Abstracts 2003
Dendritic Cell Immunotherapy for Cancer and HIV : Reality and PromisesViggo F.I. Van Tendeloo, PhD, University of Antwerp (UA), Laboratory of Experimental Hematology, Antwerp University Hospital
Since their original discovery in 1973 by Steinman and Cohn, dendritic cells (DCs) have advanced from a curiosum to nature’s adjuvant in immune–based strategies. In the 1990s, methods became available to generate large numbers of DCs in vitro, which allowed to study their role in the immune system and energized their use in clinical trials for immunization of humans against cancer and infectious diseases. The use of in vitro cultured DCs in cellular vaccines entails a number of functions critical to the success of DC vaccines: (i) DCs, more than any other APC, express extremely high levels of important adhesion and costimulatory molecules critical for (naive) T cell activation; (ii) direct ex vivo loading of DCs with relevant tumor or viral antigens bypasses the need for in vivo transfer of antigens from the tumor or virally infected cell to the DC; and (iii) antigen–loaded DCs have the ability to migrate to the T cell areas of lymphoid organs to get in contact with rare antigen–specific T cells. The form of antigen transferred to the DCs can range from minimal MHC class I–restricted peptides to proteins, cell lysate or even whole tumor cells. We have explored ex vivo gene transfer using RNA and DNA transfection to introduce genes encoding antigen(s) and have demonstrated that electroporation of DCs with in vitro transcribed antigen–encoding mRNA results in very high transfection levels and, importantly, in efficient priming of antigen–specific CD8+ T–cells. Currently, a large number of phase I and II DC trials have already been performed in an attempt to test their safety and efficacy in cancer patients. Taken together, these trials were shown to be safe and well–tolerated and have resulted in strong immune responses in most tumor patients, even those with advanced stages of cancer. In some cases, the clinical status of the patients receiving DC therapy significantly improved ranging from complete remission to stable disease. However, the common bottle–neck in all these trials remains the lack of correlation between the detection of a strong anti–tumor immune response and the observed clinical benefit. This observation could be in part ascribed to various tumor immune escape mechanisms and the progressive disease state of the patients in phase I and II trials. More importantly, there are still critical unknowns which deserve clarification before DC vaccines can be taken to another level in humans, such as type of DCs used for vaccination, maturation/activation state of the DCs and method of antigen loading in terms of its efficacy to elicit both CD4+ and CD8+ T cell responses. Next, clinical trials will be necessary to translate answers to these challenges into the clinic. Furthermore, DC researchers are trying to establish adequate and sensitive surrogate immuno–monitoring assays that correlate with clinical outcome. In conclusion, although still in its infancy, DCs have provided immunologists new avenues for active immunization of cancer and HIV by extending the playing field from T cells and cytokines to immunogens presented by the key players of the immune system, the dendritic cells. Finally, given the complexities and costs of cellular therapy requiring ex vivo culturing of patient’s cells, a more desirable goal would be to target antigens to DCs directly in vivo . Methods to increase the frequency of DC precursors in vivo have been developed and DC targeting molecules have been identified, thereby paving the way for in vivo DC therapy.
DC Vaccination for Malignant Brain Tumors.
Steven De Vleeschouwer, Martine Adé, Stefan Rutkowski, Eckhard Kaempgen, Johannes Wolff, Joachim Kuehl, Stefaan Van Gool
Prof. Dr. Stefaan Van Gool, Pediatric hemato–oncology and neuro–oncology, Laboratory of Experimental Immunology, University Hospital Gasthuisberg Stefaan.VanGool@uz.kuleuven.ac.be
Glioblastoma multiforme (GBM) is the end of a spectrum of glial tumors according to malignancy. The pilocytic astrocytoma (grade I) and fibrillary (or protoplasmic or gemistocytic) astrocytoma (grade II) are considered as low–grade glioma's. In contrast, the anaplastic astrocytoma (grade III) and glioblastoma's are high–grade malignant glioma's. The yearly incidence of the high–grade glioma's is about 2.4 per 100000 adults. Although the combination of surgical resection, radiotherapy and chemotherapy can create a significant reduction of the tumor volume, the risk for relapse is high. The ultimate prognosis of these patients is very poor, with a median survival of less than 2 years. Therefore, new treatment strategies have to be developed at the preclinical level, and, at a later stage, transferred into clinical practice.
DC Loading with Tumor Proteins.
Dendritic cells (DC) are professional antigen–capturing and -presenting cells (APC). Their fundamental role in initiating and directing a primary immune response is well established, and is used to design tumor vaccination strategies. Successful pulsing of DC with tumor antigens has been reported using different sources: RNA, peptides, proteins from tumor lysate or homogenate, or apoptotic or necrotic cell bodies. We studied whether the protein fraction from glioblastoma tumor homogenate was taken up and handled appropriately by DC. Freshly resected tumor tissue from patients with glioblastoma multiforme was homogenised mechanically. The tumor protein content in the homogenate was measured with Coomassie blue technique. These tumor proteins were labelled with FITC and kept frozen. DC were differentiated out of freshly isolated human PBMC from volunteers (MoDC) in the presence of rIL–4 and rGM–CSF. After 7 days, the FITC–labelled homogenate was added to the DC culture. The uptake and handling of the FITC–labelled tumor proteins was measured using FACS and confocal microscopy. We showed a time–dependent uptake and re–appearance on the cell surface of the FITC–proteins by the MoDC. Uptake and processing of the FITC–proteins was dependent on cell metabolic activity, and was not present when the cells were kept at 4°C. The re–appearance on the membrane was polarised. The uptake and re–appearance on the membrane surface of the MoDC was extremely strong in case of overnight loading. The appearance of FITC–proteins on the cell surface remained stable, even when loaded DC were kept frozen and were thawed afterwards. Addition of the DC maturation cytokines rTNF–a, rIL–1ß and PGE2 at time of loading did not interfere with the uptake and processing of FITC–proteins. To our knowledge, this is the first report to show that the protein fraction of a crude glioblastoma tumor homogenate is taken up and processed by MoDC, resulting in a polarised re–appearance on the cell surface. These data support the use of MoDC, which are loaded with tumor homogenate proteins, as tumor vaccine to treat patients with glioblastoma multiforme according to our phase I/II trial HGG–IMMUNO–2003.
In Vitro Stimulation Experiments.
T cells were stimulated with autologous MoDC, loaded with tumor proteins from GBM tumor cells. First stimulation was performed in the presence of rIL–6 and rIL–12, second stimulation was performed in the presence of rIL–2 and rIL–7. Read–out systems of T cell activation consisted of T cell proliferation, measured by thymidine incorporation; cytokine production (IL–10, IL–5, IL–13, IFN–g), measured by ELISA; cytotoxic activity, measured by MTT assay or by CTO assay. After two stimulations, there was a significant shift towards Th1 cytokine production profile when T cells were stimulated with loaded MoDC, as compared to stimulation with unloaded DC or cytokines without DC. Cytotoxic activity of stimulated T cells could be demonstrated against CTO–labelled tumor cells, which expressed annexin–V and 7–AAD after 4 hours co–culture. Two patterns of T cell activity could be detected. According to pattern 1, T cells significantly suppressed tumor growth on the MTT assay, had proliferative capacity and produced IL–10. In contrast, according to pattern 2, T cells did not suppress tumor growth on MTT assay, had no proliferative capacity and did not produce IL–10. These data suggest that MoDC loaded with tumor proteins are able to stimulate T cells that have antitumoral activity. The reason why two patterns of responses can be distinguished is not yet clear.
Results of the Pilot trial of HGG–IMMUNO–2003 (status September 2003).
The Leuven–Würzburg group has treated thusfar 12 patients (median age of 36 years (range: 11 – 78 y)) with autologous DC loaded with autologous tumor proteins according to a pilot protocol ("Heilversuch"), which was approved by the local ethical committees in both vaccination centers. All patients had a malignant glioma histology. Eight patients were vaccinated at first relapse, while 4 patients had more than one malignant event prior to vaccination. Vaccination was given at week 1, 3, and further each 4 weeks. A median of 5 (range: 2 – 7) vaccines was given. In 5/12 patients, a response was observed. Three out of 6 patients in whom the tumor was completely resected are in continuous complete remission with a follow up of 28, 27 and 15 months. In the 6 patients with residual tumor load after surgery, vaccination induced 1 stable disease, and 1 partial response. Seven patients died after a median postoperative survival of 7 months (range: 4 – 12 m). Five other patients are still alive with a median postoperative follow up of 21 months (range = 12 – 28 m). There were no major side effects unless in one patient who had gross tumoral disease prior to vaccination and who developed repetitively vaccine–related peritumoral oedema. Immune therapy with autologous mature dendritic cells loaded with crude tumor homogenate for patients with relapsed malignant brain tumors is feasible without major side effects. Although it is too early to assess efficacy, some interesting case histories are promising and support the continuation of the development of dendritic cell immune therapy for malignant brain tumors, especially in patients with minimal residual tumor burden.
