Search PubMed⌕ Search

Biomedical subjects

M Welling

Publications and source records attributed to M Welling.

13 recordsLinked to original sources
PubMed2001

A constrained EM algorithm for independent component analysis.

We introduce a novel way of performing independent component analysis using a constrained version of the expectation-maximization (EM) algorithm. The source distributions are modeled as D one-dimensional mixtures of gaussians. The observed data are modeled as linear mixtures of the sources with additive, isotropic noise. This generative model is fit to the data using constrained EM. The simpler "soft-switching" approach is introduced, which uses only one parameter to decide on the sub- or supergaussian nature of the sources. We explain how our approach relates to independent factor analysis.

Algorithms↗
PubMed1997

Detection of experimental infections with 99mTc-labeled monoclonal antibodies against TNF-alpha and interleukin-8.

This study was designed to assess monoclonal antibodies (MAbs) directed against tumor necrosis factor-alpha (TNF-alpha) (anti-TNF) or interleukin-8 (anti-IL-8) as radioactive agents for the detection of Staphylococcus aureus-or Klebsiella pneumoniae-infected thighs in mice. At 5 min (acute infection) or 20 h (established) post-infection, 20 micrograms of the 99mTc-labeled MAbs were injected. At various time intervals, the accumulation of the radiotracer in the infected thighs was assessed and expressed as a target-to-nontarget (T/NT) ratio. The binding of 99mTc-labeled MAbs to circulating mononuclear cells and granulocytes was quantitated 20 h after injection. The pharmacokinetics of the MAbs, in relation to the control agents 99mTc-labeled polyclonal human immunoglobulin (IgG) and a 99mTc-labeled nonspecific IgG1 MAb, were also studied. In acute infections, 99mTc-anti-TNF accumulated to a higher extent (p < 0.05) in S. aureus-infected thighs in mice until 4 h after the injection than 99mTc-IgG and was higher at 0.25 h in K. pneumoniae-infected mice (p < 0.03) compared with 99mTc-IgG. In established S. aureus and K. pneumoniae infections, 99mTc-anti-IL-8 detected the infection more intensely than 99mTc-IgG until 1 h after injection. In both S. aureus and K. pneumoniae infections, localization of sites of infection correlates (p < 0.05) with increased binding of the 99mTc-labeled MAbs to granulocytes and mononuclear cells in both acute and established infections. It was concluded that 99mTc-labeled MAbs, directed against TNF-alpha and IL-8, accumulate in bacterial infections in mice to a higher extent than does 99mTc-IgG after infection and is related to the binding of the antibodies to blood leukocytes. With these 99mTc-labeled MAbs, information might be gained about the development of an infection.

Animals↗
PubMed1997

Localization of a bacterial infection with 99Tcm-labelled human IgG: further improvement with enriched IgG subclass preparations.

The aim of this study was to determine the contribution of various IgG subclasses to the scintigraphic detection of a staphylococcal infection. An experimental thigh infection in mice was used to determine the accumulation of the various 99Tcm-labelled IgG preparations with enriched IgG1, IgG2 or IgG4 subclass. Multiple-regression analysis was used to investigate a relationship between the IgG subclasses and the time-dependent accumulation in infected sites. Eighteen hours after infection with Staphylococcus aureus bacteria 20 micrograms of 99Tcm-labelled IgG preparations enriched with one of the IgG1, IgG2 or IgG4 subclasses by thiophilic absorption were administered intravenously and target-to-nontarget (T/NT) ratios were determined at 15 min, 1 h, 4 h and 24 h after injection of the tracer. Moreover, the binding of these preparations to S. aureus was assessed using an in vitro bacterial pellet model as an indication for the potency of detecting infections. As a control agent, 99Tcm-labelled polyclonal IgG (HIG) was used. In vivo, the T/NT ratios were significantly (P < 0.05) higher for the IgG1-enriched preparation at all time points, and for the IgG2-enriched preparation at 4 h and 24 h after injection, compared with HIG. In contrast, IgG4 did not yield higher T/NT ratios at any time. Using multiple-regression analysis, it became evident that IgG3 at all time intervals, IgG1 for early scans (up to 4 h) and IgG2 for late scans (24 h) contribute significantly (P < 0.05) to the accumulation. The abundance of IgG subclasses in the various preparations appeared to influence the accumulation of tracer at infected sites. The percentage of binding to S. aureus in vitro was significantly (P < 0.05) higher for enriched IgG subclass preparations than for HIG. We conclude that specific subclass enrichment of 99Tcm-labelled IgG preparations improves the scintigraphic detection of staphylococcal infections at various time intervals post-injection.

