of NG2 on both proliferating and non-proliferating cells, at the mRNA level by reverse transcription-polymerase chain reaction analyses, and at the protein level by immunocytochemistry and immunoblotting. We further observed that human cerebral microvessels in nonmalignant CNS tissues immunoreacted with NG2. NG2 protein was detected using both a rabbit antibody raised against the rodent NG2 and a monoclonal antibody raised against the human NG2 (9.2.27). Our findings further define the range of resident cells of the CNS that can express NG2 and indicate that expression of NG2 by endothelial cells is not restricted to proliferating CNS endothelial cells or to endothelial cells found in brain tumors.