Cellular analysis of broncho–alveolar lavage fluid in patients with interstitial lung disease.
Heleen van Velzen–Blad, Medical Microbiology and Immunology, Antonius Ziekenhuis, Nieuwegein, on behalf of the Dutch Working Party Bronchoalveolar Lavage
To assess the inflammatory reaction in the lower airways, a Broncho–Alveolar Lavage (BAL) can be performed. BAL is a technique that allows the recovery of both cellular and non–cellular components from the epithelial surface of the lower respiratory tract. BAL has a role in the diagnosis of Interstitial Lung Disease (ILD). Data should be interpreted in the context of clinical, radiological and other laboratory results. Further, BAL is also an important, low invasive tool in our understanding of the patho–physiology of ILD.
Sarcoidosis and extrinsic allergic alveolitis (EAA) are two frequently encountered interstitial lung diseases. Although the patho–physiology of both diseases is very different, their clinical presentation can be similar. Sarcoidosis is a multisystem granulomatous disorder of unknown aetiology and characterised pathologically by the presence of non–caseating granuloma in the lung and /or other involved organs. EAA is a (type III and IV) hypersensitivity reaction of the lung parenchyma with inflammation in the alveoli and interstitial spaces; it is induced by the acute or chronic inhalation of a wide variety of inhaled materials. Bird fancier’s disease is probably the most well known example of EAA.
In general both diseases show characteristic findings in the cellular differentiation of the recovered BAL leukocytes, which are thought to represent the cells present at the alveolar level. Whereas macrophages are the predominant cell population in healthy individuals, the BAL fluid of patients with sarcoidosis or EAA show a lymphocytic alveolitis. The BAL lymphocytes can be differentiated in T–, (CD4+ and CD8+ T cell subsets) B– and NK– cells by flowcytometric immune–phenotyping. It has been shown that different T cell subset patterns can be found in particular ILDs, although these patterns lack a high degree of disease specificity.
The technical procedures of BAL during bronchoscopy as well as the handling/processing of the BAL–fluid have recently been standardised by the Dutch working party Bronchoalveolar Lavage. In this consensus it was agreed that a fixed minimal program to identify lymphocyte (sub–) populations for the diagnostic work–up of ILD is used. Also reference values for BAL–cells were determined and a nation wide quality control (QC) program has been developed. As a consequence of the standardisation procedure and the QC–program, results from different laboratories are now comparable and BAL material can be exchanged between laboratories for diagnostic and research purposes.
This protocol also enables multi–centre studies in which new and promising markers can be evaluated.
POSTERS
Cytotoxicity of Spirulina Extracts on Human Cancer Cell Lines.
Sarah Baatout (1,*), Matthias Deruelle (2), Hanane Derradji (1,2), Max Mergeay (1), PatrickVan Oostveldt (2), Sofie Bekaert (2,*).
(1) Laboratory of Radiobiology, Belgian Nuclear Research Centre, SCK–CEN, Mol, Belgium; (2) Laboratory for Biochemistry and Molecular Cytology, Department for Molecular Biotechnology, FLTBW –Ghent University, Belgium.
Corresponding author: Sarah Baatout, sbaatout@SCKCEN.BE
The cyanobacteria A. platensis (also named spirulina) is known for its potential health effects and is reported in the literature to diminish cholesterol level, to be active against diabetes, hypertension, anemia and iron bioavailability, to reduce kidney toxicity from drugs and heavy metals, to help to loose weight, to inhibit HIV replication and to reduce the effects induced after Chernobyl radiation. Some of the beneficial effects are attributed to its natural amount of antioxidants (spirulina is the richest beta–carotene (main source of vitamin A) food known (10x > carrots) as well as iron (58 x > raw spinach and 28 x > raw beef liver). However, spirulina also contains toxins that could be responsible for its cytotoxicity.
In this study, we were interested in studying the threshold of intrinsic cytotoxic effect of spirulina on human cancer cells and its cell type dependency. For that purpose, we used flow cytometry to estimate apoptosis and necrosis in three human leukaemic cell lines (HELA : cervix carcinoma; IM–9 : multiple myeloma; K562 : chronic myelogenous leukaemia). Cells were cultured in the presence of spirulina (concentrations ranging from 0 to 500 µg/ml) between 15 to 40 hours. Apoptosis and necrosis were evaluated by the annexin–V–PI staining, cell size and granularity. Early apoptosis was checked by analysing the maintenance of mitochondrial membrane potential (DioC(6)3) and the production of superoxides (hydoethidine) whereas advanced apoptosis was studied by measuring the propensity of the sub–G1 peak, esterase activity (fluorescein diacetate) and intracellular pH (carboxyfluorescein diacetate).
The three cell lines had a different sensitivity to spirulina extracts (HELA most resistant > K562 > IM–9 most sensitive). Indeed, concerning the HELA cell line, a significant effect on enzymatic activity was observed only at 500 µg/ml spirulina whilst for K562 cells, spirulina extracts showed a significant effect on K562 enzymatic activity, cell size, cell granularity and intracellular pH from 250 µg/ml. IM–9 showed a significant induction of apoptosis and necrosis when cultured in the presence of 20 µg/ml onwards.
These results suggest that, depending on the concentration, spirulina extracts could enhance apoptosis in leukaemic cell lines and may open a perspective for cancer therapy at the condition to also address the intrinsic toxicity of spirulina on normal cells.
(*) These authors equally contributed to the work.
This work is partially supported by ESA/ESTEC (contract number n°15680/01/NL/ND).
Changes Associated with Temperature, H2O2 and pH Stresses on the Physiology of R. rubrum.
Sarah Baatout (1,*), Ruddy Wattiez (3,*), Larissa Heindrickx (2), Florence Marty (2), Max Mergeay (1, 2).
(1) Laboratory of radiobiology and (2) Laboratory for microbiology,
Belgian Nuclear Research Center, SCK–CEN, Mol and (3) Laboratory of biological chemistry, University of Mons–Hainaut, Belgium.
Corresponding author: Sarah Baatout, sbaatout@SCKCEN.BE
Flow cytometry and cell intact mass spectrometry provide powerful means to measure a wide range of cell characteristics in microbiological research. In order to estimate fine physiological changes associated with temperature, H2O2 and pH stresses, flow cytometry and cell intact mass spectrometry were employed to estimate the extent of damage on the maintenance of membrane integrity and potential, esterase activity, intracellular pH and production of superoxides in Rhodospirillum rubrum, chosen for its potential use in life support systems in space.
Suspensions of R. rubrum were submitted to a 1–hour temperature stress (from –170°C to 70°C), H202 stress (at various concentrations from 0 to 880 mM) or pH stress (from 2 to 12). For flow cytometry, fluorochromes, including propidium iodide, rhodamine–123, 3,3'–dihexyloxacarbocyanine iodide, fluorescein diacetate, carboxy–fluorescein diacetate or hydroethidine were chosen as analytical parameters for identifying the physiological state and the overall fitness of individual cells. An individual cell's physiological state was assessed with a Coulter EPICS XL analyser based on the differential uptakes of these fluorescent stains. The laser desorption technologies could be used for the characterization of intact micro organisms by generation of different ion spectra of molecules (especially proteins) desorbed from bacteria. Specific cell fingerprint of R. rubrum were analysed in different stress conditions in linear MALDI configuration. R. rubrum exhibited varying staining intensities following temperature, H2O2 or pH stress. Membrane permeability and potential, esterase activity, intracellular pH and production of superoxide anions were increased to high levels after freezing (–170°C, –80°C and sometimes –20°C) or heat (45°C, 50°C, 60°C and 70°C) treatments. However, almost no change could be detected between 4°C and 37°C (except DioC(6)3 staining). Following oxidative stress, membrane permeability, membrane potential (measured following DioC(6)3 staining), esterase activity and superoxide anion production were increased in function of the H202 concentration. The physiological effects of pH stress on R. rubrum were pronounced at extreme pH conditions (acidic or alkaline). However, no change in physiology could be detected from pH 5 to pH 9.
In conclusion, we showed that there is a range of significant physiological alterations that occur after temperature, H2O2 or pH stress. Fluorescent staining methods coupled with flow cytometry as well as cell intact mass spectrometry are useful for monitoring physiological changes induced not only by temperature, H2O2 and pH stresses but also radiation or pressure that we are extensively studying in our laboratories.
(*) These authors equally contributed to the work.
This work is currently supported by ESA/ESTEC (contract number n°15680/01/NL/ND).