Animals↗
PubMed1996

99Tcm-HIG accumulates in the synovial tissue of rats with adjuvant arthritis by binding to extracellular matrix proteins.

Our objective was to investigate the mechanism of accumulation of 99Tcm-labelled non-specific polyclonal human immunoglobulin (99Tcm-HIG) in inflamed synovial tissue (ST) in an experimental animal model of arthritis. Following 99Tcm-HIG scintigraphy, the in vivo localization of 99Tcm-HIG in the ST of knee joints of rats with adjuvant arthritis was studied using immunohistochemical techniques. In addition, the in vitro binding of 99Tcm-HIG to extracellular matrix proteins was analysed by means of immunohistochemistry and enzyme-linked immunosorbent assay (ELISA). After 99Tcm-HIG scintigraphy, 99Tcm-HI was detected in the ST of rats with adjuvant arthritis. 99Tcm-HIG was diffusely distributed and not bound to cells. In vitro incubation of 99Tcm-HIG on the ST of rats with adjuvant arthritis revealed binding of 99Tcm-HIG to inflamed, but not to non-inflamed, ST. In addition, specific binding of 99Tcm-HIG to fibronectin, fibrin, collagen type I and III was demonstrated by ELISA. We conclude that the accumulation of 99Tcm-HIG in inflamed ST can be explained by the binding of 99Tcm-HIG to extracellular matrix proteins.

Animals↗
PubMed1995

Contribution of phagocytic cells and bacteria to the accumulation of technetium-99m labelled polyclonal human immunoglobulin at sites of inflammation.

The purpose of this study was to assess the contribution of phagocytic cells and bacteria to the accumulation of technetium-99m labelled polyclonal human immunoglobulin (HIG) at sites of inflammation. Mice were intraperitoneally injected with Staphylococcus aureus (SA animals), with heat-inactivated newborn calf serum (NBCS, to mimic a non-bacterial inflammation) or with physiological saline (controls); 1 h thereafter they received HIG. At various intervals after the administration of HIG the mice were killed, and the percentages of radioactivity in the peritoneal effluent and attached to the cellular and bacterial fraction thereof were established. Furthermore, the total number of cells and that of bacteria in the fluid were quantitated. The percentage of activity in the effluent in the SA animals was (P < 0.02) higher than those in the NBCS-injected animals and controls from 4 h onwards. In all groups of mice this percentage was highest at 4 h and decreased (P < 0.01) afterwards. The percentage of cell-bound activity and the total number of cells remained fairly constant or increased with time in the SA animals (P < 0.01). The bacteria-bound activity remained rather constant throughout the experiment and ranged between 4% and 6%. In the SA-infected animals the percentage of cell-bound activity was correlated with the total number of cells (macrophages but especially neutrophils) but even more strongly with the number of cell-associated bacteria. In the NBCS-injected animals a correlation was demonstrated between the cell-bound activity and the total number of cells (only neutrophils).(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗
PubMed1995

A new 99mTc labelling method for leucocytes: in vitro and in vivo comparison with 99mTc-HMPAO.

A new method for the labelling of mixed leucocytes with 99mTc-tropolone was optimized and compared with a 99mTc-HMPAO leucocyte labelling procedure in vitro and in vivo. In the present study, leucocytes obtained from patients suffering from Crohn's disease, were isolated and labelled with 99mTc-HMPAO or labelled according the new 99mTc-tropolone procedure using 9.8 mM tropolone, 1 microM stannous chloride and 0.8 mM potassium borohydride (KBH4) at pH 5.5-6. Labelling efficiency with 99mTc-tropolone yielded 92 +/- 3%, which is higher compared to the 99mTc-HMPAO labelling procedure (64 +/- 13%) using 10(8) of leucocytes. In vitro stability and viability of both the tropolone and the HMPAO labelled cells was investigated. The viability test of the 99mTc-labelled leucocytes was performed in autologous plasma at 37 degrees C and compared with unlabelled leucocytes. After 18 hours of incubation a significant (P < 0.05) higher stability was observed for 99mTc-tropolone labelled leucocytes (84 +/- 5%) compared with that of 99mTc-HMPAO labelled leucocytes (73 +/- 5%). The viability of the 99mTc-labelled leucocytes observed for both labelling procedures was similar to unlabelled leucocytes. In vivo experiments were performed in mice. 99mTc-tropolone or 99mTc-HMPAO labelled murine mixed leucocytes were injected in mice, with a Staphylococcus aureus ATCC 25923 thigh infection. Analysis of scintigraphic images yielded a faster clearance of the 99mTc-tropolone labelled leucocytes. This was most likely due to a significant (P < 0.02) higher liver uptake at 4 hours after administration of the 99mTc-tropolone labelled leucocytes (19%) in comparison with 99mTc-HMPAO labelled cells (9%). Faster and significant (P < 0.02) higher accumulation of the 99mTc-tropolone labelled leucocytes was observed at the site of infection compared with 99mTc-HMPAO labelled leucocytes at all time-intervals after the administration of the 99mTc-labelled leucocytes. The new 99mTc-tropolone leucocyte labelling procedure, offers an attractive low-cost agent for research purposes.