Temperature–induced Changes on Bacterial Physiology.
Sarah Baatout (1), Patrick De Boever (2), Max Mergeay (1,2), (1) Laboratory of Radiobiology and (2) Laboratory for Microbiology,
Belgian Nuclear Research Center, SCK–CEN, Mol, Belgium.
Corresponding author: Sarah Baatout, sbaatout@SCKCEN.BE
Flow cytometry provides a powerful means to measure a wide range of cell characteristics in microbiological research. In order to estimate physiological changes associated with temperature stress, flow cytometry was employed to estimate the extent of damage on the maintenance of membrane integrity and potential, esterase activity, intracellular pH and production of superoxides in four bacterial strains (Ralstonia metallidurans, Escherichia coli, Shewanella oneidensis and Deinococcus radiodurans). The physiology of the bacterial strains is being studied in order to understand their behaviour and resistance under extreme conditions (such as oxidative stress). This knowledge is of importance in the light of the potential use of these strains for bioremediation.
Suspensions of R. metallidurans, E. coli, S. oneidensis and D. radiodurans were submitted to a 1–hour temperature stress (–170°C, –80°C, –20°C, 4°C, 15°C, 28°C, 37°C, 45°C, 50°C, 60°C or 70°C). Fluorochromes, including propidium iodide, rhodamine–123, 3,3'–dihexyloxacarbocyanine iodide, fluorescein diacetate, carboxy–fluorescein diacetate or hydroethidine were chosen as analytical parameters for identifying the physiological state and the overall fitness of individual cells. An individual cell's physiologic state was assessed with a Coulter EPICS XL analyser based on the differential uptakes of these fluorescent stains.
The four bacterial strains exhibited varying staining intensities. For all bacterial strains, the physiological status was not affected between 4°C and 37°C in comparison with 28°C, which represents the reference growth temperature. Moderate physiological damage could be observed at 45°C. In all four strains, membrane permeability and potential, esterase activity, intracellular pH and production of superoxide anion production were increased to high levels after freezing (–170°C, –80°C and –20°C) or heat (+50°C, +60°C and +70°C) treatments.
It is apparent that a range of significant physiological alterations occurs after temperature stress. Fluorescent staining methods coupled with flow cytometry are useful for monitoring physiological changes induced not only by temperature stress but also oxidative stress, radiation, pressure, pH etc. These studies are currently being investigated in our laboratories.
This work is partly supported by ESA/ESTEC (contract number n°15680/01/NL/ND).
Estimation of Telomere Length in Human Cancerous Cell Lines Using Flow Cytometry Associated with in situ Hybridization
H. Derradji (1,2), S. Bekaert (2), P. Van Oostveldt (2), M. Mergeay (1) and S. Baatout (1)
(1) Laboratory of Radiobiology, Belgian Nuclear Research Centre, SCK–CEN, Mol, Belgium; (2) Laboratory for Biochemistry and Molecular Cytology, Department for Molecular Biotechnology, FLTBW –Ghent University, Belgium.
Corresponding author: Sarah Baatout, sbaatout@SCKCEN.BE
The nucleoprotein structures present at the very tip of the chromosomes are called telomeres; their proper functioning is vital for cell proliferation and essential in several biological functions. The telomeres cap the chromosome ends preventing them from end–to–end fusion and from being recognized as broken DNA ends.
Estimating average telomere length within cells is commonly achieved performing Southern blotting and densitometry. Telomere length of individual chromosomes within individual cells can be determined via quantitative fluorescence in situ hybridization (Q–FISH) in combination with fluorescence microscopy. These methods are both very elaborate and time consuming. The combination of Q–FISH and flow cytometric techniques give rise to a flow–FISH technique, which is a new approach regarding the determination of the telomere length distribution among each individual cell, with the advantage to be able to perform fast analysis of proliferating cells. In our setup we used a fluorescein labelled PNA probe (for telomere length) and a DNA counterstaining with propidium iodide (cell cycle and calibration). Different methodological setups were compared.
The flow–FISH was performed on three human cancerous cell lines: K–562 (chronic myelogenous leukaemia), IM–9 (multiple myeloma) and 1301 (T cell lymphoblastic leukaemia). Different fixation procedures (ethanol 80%, methanol 80% and formaldehyde 4%) were tested together with different fixation times (15', 30' and 60') as well as hybridization times (2h and overnight). For each of these protocols the following parameters were compared: forward scatter (related to the cell size), side scatter (related to the cell granularity), DNA (FL3 and FL4 fluorescence) and PNA content (FL1 fluorescence) using an EPICS XL flow cytometer. Regarding the fixation procedures, we preferred methanol followed by ethanol and formaldehyde, with respect to the efficiency to measure the different parameters. The duration of the fixation did not show significant interference in the results quality. The overnight hybridization appeared to be more effective when compared to the 2h–hybridization. In conclusion, we show that the development of a flow–FISH technology can allow on a single cell level, a rapid and quantitative estimation of telomere length and other cellular parameters (size, granularity, DNA, ...) on thousands of cells simultaneously. The development of this methodology is under progress in our laboratory in order to extend it to other cell types (lymphocytes, fibroblasts, ...).
Cytometric Analysis of Radiation–induced Damage and Death in Human Cancer Cells
Sofie Bekaert (2,*), Matthias Deruelle (2), Hanane Derradji (1,2), Max Mergeay (1), Patrick Van Oostveldt (2), Sarah Baatout (1,*)
(1) Laboratory of Radiobiology, Belgian Nuclear Research Centre, SCK–CEN, Mol, Belgium; (2) Laboratory for Biochemistry and Molecular Cytology, Department for Molecular Biotechnology, FLTBW –Ghent University, Belgium.
Corresponding author: Sarah Baatout, sbaatout@SCKCEN.BE
The anti–tumorigenic, anti–inflammatory and antioxidant effects of several naturally occurring substances (spirulina, tea catechins, resveratrol, curcuma) are currently being examined in the authors' laboratories. Besides their presumed inherent protective effect on stress induced by reactive oxygen species (byproducts of normal cell metabolism), the extent to which these agents have radio–protective properties for normal cells and/or radio–sensitizing properties for cancer cells is being studied.
In these initial experiments the physiological effects and dose–dependency of X–irradiation on cancer cell lines with different radiation sensitivities (HELA : cervix carcinoma; IM–9 : multiple myeloma and K–562 : chronic myelogenous leukaemia) were studied. The parameters investigated were the extent of cellular and nuclear damage, and the induction of cell death (apoptosis and necrosis) as a response to the irradiation stress. The cell lines were cultivated under normal conditions and subsequently X–irradiated (doses between 0 and 8 Grays) in log phase. Various flow cytometric analyses were performed after 24h or 48h of culture. Cell size, granularity and annexin–V– propidium iodide staining were monitored to evaluate apoptosis and necrosis. A distinction was made between the early apoptotic effects (via DioC(6)3 and ROS measurements) and the manifestations of advanced cell death (via esterase activity quantification, intracellular pH and occurrence of sub–G1 peak).
The preliminary results, in agreement with results of similar experiments, showed a significant, cell–dependent difference in radiation sensitivity (Hela most radiation resistant >K–562>IM–9 most radiation sensitive) and dose dependency (especially at high doses). In all cell types, the proportion of apoptotic cells (as measured by Annexin V–propidium iodide staining), and the effect on metabolic activity was greater 48h after irradiation than 24h after. In contrast to IM–9 and K–562, no increase in reactive oxygen species production or significant effects on intracellular pH (carboxyfluorescein diacetate) were measured in Hela cells. However, the mitochondrial membrane potential (DioC(6)3) was only preserved in Hela cells for 24h, whereas in IM–9 it remained intact after 48h.
In addition, cell proliferation, morphology and polyploidy levels during extended growth are being monitored by trypan blue exclusion tests and microscopical analysis. Measurement of telomere length and detection of DNA single stranded breaks (S1 nuclease test) is performed every 10 population doublings.
The final goal is to repeat this setup in the presence of radio–protective substances.
(*) These authors equally contributed to the work.
Angiogenic Switch and Angiogenic Heterogeneity in Multiple Myeloma
Kewal Asosingh1, Hendrik De Raeve2, Eline Menu1, Ivan Van Riet1, Benjamin Van Camp1, Karin Vanderkerken1
1Department of Hematology and Immunology, Vrije Universiteit Brussel (VUB); Brussels, Belgium; 2Department of Pathology, University of Antwerp, Antwerp, Belgium. K.A. and K.V. are postdoctoral fellows of FWO–Vlaanderen.
Corresponding author: Kewal Asosingh, kewal@vub.ac.be
The important role of angiogenesis is well recognized in solid tumors. Like normal cells, cancer cells need an adequate supply of oxygen and nutrients and an effective way to dispose waste products. Early tumor stages are characterized by a pre–angiogenic stage of slow tumor progression. During this stage the pre–existing host vasculature is sufficient to maintain the tumor. However, for progressive tumor growth, a tumor vasculature must be induced. During the pre–angiogenic phase, the tumor population consists of only few cancer cells with the capacity to induce angiogenesis. When enough cancer cells become angiogenic, angiogenic switch occurs, followed by an angiogenic stage of progressive tumor expansion.
Angiogenesis has been observed in multiple myeloma (MM), however its significance in myelomapathobiology is unclear. At clinical presentation, both the myeloma disease as well as the associated angiogenesis are well established. Insights into the pre–clinical stages would illuminate the role of angiogenesis in MM. Such an investigation requires an in vivo model and the 5T2MM experimental mouse model was used. Tumor inoculated mice were analyzed at weekly intervals. Bone marrow (BM) tumor load and microvessel density (MVD) were assessed by FACS staining with anti–5T2MM idiotype antibodies and CD31 staining of BM sections for histological analysis, respectively. As reported in human MM, a good correlation was observed between the MVD and the tumor load (R2= 0.987), serum M–component (R2= 0.981) and BrdU incorporation (R2= 0.816). In the early disease phases (first 6 weeks), a pre–angiogenic stage of small tumor aggregates was observed within the pre–existing network of BM sinusoidal capillaries. During this stage the tumor growth was slow, as indicated by the evolution of the tumor load and by histological analysis. Between week 6 and week 7 the MVD in tumor infiltrated areas started to increase, indicating the occurrence of an angiogenic switch. Subsequently an angiogenic stage of progressive tumor growth and large, confluent tumor nodules were observed. Angiogenic heterogeneity has been reported in solid tumors and MM cells have been reported to have a heterogeneous CD45 expression. Therefore we analyzed whether there was any association between CD45 expression and angiogenesis. FASC analysis indicated that during the pre–angiogenic stage, the majority of the MM–cells were CD45+. There was a gradual increase of percentage CD45– MM during the disease progression and at the time point of angiogenic switch, the MM–population consisted of CD45– MM–cells mainly. Tumor–derived vascular endothelial growth factor (VEGF) is known to play a crucial role in tumor neovascularization. Quantitative RT–PCR of sorted CD45 subsets indicated four fold higher VEGF120 and VEGF164 transcript levels in CD45– MM cells. VEGF ELISA revealed high VEGF protein secretion by CD45– MM cells (179–478 pg/ml), but no detectable (< 3 pg/ml) secretion by the CD45+ MM cells, indicating angiogenic heterogeneity among the MM cells.
Our data indicate important similarities between angiogenesis in solid tumors and angiogenesis in MM and suggest that as in solid tumors, angiogenesis is an active and important process in myeloma pathogenesis.
Tracking the Follicular Lymphoma Cells in Flow Cytometry; Usefulness Of A New Antibody Combination CD44 And CD38.
G. Detry(1), B. Drenou(2), A Ferrant(3), I. Théate(4), L. Michaux(3), J–M Scheiff(1),
D. Latine(1), P. Leveugle(1), A–M Mazzon(1), V. Deneys(1)
(1)Laboratoire d’immunohématologie, Clinique St–Luc, Bruxelles ; (2)Laboratoire d’hématologie immunologie, CHU de Rennes ; (3) Service d’hématologie, Clinique St–Luc, Bruxelles ; (4)laboratoire d’anatomo–pathologie, Clinique St–Luc, Bruxelles.
Corresponding author : deneys@sang.ucl.ac.be
Follicular lymphoma (FL) is one of the most common adult Non–Hodgkin’s Lymphoma (NHL) in Western Europe and North America1. Although histologically low grade and clinically indolent, FL remains mostly incurable. It is defined as a neoplasic proliferation of follicular B cells ,centrocytes and centroblasts, that form, at least partially, a nodular pattern. The diagnosis of this lymphoma is based on morphology and can be confirmed by cytogenetics and molecular studies. Nevertheless morphology studies can be extremely difficult in extra–nodular lymphoma. Furthermore cytogenetic and molecular studies can be confusing since the closely associated t(14;18) translocation can be found in other non Hodgkinian lymphoma or in normal blood donors. The study in flow cytometry of the expression of CD10, the hallmark of follicular cells, is not sufficient as normal B lymphoid progenitors express CD10 and as some FL do not express this marker.
Aim of the study:
Characterisation of the expression of CD44 in flow cytometry which, in combination with the study of the expression of CD38 allows us to target the follicular lymphoma cells.
Material and methods:
114 samples from 98 patients of mean age 65 years (range 36–98) and 10 samples from blood donors were collected (table 1).
Results:
The expression of CD44 is decreased in germinal centre cell and this lower fluorescence expression determine the follicular origin of the neoplasic cells. The combination of CD44 and CD38 is usefull in bone marrow sample as CD38 expression allows the separation of immature CD44low B lymphocytes from the neoplastic FL cells. Using this simple approach we detect follicular cells with an excellent sensibility (94.5%) and an excellent specificity (97.2%).
Engraftment of SCID–Repopulating Cells Is Independent of VLA–4 and Mediated by VLA–5 after Short–Term Expansion Culture.
Jacques Foguenne, Olivier Giet, Ivano Di Stefano, Yves Beguin and André Gothot.
Laboratory and Clinical Hematology, University of Liège, Liège, Belgium.
Corresponding author: J.Foguenne@student.ulg.ac.be
The pivotal role of VLA–4 in mediating hematopoietic stem cell (HSC) lodgment in the bone marrow (BM) is well documented while the importance of VLA–5 appears to be less significant. Several studies indicate that ex vivo cytokine stimulation induces defective HSC engraftment. We used NOD/SCID b 2m null mice repopulating cell (SRC) assays to compare the activity of VLA–4 and VLA–5 integrins in engraftment of unmanipulated and cytokine–treated human cord blood (CB) HSC. Recipient mice were transplanted with 150–200x103 freshly isolated CB CD34+ cells or their expansion product following 3 days of serum–free culture supplemented with SCF, FL, TPO, IL–6 and G–CSF. Integrin function was assessed by incubating grafts with neutralizing antibodies P4C2 (anti VLA–4) or P1D6 (anti VLA–5) prior to infusion. Control cells were treated with anti–CD34 antibody. Human chimerism in recipient BM was determined by flow cytometric detection of human CD45+ cells. Co–expression of CD19 or CD33 was used to evaluate multilineage repopulation. After transplantation of uncultured control CD34+ cells, human chimerism was 37.6+/- 11.4% CD45+ cells. VLA–4 neutralization resulted in decreased engraftment (1.6+/- 0.9% CD45+ cells, P<0.05), while VLA–5 neutralization had no significant effect (22.9+/- 9.9% chimerism). After expansion culture, BM repopulation by control cells was at 36.3+/- 9.3%. Prior incubation of expanded cells with anti VLA–4 did not affect SRC activity (48.9+/- 7.4% chimerism) whereas VLA–5 neutralization reduced engraftment down to 2.7+/- 1.1% CD45+ cells (P<0.05). All mice were reconstituted with lymphoid and myeloid cells in similar ratios, indicating that VLA–4 and VLA–5 blocking antibodies did not target populations of committed progenitors but rather inhibited engraftment of multilineage reconstituting cells. When direct homing of CD34+ cells in the recipient mice BM was determined at 20 hours after transplant, similar changes in integrin activity were detected. BM homing of uncultured CD34+ cells (control value: 0.98+/- 0.09% of infused cells) was significantly inhibited by VLA–4 neutralisation (0.06+/- 0.01%, P<0.05) while VLA–5 neutralisation had no such effect (1.10+/- 0.17%). On the contrary, homing of expanded CD34+ cells (control value: 0.75+/- 0.19%) was not significantly affected by blocking VLA–4 (0.42+/- 0.03%) but was markedly reduced after incubation with VLA–5 blocking antibody (0.15+/- 0.04%, P<0.05). In conclusion, while homing and engraftment of native human SRC are VLA–4 dependent, ex vivo expansion is associated with VLA–4 inactivation which uncovers the role of VLA–5 in mediating in vivo hematopoietic reconstitution.
Flow–Cytometric Evaluation of Chemokine–Induced Transwell® Cell Migration
S. Hatse1, K. Vermeire1, K. Princen1, D. Huskens1, B. De Klerck2, P. Matthys2 and D. Schols1
1Laboratory of Virology and Chemotherapy and 2Laboratory of Immunobiology, Rega Institute for Medical Research, Minderbroedersstraat 10, B–3000 Leuven, Belgium.