Abscess↗
PubMed1994

Optimized localization of bacterial infections with technetium-99m labelled human immunoglobulin after protein charge selection.

To improve the scintigraphic detection of bacterial infections a protein charge-purified fraction of polyclonal human immunoglobulin was applied as a radiopharmaceutical. This purification was achieved by attaching the immunoglobulin to an anion-exchanger column and by obtaining the column-bound fraction with buffer. The binding to bacteria in vitro and the target to non-target ratios of an experimental thigh infection with Staphylococcus aureus or Klebsiella pneumoniae in mice were evaluated to compare the purified and the unpurified immunoglobulin. The percentage of binding to all gram-positive and gram-negative bacteria used in this study was significantly (P < 0.03) higher for the purified than for the unpurified immunoglobulin. For the in vivo study, mice were infected in the thigh muscle with Staph. aureus or K. pneumoniae. After 18 h 0.1 mg of technetium-99m labelled polyclonal immunoglobulin or 99mTc-labelled protein charge-purified polyclonal human immunoglobulin was administered intravenously. At all time intervals the target (infected thighs) to non-target (non-infected thighs) ratios for both infections were significantly higher (P < 0.03) for protein charge-purified polyclonal immunoglobulin than for unpurified polyclonal human immunoglobulin. Already within 1 h the infected tissues could be detected by the purified immunoglobulin. It is concluded that 99mTc-labelled protein charge-purified immunoglobulin localizes both a gram-positive and a gram-negative thigh infection more intensely and faster than 99mTc-labelled unpurified immunoglobulin.

Animals↗
PubMed1993

Improved detection of a staphylococcal infection by monomeric and protein A-purified polyclonal human immunoglobulin.

The present study was undertaken to compare the technetium-99m labelled non-specific polyclonal human immunoglobulin (Ig) with 99mTc-labelled monomeric human immunoglobulin (m-Ig), 99mTc-labelled, protein A-purified, human immunoglobulin (A-Ig) and 99mTc-labelled monomeric, protein A-purified, human immunoglobulin (mA-Ig) as tracer agents for the detection of a thigh infection with Staphylococcus aureus. In vitro the binding of the various tracer agents to bacteria at various intervals was determined. For the in vivo evaluation, mice were infected and received one of the various labelled proteins. Scintigrams were made 0.25, 1, 4 and 24 h later. All 99mTc-labelled Igs bound to bacteria in vitro: the percentages of binding for the m-Ig (from 1 h onwards) and A-Ig and mA-Ig (from 3 h onwards) were significantly higher than that for Ig. The in vivo target-to-non-target (T/NT) ratios were significantly higher from 4 h onwards for all purified Igs than for Ig. Protein A-purified Igs yielded higher T/NT ratios than m-Ig. Furthermore, the amount of activity in the liver was significantly lower 24 h after administration of m-Ig, A-Ig and mA-Ig than after administration of Ig. It is concluded that in this experimental infection 99mTc-labelled monomeric Ig localizes a staphylococcal thigh infection better and faster than 99mTc-labelled unpurified Ig. However, the accumulation obtained with protein A-purified Ig or protein A-purified monomeric Ig was the highest of all tracer agents tested.

Animals↗
PubMed1990

Effects of an antifibrin monoclonal antibody and fragments thereof on some properties of fibrin.