Corresponding author : sigrid.hatse@rega.kuleuven.ac.be
Chemokines are small cytokines that direct the migration of leukocytes and thus play an important role in numerous physiological and pathological processes. In particular, the chemokine/receptor pair stromal cell–derived factor–1(SDF–1, CXCL12)/CXCR4 was recently shown to be involved in metastasis of several types of cancer, as well as in certain inflammatory autoimmune disorders like rheumatoid arthritis. Therefore, an intense search is currently in progress to identify potent and specific CXCR4 antagonists that are able to block SDF–1–induced cell migration. Until recently, the most widely used method to study chemotaxis in vitro is the Boyden microchamber assay with microscopic counting of migrated cells at the lower surface of the membrane. This is a laborious and time–consuming technique which also does not allow further phenotypic characterization of the migrated cells. Here, we describe a more convenient chemotaxis assay using 24–well plate tissue culture inserts with 5 µm pore polycarbonate membranes (Transwell, Costar) and subsequent flow–cytometric counting and up to four–color phenotypic analysis of the migrated cells. In this system, we have evaluated the inhibitory effect of the CXCR4 antagonist AMD3100 on SDF–1–induced chemotaxis of different cell types, including CXCR4+ T– and B–lymphoid tumor cell lines. We also studied the chemotactic response of the different subsets of human peripheral blood lymphocytes to SDF–1. In addition, this assay was used to demonstrate the central role of SDF–1 and the therapeutic potential of CXCR4 inhibitors in the pathogenesis of murine autoimmune collagen–induced arthritis (CIA), a study model for rheumatoid arthritis in man. We found that SDF–1, which was shown to be present in the inflamed synovial joint tissue, strongly attracts CD11b+/Gr–1+/CXCR4+ neutrophils and that this migration is specifically inhibited by CXCR4 inhibitors.
The human immune response to pneumococcal capsular polysaccharides is dependent on the CD40–CD40L interaction.
A. Jeurissen1, G. Wuyts1, L. Boon2, J.L. Ceuppens3, X. Bossuyt1
1Experimental Laboratory Medicine, UZ Gasthuisberg, Leuven
2MacroZyme B.V., Amsterdam
3Experimental Immunology, UZ Gasthuisberg, Leuven
Corresponding author: axel.jeurissen@uz.kuleuven.ac.be
Introduction:
Streptococcus pneumoniae is a major cause of serious infections. Protection against S. pneumoniae is mediated by antibodies against capsular polysaccharides (caps–PS), which are classified as Thymus Independent type 2 (TI–2) antigens. T lymphocytes are not required for antibody production to TI–2 antigens, but they do have a regulatory effect. Until now, the molecular mechanisms of the T lymphocyte mediated modulation of the anti–caps–PS immune response are not fully understood.
Aim:
To study whether the human immune response to caps–PS is regulated by CD4(+) and/or CD8(+) T lymphocytes, and whether this T lymphocyte–mediated modulation is mediated by CD40L.
Methods:
Human peripheral blood mononuclear cells (PBMC) were cultured in vitro in the presence of caps–PS and anti–caps–PS IgM was measured using ELISPOT. For in vivo studies, SCID/SCID mice in which human PBMC were transferred i.p., were immunized with caps–PS (Pneumovax®) and after 14 days human anti–caps–PS IgM and IgG were measured by ELISA. To study the effect of the CD4(+) and CD8(+) T lymphocytes, depletion of cell fractions with Dynabeads was used. To study the CD40–CD40L interaction, PBMC from a patient with hyper IgM syndrome and a CD40L blocking monoclonal antibody (LL55) or a chimeric CD40 blocking (ch5D12) antibody was used.
Results:
We showed that in the absence of T lymphocytes, human B lymphocytes generate only a weak anti–caps–PS antibody response, which is strongly enhanced by CD4(+) T lymphocytes. Depletion of CD8(+) T lymphocytes from total lymphocytes resulted in an enhanced anti–caps–PS immune response, indicating an inhibitory role for CD8(+) T lymphocytes. Stimulation with caps–PS resulted in increased expression of CD40L. Using antagonist anti–CD40 and antagonist anti–CD40L monoclonal antibodies we found that the CD4(+) T lymphocyte–mediated stimulation as well as the CD8(+) T lymphocyte–mediated inhibition was dependent on the CD40–CD40L interaction. The crucial role of CD40L and T lymphocytes in the human immune response was further illustrated by the observation that CD4(+) T lymphocytes which lack CD40L, obtained from a patient with hyper IgM syndrome, were unable to enhance the immune response to caps–PS. Furthermore, CD4(+) T lymphocytes from cord blood, which, in contrast to adult CD4(+) T lymphocytes, did not express CD40L in response to stimulation with caps–PS, failed to stimulate the antibody response of adult B lymphocytes to caps–PS. These in vitro findings were confirmed by in vivo experiments, in which natural killer cell–depleted SCID/SCID mice were reconstituted with human mononuclear cells.
Conclusion:
The human immune response to caps–PS, which are considered TI–2 antigens, is markedly regulated by T lymphocytes and this non–cognate T lymphocyte–mediated modulation is dependent on CD40–CD40L interactions.
Development of a Cell Lysate Immunometric Assay (CLIA) for Heat Shock Protein 70 Detection
Rose Njemini1, Christian Demanet2 and Tony Mets1
1Geriatric Unit, Academic Hospital, Free University of Brussels (VUB), Laarbeeklaan 101, B–1090 Brussels, Belgium
2HLA and Molecular Hematology Academic Hospital, Free University of Brussels (VUB), Laarbeeklaan 101, B–1090 Brussels, Belgium
Corresponding author: rnjemini@vub.ac.be
Heat shock proteins (Hsp) are highly abundant proteins that play fundamental roles in promoting cellular survival and maintenance of normal cellular function. They are expressed in small quantities under normal physiological conditions, and their synthesis can be strongly induced in response to a plethora of stress signals including hyperthermia, physical stress, and various disease states. Several studies have shown the expression of Hsp, especially Hsp 70, during clinically relevant situations, and evidence is mounting that the ability to survive and adapt to severe systemic physiological stress is critically dependent on the ability of cells to mount an appropriate compensatory stress response. The need for a simple assay for determining Hsp 70 in cells has emerged with the increasing interest in therapeutic manipulation of Hsp for clinical trials. In this study we have developed a new technique for the investigation of intracellular Hsp 70. This cell lysate immunometric assay (CLIA) uses a combination of two distinct monoclonal antibodies that recognize different epitopes on the Hsp 70 molecule. A recombinant human Hsp 70 was used as the standard material. The detection range of the CLIA was 4 –4000 ng/mL. the intra– and interassay coefficients of variation were, on average, 5 and 12, respectively. The recovery varied between 81% and 116%. The Hsp 70 levels assayed after serial dilution of cell lysates varied linearly with dilution (between 97% and 120%). The reliability of the CLIA for the detection of Hsp 70 was assessed by comparing the values determined by flow cytometric procedure; these two sets of values showed a highly significant correlation (r=0.896, P<0.0001), indicating that the two methods are comparable. In conclusion, the described CLIA can be considered as a low–cost alternative of the flow cytometric technique that may be relevant in the understanding of disease conditions that are accompanied by Hsp 70 production.
Despite Inhibition of Hematopoietic Progenitor Cell Growth in vitro, the Tyrosine Kinase Inhibitor sti571 does not Impair Engraftment of Human cd34+ Cells into Nod/scidb2null Mice.
L.Pirson, F. Baron, O.Giet, N. Meuris, V. Bours, A. Gothot, G.Fillet, Y. Beguin Department of Medicine, Division of Hematology; University of Liège, Liège, B4000, Belgium.
Corresponding author: lpirson@student.ulg.ac.be
Purpose:
There is a strong interest for combining non myeloablative stem cell transplantation (NMSCT) and STI571 in order to maximize anti–leukemic activity against chronic myeloid leukemia (CML) and to reduce the high incidence of graft rejection observed after NMSCT in CML patients. However, as stem cell factor (SCF) plays a major role in the homing and the proliferation of hematopoietic stem cells (HSC), it remains to be demonstrated whether the inhibition of its receptor (c–kit) by STI571 does not impair hematologic recovery after hematopoietic stem cell transplantation (HSCT).
Material and methods:
CFU–GM and BFU–E assays were performed in Methocult H4435 supplemented with SCF, GM–CSF, IL–3 , IL–6, G–CSF and Epo. LTC–IC were performed in H5100 medium during 5 weeks and in Methocult H4434 during 2 weeks. The expression of VLA–4, VLA–5 and CXCR4, as well as the percentage of cycling cells (G2/M + S) were determined by FACS analysis. Migration towards SCF–1 conditioned medium was studied in BSA– and fibronectin (Fn)–coated transwells. For in–vivo experiments, NOD/SCID/B2mnull mice were irradiated (3 Gy) on day 0 and then i.v. inoculated with 6 x 105 human CD34+ cells. Mice were started on day 0 on a placebo or on STI–571 (150 mg/kg/day) by gavage and were sacrificed 6 weeks after xenotransplantation.
Results:
STI571, at 0.5, 1 and 2 µM decreased CFU–GM formation from mobilised peripheral blood cells by a mean of 12% (p=0.030), 22% (p=0.030) and 44% (p=0.001), respectively and BFU–E formation by 13% (NS), 41% (p=0.003) and 55% (p=0.008), respectively. When CD34+ were cultured 48 hours preliminary with STI571, at same concentrations, BFU–E and CFU–GM formation was not modified significantly. Similarly, STI571 didn’t significantly decrease the formation of LTC–IC. STI571 at 0.5, 1 and 2 µM decreased the number of CD34+ cells in cycle after 48 hrs culture in a SCF–containing medium by a mean of 30% (p=0.01), 46% (p=0.01) and 47% (p=0.003), respectively. STI571 did not alter VLA–4, VLA–5 or CXCR4 expression by CD34+ cells. Moreover, STI571 did not modify the ability of CD34+ cells to adhere to Fn or to migrate through BSA– and Fn–coated filters towards a medium containing SDF–1. Finally, human chimerism was 15+4% in NOD/SCID/B2mnull mice receiving placebo versus 36.3+0.2% in mice treated with STI571 (p=0.006). Overal survival was decreased when mice were treated with STI571. Histo–pathological studies showed anaemia in thighbone marrow of mice treated.
Conclusion:
Although progenitor growth and cell transit into cycle appear to be reduced by STI571, the drug does not alter the engraftment ability of human hematopoietic stem cells into NOD/SCID/B2mnull mice.
Zap–70 Evaluation by Flow Cytometry and Correlation with IgVH Mutational Status in B–CLL
B. Leus, A. Piette, I. Mollet, F. Van Bockstaele, A. Janssens, F. Offner, J. Philippé
University Hospital Ghent, Belgium
Corresponding author: jan.philippe@ugent.be
Introduction
Although generally regarded as an indolent disease, CLL has a considerable variability in clinical course. The prognostic significance of IgVH gene rearrangements is well established, those patients with unmutated IgVH genes having a worse prognosis than those with somatic mutations. Zap–70, a tyrosine kinase involved in T–cell receptor signalling, was found by gene expression profiling, to be expressed in the unmutated subgroup. The analysis by flow cytometry of the latter marker is much easier than the elaborate and costly sequencing technique. Therefore we evaluated the zap–70 expression by flow cytometry and compared these results with the mutational analysis.
Materials and Methods
23 B–CLL patients were included. In two patients analyses were performed repeatedly. 5.105 cells were fixed and permeabilized with use of the Fix&Perm kit (Caltag) and 1.5 µg of anti–zap–70 antibody (clone 2F3–2, Upstate Biotechnology) was used and further stained with GAM–PE. Then cells were washed and incubated with normal mouse serum for 5 minutes and finally staining with CD19–FITC, CD3PerCP and CD56–APC/CD5–APC. For each sample a D–value was established as a result of the Kolmogorov–Smirnov statistics comparing the zap–70 expression of T–cells (CD3+) versus the zap–70 expression of the B–CLL cells (CD19+/CD5+, or all CD19+). Flow cytometric results were compared with the IgVH mutational status.
Results: Seven patients had unmutated VH–genes (<2% mutations) and 16 patients showed a mutated profile. D–value was 0.81 +/- 0.07 (mean +/- SD) for the unmutated patients versus 0.89 +/- 0.06 in the mutated patients (p=0.01). As cut–off for the D–value several options were possible: 0.855 with one patient with D=0.92 having an unmutated profile and 5 patients with D<0.855 showing a mutated profile, or 0.815 with 3 misclassifications for each group (3 false positives and 3 false negatives). Of the three unmutated patients with ‘false high’ D–values one patient (D=0.92) has a stable disease during more than 30 years, another (D=0.84) has a stable disease for 10 years, and one patient was recently diagnosed.
Discussion
A cut–off value of 0.815 for D seems most promising. A prospective study is ongoing in order to further evaluate this flow cytometric assay.
Multiparameter Flow Cytometric Detection of Residual Neuroblastoma Cells.
Swerts K1,2, De Moerloose B1, Dhooge C1, Benoit Y1, Laureys G1 and Philippé J2
Department of Pediatrics1, Ghent University Hospital, Ghent, Belgium
Department of Clinical Chemistry, Microbiology and Immunology2, Ghent University Hospital, Ghent, Belgium
Corresponding author: jan.philippe@ugent.be
Introduction:
Neuroblastoma, the most common extracranial cancer in children, is derived from the neural crest. It shows a wide range of biologic, genetic and morphologic characteristics and exhibits a diverse clinical behavior. Approximately 40% of the neuroblastoma patients suffer from high risk stage 4 disease with bone marrow involvement. These children have a poor clinical outcome. The presence of neuroblastoma cells in bone marrow during therapy can predict relapse or clinical outcome. The detection of bone marrow involvement is thus critical for accurate staging and risk assessment. Moreover, myeloablative chemotherapy in combination with autologous stem cell transplantation is commonly used to treat children with stage 4 disease. Residual neuroblastoma cells can contaminate the stem cell product and their reinfusion can cause the recurrence of the disease. Several techniques can be used to detect the residual neuroblastoma cells. In this study, we developed a four color flow cytometric assay to evaluate minimal residual disease (MRD) in NB patients. The test is based on the combination of CD9, CD81, CD56 and anti–GD2. CD45 is used as a negative marker.
Materials and methods:
Spiking experiments with cells from a neuroblastoma cell line were used to evaluate the sensitivity of the technique. Ten bone marrow samples from adults without malignant disease were used as negative controls. Twelve tumor samples, twenty–eight bone marrow samples and two stem cell preparations from twenty–two neuroblastoma patients were analyzed and the percentage of GD2+/CD81+/CD56+/CD45– and CD9+/CD81+/CD56+/CD45– cells was calculated. The samples were taken at diagnosis or during therapy. Finally, we compared the flow cytometric results with the results of an immunocytochemical APAAP technique. This assay has a sensitivity of one neuroblastoma cell in 105–106 mononuclear cells.
Results:
In a dilution experiment, the multiparameter flow cytometric assay reached a sensitivity of one neuroblastoma cell in 104 –105 mononuclear cells. All tumor samples were CD9+, CD81+, CD56+ and CD45–. All except one showed GD2 disialoganglioside expression. No positive cells were found in the negative controls. Concordant results between the flow cytometric assay and the immunocytochemical APAAP test were found in 26 out of 30 bone marrow samples and stem cell preparations. Four samples were negative and 22 samples contained residual neuroblastoma cells. In one stem cell preparation and three bone marrow samples, discordant results were found. These samples were taken during therapy and contained only a few residual neuroblastoma cells.
Conclusion:
The multiparameter flow cytometric assay using a combination of CD9, CD81, CD56, anti–GD2 and CD45 can be used to detect residual neuroblastoma cells in bone marrow, peripheral blood or stem cell preparations. It allows the detection of one NB cell in 104–105 mononuclear cells. Consequently, the assay is less sensitive than the immunocytochemical test. However, further investigations are necessary to determine the clinical relevant detection limit.
Analysis of Donor–recipient Chimerism after Nonmyeloablative Stem Cell Transplantation (NMSCT) : A Method Allowing CD3+ T Cell Selection in Patients PB with Low WBC and Lymphocytes Counts.
Nicole Schaaf–Lafontaine°, Roland Greimers+, Jean–Claude Grosdent°, Jeanine Comte°, Michel Humblet° and Christian Herens*.
Laboratory of Hematology°, Laboratory of Pathology+ and Laboratory of Cytogenetics*. CHU of Liege, Belgium.
Corresponding author: Nicole Schaaf–Lafontaine, my747@yahoo.com
A nonmyeloablative human stem cell transplantation (HSCT) approach has been adopted with the aim of reducing the graft versus–host disease (GVHD)and obtaining efficient graft versus leukaemia (GVL) effects. The evaluation of the chimerism during haematological reconstitution is performed on day 28, 100, 180 and 365. The samples of patients whole peripheral blood (PB) and of selected CD3+ lymphocytes and CD13+/CD33+ granulocytes are analysed. According to the kind of transplant, fluorescence in situ hybridisation (FISH) is used, to detect X and Y chimerism for recipients of sex–mismatched transplants recipients, and polymerase chain reaction (PCR) of polymorphism microsatellite regions for recipients of sex–matched transplants. The positive selection of peripheral CD3+ lymphocytes and of CD13+ granulocytes respectively is performed following hemolysis, according to standard technique of fluorescence activated cell sorting.