Antifibrin monoclonal antibody Y22, of IgG1-subclass, has its epitope in the D-domain of fibrin. In a thrombin time assay, Y22 and its F(ab)2 fragments interfere with clotting of citrated plasma. Transmission and scanning electronmicroscopic studies show that clotting of citrated blood or plasma in the presence of Y22 results in formation of thin, short fibrin fibres. The (smaller) Fab fragments of Y22 did not have an anti-clotting effect. This suggests that the anticoagulant effect of Y22 is due to steric hindrance of the association of fibrin monomers. A control antibody and its F(ab)2 and Fab fragments have no effect on fibrin formation. In a parabolic rate assay, Y22 Fab fragments interfered strongly with the fibrin-induced enhancement of the t-PA-catalyzed plasminogen activation, whereas intact Y22 and a control antibody did not. In contrast with their effects on the fibrin assembly, the effects of Y22, Y22-F(ab)2 and Y22-Fab on the capacity of fibrin to act as a rate-enhancer in the plasminogen activation by t-PA appears to decrease with the size of the immunoreactive entity. As is discussed, this may be due to the differential accessibility of sites involved in stimulation and polymerization which are located in the fibrin D-domain.

Amino Acid Sequence↗
PubMed1984

Quantitation of carcinogen-DNA adducts by a standardized high-sensitive enzyme immunoassay.

A highly sensitive competitive enzyme immunoassay has been developed, allowing accurate determination of DNA containing the adducts N-(deoxyguanosin-8-yl)-N-acetyl-2-amino-fluorene, N-(deoxyguanosin-8-yl)-2-aminofluorene, 1-[6-(2,5-diamino-4-oxopyrimidinyl-N6-deoxyriboside)]-3-(2-fluo renyl)urea or trans-(7R)-N2-[10-(7 beta,8 alpha,9 alpha-trihydroxy-7,8, 9,10-tetrahydrobenzo[a]pyren)yl]deoxyguanosine. Standard amounts of the carcinogen-modified DNA preparations (25 fmol/500 ng) were coated on the wells of microtitre plates. Various amounts of the modified DNA preparations were used as inhibitors and added before binding of the antibodies (dilution 1:10(6). As second antibody, goat anti-rabbit IgG coupled to alkaline phosphatase was used. The amount of enzyme was determined with 4-methylumbelliferyl phosphate as substrate (Van der Laken et al., 1982). The 50% inhibition values of the standard curves are in the range of 2-16 fmol for the enzyme immunoassays. It is possible to add up to 50 micrograms of unknown DNA to the wells, allowing for comparison with the same quantity of modified DNA. This extends the limit of detection to 1-6 adducts per 10(8) nucleotides. The sensitivity of the assay seems sufficient to demonstrate exposure of humans to known chemical carcinogens.

Animals↗
PubMed1982

Measurement of O6-ethyldeoxyguanosine and N-(deoxyguanosin-8-yl)-N-acetyl-2-aminofluorene in DNA by high-sensitive enzyme immunoassays.

Antibodies raised in rabbits against the bovine serum albumin conjugates of O6-ethylguanosine and N-(guanosin-8-yl)-N-acetyl-2-aminofluorene have been used to develop a high-sensitive enzyme-linked immunosorbent assay (HS-ELISA) for the quantification of adducts in DNA modified by ethylating agents and N-acetyl-2-aminofluorene. Linear dose-response relations were obtained in the non-competitive HS-ELISA between 0.5 fmol and 50 fmol O6-ethyldeoxy-guanosine per 2.8 microgram DNA, and between 0.1 fmol and 20 fmol N-(deoxyguanosin-8-yl)-N-acetyl-2-aminofluorene per 0.8 microgram DNA. The sensitivity of an ultrasensitive radioimmunoassay was in the same order of magnitude. Modification levels as low as 0.1 mumol of adduct/mol DNA-nucleotides (1: 10(7)) can be detected by each assay.

2-Acetylaminofluorene↗
PubMed1982

Immunological detection and quantification of the reaction products of 2-acetylaminofluorene with guanine in DNA.

Rabbits were immunized with N-(guanosin-8-yl)-N-acetyl-2-aminofluorene bound to bovine serum albumin. The presence of specific antibodies was determined in a radioimmunoassay by competing N-([5'-3H]-guanosin-8-yl)-N-acetyl-2-aminofluorene (specific radioactivity, 25 Ci/mmol) as tracer against various carcinogen-modified and normal DNA components. Highly specific antibodies were obtained (affinity constant, 6.0 X 10(9) l/mol) and have been employed in a sensitive radioimmunoassay to determine N-(deoxyguanosin-8-yl)-N-acetyl-2-aminofluorene directly in enzymatic hydrolysates of DNA modified with N-acetoxy-N-acetyl-2-aminofluorene. In a similar fashion, the modified nucleoside could be detected in hydrolysed rat liver DNA following administration of N-acetyl-2-aminofluorene, with a detection limit of one carcinogen-modified deoxyguanosine moiety per 10(6) deoxynucleotides in 150 micrograms of DNA.

2-Acetylaminofluorene↗