This method becomes laborious, time consuming for several samples and of poor cell recovery for CD3+ lymphocytes at some delays of reconstitution because of low WBC and lymphocytes counts. This challenged us to develop another method allowing the CD3+ cells selection.
We get efficient numbers of lymphocytes by using a negative selection method with the CD3+ ‘cocktail’ of tetrameric antibody complexes (RosetteSep) from StemCell Technologies. One to 5ml of fresh heparinized blood samples are mixed with the cocktail reagent, then layered on Ficoll or Density Medium for Lymphocytes separations (DML) for standard density gradient centrifugation. Cells of interest are harvested from the interphase and washed twice. Following numeration, selected cells are phenotyped (CD3, CD13, CD19, CD56 and CD14). This approach offers great specificity, rapidity and efficient recovery of viable cells. The CD3+ selected are able to proliferate in vitro after PHA stimulation. The amount of selected CD3+ cells allows subsequent FISH or PCR analysis. In situations of pronounced WBC deficiency, by combination of both RosetteSep CD3 cocktail and granulocyte depletion cocktail, we can handle enough lymphocytes to perform chimerism analysis.
Efficient Removal of LoxP–flanked Genes by Electroporation of Cre–recombinase mRNA
Peter Ponsaerts1, Dave Van den Plas2, Viggo Van Tendeloo1, Dirk R. Van Bockstaele1,
Zwi N. Berneman1, Joseph Merregaert2
1Laboratory of Experimental Hematology, Faculty of Medicine, University of Antwerp, Antwerp University Hospital, Wilrijkstraat 10, B–2650 Edegem, Belgium
2Laboratory of Molecular Biotechnology, University of Antwerp, Universiteitsplein 1, B–2610 Wilrijk, Belgium
Correspondent author: Peter.Ponsaerts@uza.be
Cre–recombinase–mediated excision of LoxP–flanked DNA sequences is a powerful technology in transgenesis and in various gene therapy strategies. Current technologies to establish stably–transfected cells or cell lines involve co–selection for a drug selection marker or a fluorescent protein. However, if downstream applications for these transfected cells include animal experiments or human clinical trials, it might be advisable to remove the selection marker. For example, commonly used selection markers, e.g. neomycin resistance and fluorescent proteins, are antigenic targets for cytotoxic T–cells upon re–introduction of gene–modified cells in immune competent animals or humans. For these reasons, technologies were developed to remove undesired DNA sequences from stably–transfected cells. Introduction of LoxP–sites in DNA constructs for generation of stable clones has become a widely used application. Cloned cells can be screened for integration in undesired DNA sequences, in order to avoid insertional mutagenesis, resulting in uncontrolled division of cells in vivo. To remove loxP–flanked DNA sequences, current technologies are based on the transfection of plasmid DNA encoding Cre–recombinase or adenoviral–mediated transduction and expression of Cre–recombinase. With these transfection techniques, additional integration of DNA sequences into the genome cannot be excluded, which might result in undesired side effects, although this is likely to occur with low frequency.
We have developed a non–viral non–DNA technique for rapid and efficient excision of LoxP–flanked DNA sequences in bulk cell populations based on electroporation of mRNA encoding Cre–recombinase. The use of mRNA for transfection of cell lines and primary cells has previously been demonstrated and is considered as a safe, highly efficient, and clinical applicable alternative to DNA–mediated gene transfer. Moreover, electroporation of mRNA is not associated with the risk of additional DNA integration, and protein expression will rapidly disappear, thereby leaving cells in an untouched condition. We created an experimental model system, where human leukemic K562 cells were stably transfected with a DSRed reporter gene flanked by two LoxP sites. An EGFP coding sequence placed downstream of the DSRed reporter gene could only be transcribed when the coding sequence for DSRed was removed. These cells, described as K562–DSRed[EGFP] cells, were electroporated with in vitro transcribed mRNA encoding Cre–recombinase, in order to remove DSRed sequences. The presented data show recombination efficiencies, as measured by stable appearance of EGFP fluorescence and disappearance of DSRed fluorescence, reaching more than 80% in Cre–recombinase mRNA–electroporated K562–DSRed[EGFP] cells, as measured by flowcytometric analysis.
The Anti–HIV Activity of the CXCR4 Inhibitor AMD3100 Is Independent of the CXCR4/HIV Coreceptor Level
Katrien Princen1, Sigrid Hatse1, Kurt Vermeire1, Gary J. Bridger2, Renato T. Skerlj2, Erik De Clercq1, and Dominique Schols1
1Rega Institute for Medical Research, Katholieke Universiteit Leuven, B–3000 Leuven, Belgium. 2AnorMed, Langley, BC V2Y 1N5, Canada
Corresponding author: Katrien.Princen@rega.kuleuven.ac.be
The chemokinereceptor CXCR4 is the main coreceptor used by T–tropic X4 HIV–1 strains to infect its target T–cells. It has been proven that the CXCR4 expression level in T–cells is strongly upregulated by IL–4, a Th2–type cytokine that is secreted preferentially in HIV–infected patients in a later stage of disease. This results in an enhancement of HIV–1 replication in CD4+ T–lymphocytes. We have now evaluated the potency of the CXCR4 antagonist AMD3100 in PHA/IL–2– versus PHA/IL–4–activated T–cells in order to determine whether the compound has comparable CXCR4–antagonistic and anti–HIV–1 effects under these different cytokine treatments. We analysed the CXCR4 expression level and the dose–dependent inhibition of CXCR4 expression by AMD3100, by monitoring the binding of an anti–CXCR4 mAb (clone 12G5). We also determined SDF–1–induced intracellular calcium signaling and HIV–1 replication in these cells in the absence and presence of AMD3100. The CXCR4 expression level in PHA/IL–4–stimulated cells was much higher than in PHA/IL–2– stimulated cells. However, the potency of the bicyclam AMD3100 to block anti–CXCR4 mAb binding, SDF–1–induced intracellular calcium signaling and X4 HIV–1 replication, remained unchanged. Our data indicate that CXCR4 antagonists such as AMD3100 act independently of the HIV–1 coreceptor expression level. These compounds should therefore be useful in suppressing HIV–1 infection in all stages of the disease.
Flow Cytometry and Molecular Analysis: Integration of Both as an Improved and Robust Exploratory Tool in the Diagnosis of Some Unexplained Conditions.
Dirk R. Van Bockstaele, Goedele Caethoven, Eva Steel, Ann Van de Velde and Zwi N. Berneman, Antwerp University Hospital, University of Antwerp, Antwerp, Belgium
Corresponding author: Dirk.Van.Bockstaele@uza.be
Careful examination of flow cytometric listmode data acquired using an exploratory panel of triple combinations of monoclonal antibodies will in some cases provide the only first clue towards the presence of an aberrant hematopoietic cell population in samples of patients, who sometimes only have vague or unexplained complaints. A particular case in point is the
finding of a flow cytometric "bimodal" antigen density pattern (i.e. one cell population with normal density and one cell population with decreased density), especially regarding "important" antigens, eg. CD3, CD4, CD8, CD45 and surface kappa or lambda.
Several examples (two of which are further introduced below) will be presented that illustrate conditions where discrete abnormalities in flow cytometry were found in the absence of marked cytopenia or increased blood cell counts and were confirmed by molecular results.
In a female patient with itching persistent for 3 years, flow cytometric analysis showed bimodal CD4 antigen density on CD3+ T–lymphocytes.
Additional flow cytometric analysis demonstrated restricted TcR V–beta usage on CD4+ lymphocytes with decreased CD4 antigen density and TCR–gamma chain gene rearrangement analysis demonstrated a monoclonal T–cell population. Similar findings were made in the bone marrow. The diagnosis of a CD4+T–cell chronic lymphocytic leukemia was made, even if the patient never demonstrated (T–cell) lymphocytosis in blood or marrow.
Another female patient with fluctuating fatigue had normal blood cell levels, but a double– positive CD4+CD8+ T–cell population was found in blood and marrow. CD8 antigen density was decreased on these double–positive cells. Additional flow cytometric analysis disclosed that these cells were CD1–, CD2+, CD3+, CD7+, CD5+, CD56+ and CD16+, compatible with a NK/T–cell phenotype. Spectratype / immunoscope (GeneScan) analysis of TCR–gamma gene
rearrangements demonstrated a monoclonal population, compatible with a NK/T–cell neoplasm.
This illustrates that starting from a rapid exploratory flow cytometric examination, further measures can be taken to substantiate abnormal findings. Especially further molecular analysis (RT–PCR for fusion transcripts or PCR of IgH and TCR–gamma chain gene rearrangements for
clonality evaluation) can serve as an important confirmation of the flow cytometric findings. The combined presence of flow cytometric, cytological and molecular expertise within the same setting greatly facilitates and expedites the flow of information in that in many cases the left–over of the flow cytometric sample can be immediately used for DNA/RNA isolation, without having to request an additional sampling from the patient.
Quality Control Issues in Flow Cytometry (FCM) for the Immunophenotyping of Leukaemia's and Lymphomas: what is available, necessary, redundant and missing?
Dirk R. Van Bockstaele, Antwerp University Hospital, UA.
Corresponding author: Dirk.Van.Bockstaele@uza.be
A quarter of a century ago, the FCM technique was developed for and by the research laboratories in order to obtain a fast individual–cell measurement technique that is able to generate parameter distributions (in contrast to the bulk methods of the past) of physical, bio– and immunochemical properties of cell populations. The hydrodynamic focusing principle still
is the heart of the system while the instrumentation gradually increased in complexity.
This increased demand for multiparametric analysis is a consequence of the clinical relevance of evaluating the combined presence, absence and density of a myriad of antigens, that became detectable by the parallel availability of vast numbers of monoclonal antibody.
The perception of this clinical relevance led to the introduction of FCM within specialised oncohematological centres that had the necessary scientific, technical and financial back–up and where peoples' expertise increased in parallel with the increasing instrumentation complexity.
The outbreak of the HIV–epidemic triggered the introduction of FCM into the clinical laboratories, where it was somewhat misintroduced as a black–box technology in line with the chemistry and immunochemistry robots. Although the machines were externally simplified and looked like a pushbutton system, the complexity now was hidden in the software. The flow cytometer is the only machine in the clinical laboratory that is fully operator adjustable in order to accommodate a variety of applications. By this misintroduction one is prone to believe that the QA/QC measures, common in the field of automatic test results, could be directly implemented in the FCM field as well. This is a misconception that gives rise to a number of QA–habits that don't serve any purpose other than providing a false sense of security.
The habit of daily running standard particles to check on the instrument settings: the idea is that in this way one can check on the stability of the instrument settings. The reality is that these instrument setting are very robust and a weekly running of these particles will suffice to document on the only property that is detectable by this procedure: i.e. the (possible) long term drift.
The habit of running QC material before and after the running of a number of patients samples: the idea is that if the QC runs are acceptable, than the sample runs will be acceptable too. No QC material can however guarantee that the flow cytometer is functioning properly in between. Sudden, unexpected and transient flow disturbances that can temporarily impede on the laminar flow conditions can result in ruining the acquisition. A flow cytometric run is an experiment on its own, the quality of which should be checked within the run.
QC actions that will improve on the quality and correctness of the generated results rely on the within–run checking of the data and the scrutiny of the post–run (listmode) data file analysis. One has to be particularly very attentive to all events that appear to be biologically absurd or inconsistent. However in the field of immunophenotyping for haematological malignancies some aberrance's my pinpoint to the presence of an abnormal cell population. Two parameters that are available on modern machinery are heavily under–used and may, especially in the field of QC, be of utmost importance to discriminate between artefacts and malignant cell populations: i.e. time and pulse shape analysis.
Finally one has to be aware that the habit of relying on automatic (standard particle based) fluorescence compensation corrections can give rise to wrong results when confronted with extraordinary high fluorescence intensity data (which is not uncommon in the haematological field). One should use these patient samples as material to fine–tune the compensation corrections rather than relying on automatic routines and one should be aware of the
obsoleteness of quadrant statistics under these circumstances.
One should start to rethink QC issues depending on whether FCM is being used as a multipurpose machine for the immunophenotyping of haematological malignancies or as a machine dedicated for some specific FCM applications (such as CD4 count in HIV+ patients). The latter may fit in the "classical" thinking of automated test result QC, the former however is very much dependent on the expertise of the people involved and relies on a student/mentor type of relationship that is common in pathology and high complexity testing laboratories. Continuing education to newcomers in the field is of greater importance in assuring quality than any running of calibrators or standards. This should be preferably accomplished by independent organisations (such as the Belgian Society for Cytometry, BVC/ABC) under the auspices of official legislative organs (such as the Scientific Institute of Public Health, WIV/ISP).
Flow Cytometric Quantification of Cell Surface and Intracellular HLA Class I Antigen Processing Machinery Components.
Sonja Verheyden1, Mieke Roels1, Kathleen Stam1, Arend Mulder2, Frans Claas2, Soldano Ferrone3, Wim Renmans1 and Christian Demanet1.
1HLA laboratory, Academic Hospital VUB, Brussels, Belgium, 2Dept. of Immunohematology and Bloodtransfusion, Leiden University Medical Center, Leiden, the Netherlands, 3Dept. of Immunology, Roswell Park Cancer Institute, Buffalo, NY, USA.
Corresponding author: sonjaverheyden@hotmail.com
Downregulation of cell surface expression of human leukocyte antigen (HLA) class I antigens occur frequently in human malignant cells and may represent a mechanism of tumor escape from cytotoxic T lymphocytes (CTL) – mediated immune responses.
The HLA class I antigen processing machinery (APM) plays a crucial role in the processing and presentation of HLA class I restricted antigenic peptides. Therefore, it is possible that reduced expression of these molecules in tumor cells may affect surface expression of HLA class I molecules and subsequent recognition by CTL. Reverse transcriptase–polymerase chain reaction (RT–PCR) and Western blotting are widely used techniques for the detection of the expression of APM proteins. However, neither of them is suitable for a quantitative analysis. To circumvent this limitation, we have developed a quantitative method to detect intracellular HLA class I APM molecules in permeabilized cells by flow cytometry. The objective of this study was to optimize the intracellular staining of APM proteins allowing us to investigate the relationship between quantitative changes in the expression of these intracellular APM markers and cell surface HLA class I antigens in human leukemia.
Initially, we analyzed the cell surface expression of HLA–A, –B, –C at the allelic level and b–2 microglobulin of malignant (B–cells) and normal (T–cells) lymphocytes in B–lymphocytic leukemia patients by indirect multiparameter flow cytometry using specific monoclonal antibodies. These cells were subsequently analyzed for the intracellular expression of several members of the HLA class I APM such as LMP–2, –7, –10, delta, MB1, TAP1, TAP2, calnexin, calreticulin, ERp57 and tapasin. We coupled a fixation and permeabilization method with flow cytometric analysis to facilitate the penetration of specific monoclonal antibodies. A generalized protocol was developed for staining HLA I APM components intracellular by examining a spectrum of different fixation and permeabilization conditions that might had an impact on the optimal intracellular antigen detection. Fixing cells with 4 % paraformaldehyde followed by permeabilization in 0.1 % saponin gave optimal results for TAP1, LMP–2, delta, calnexin and calreticulin staining. The other monoclonal antibodies remained negative for unknown reasons. Furthermore, the fixation and permeabilization protocol preserved the antigenicity of cell surface markers allowing us to distinguish distinct cell populations.
In conclusion, we optimized a technique that permits the quantitative analysis of the expression of intracellular APM antigens at the single cell level, while preserving the antigenicity that allows identification of the target cell. The proposed cytometric methodology is important to supplement Western blotting or RT–PCR and can contribute to give more insights into downregulation mechanisms in tumors.
Genotyping Mutations/Polymorphisms in the Mannose–binding Lectin Gene Using 5’nuclease Assay in Combination with Minor–Groove Binder DNA Probes.
Erna Van Hoeyveld, Griet Duson*, Frans Houtmeyers, Marc Van Ranst, Norbert Blanckaert, Xavier Bossuyt
Department of Laboratory Medicine, Immunology, Gasthuisberg University Hospital, Belgium
*Applied Biosystems, Lennik, Belgium
Corresponding author: erna.vanhoeyveld@uz.kuleuven.ac.be
Structural point mutations in exon 1 at codons 52, 54 and 57 and promotor polymorphism at –221 and –550 bp of the mannose–binding lectin (MBL) gene are associated with increased susceptibility to various infectious diseases. We developed a genotyping method based on the 5’–nuclease or TaqMan® assay in combination with the use of minor groove binder (MGB) probes for screening of these mutations/polymorphisms.
The short length of these probes allows them to be used for the detection of mutations that are in close proximity to each other, as is the case for the structural mutations in exon 1 of the MBL gene. Results for structural mutations were comparable with results obtained with classical techniques such as restriction fragment length polymorphism and allele specific PCR. Polymorphisms in the promotor were confirmed by sequencing.
In conclusion, this assay is very useful for large–scale screening of point mutations/polymorphisms, even when they are in close proximity.