Specific adeno-associated virus serotypes facilitate efficient gene transfer into human and non-human primate mesenchymal stromal cells

Mesenchymal stromal cells (MSCs) show great promise for ex vivo gene and cell-mediated therapies. The immunophenotype and in vitro differentiation capacity of primary baboon MSCs was demonstrated to be near-identical to that observed in human MSCs. To optimize gene transfer efficiency, we compared the efficiency of serotypes 1, 2, 3, 4, 5, 6, and 8 of adeno-associated virus (AAV) vectors for their ability to mediate transduction of human and baboon MSCs. AAV serotype 2 vectors were the most efficient in transducing MSCs from humans and baboons. As a reference, human Ad293 cells were transduced with these seven AAV serotypes, and were found to have the highest transduction levels followed by baboon MSCs, and then human MSCs. The order of increasing transduction efficiency for the serotypes tested was similar for human and baboon MSCs, but was different for human Ad293 cells. The transduction efficiency of MSCs isolated from different individuals was comparable within the same species. We also demonstrated that baboon MSCs transduced with AAV serotype 2 vectors retain their potential to differentiate into adipocytes in vitro, and can incorporate into injured muscle tissue of NODSCID mice in vivo. We detected beta-galactosidase reporter gene expression in host muscle tissue for up to 9 weeks in this study, indicating engraftment of transduced baboon MSCs and sustained transgene expression in vivo.

Separation of adeno-associated virus type 2 empty particles from genome containing vectors by anion-exchange column chromatography

Adeno-associated virus (AAV) empty capsids typically co-purify with genome containing AAV2 vectors purified by column chromatography. This study describes a method to remove empty capsids from genome containing vector particles by anion exchange chromatography. The separation is based on the slightly less anionic character of empty particles compared to vectors. Detailed methods to achieve AAV2 vector purification and particle separation using cation exchange resin POROS 50HS followed by anion exchange resin Q-Sepharose(xl) are described. Chromatographic separation of AAV2 particles was achieved using gradients based on sodium acetate and ammonium acetate, and was optimal at pH 8.5. Efficient removal of particle surface nucleic acid impurities was found to be important to achieve good particle separation. In a large scale experiment performed using partially purified vector containing a mixture of 1.56 x 10(14)vg and 2.52 x 10(15) empty capsids as a starting material, the optimized anion exchange chromatography method resulted in a vector peak of 1.15 x 10(14)vg containing 0.25 x 10(14) empty capsids, corresponding to 74% vector yield and 86-fold reduction in empty capsids in the vector product.

Efficient AAV1-AAV2 hybrid vector for gene therapy of hemophilia

Adeno-associated virus (AAV) serotype 1 (AAV1) has been shown to be more effective than the well-studied AAV serotype 2 (AAV2) in muscle gene transfer. Replacement of amino acids 350 to 430 of AAV2 VP1 with the corresponding amino acids from VP1 of AAV1 resulted in a hybrid vector, termed AAV-221-IV, which behaved similarly to AAV1 in vitro and in vivo in muscle. Intramuscular injection of 1x10(11) vector particles per mouse of hybrid vector carrying a human FIX transgene in CD4 knockout mice resulted in an average level of human FIX in the plasma of 450 ng/ml, 4- to 10-fold higher than in mice injected with an AAV2 vector carrying the same transgene, and 80% of the transgene levels in animals treated with the same dose of AAV1. DNA analysis of injected muscle showed a 10-fold higher copy number after gene delivery by the hybrid vector compared with AAV2. A comparison of total DNA versus DNA from intact virus particles suggests a higher stability of hybrid virus particles. These results suggest that changes in the AAV capsid have an effect on virus-cell receptor interaction, and also influence trafficking and processing of the virus particle in the cell. This "hybrid vector" retains the heparin-binding sites of AAV2 and, therefore, can be purified by passage through a heparin-Sepharose column with the same efficiency as AAV2. When tested in vivo, either in CD4 knockout mice or in a hemophilic mouse model, the heparin-purified hybrid vector showed >10-fold higher activity than similarly purified AAV2. This demonstrates the utility of this hybrid vector in the performance of large-scale heparin column purification to generate a vector with a high expression profile for muscle-directed gene delivery. Initiation of clinical studies with this hybrid vector may be facilitated because it differs from AAV2 by only nine amino acids.

Successful transduction of liver in hemophilia by AAV-Factor IX and limitations imposed by the host immune response

We have previously shown that a single portal vein infusion of a recombinant adeno-associated viral vector (rAAV) expressing canine Factor IX (F.IX) resulted in long-term expression of therapeutic levels of F.IX in dogs with severe hemophilia B. We carried out a phase 1/2 dose-escalation clinical study to extend this approach to humans with severe hemophilia B. rAAV-2 vector expressing human F.IX was infused through the hepatic artery into seven subjects. The data show that: (i) vector infusion at doses up to 2 x 10(12) vg/kg was not associated with acute or long-lasting toxicity; (ii) therapeutic levels of F.IX were achieved at the highest dose tested; (iii) duration of expression at therapeutic levels was limited to a period of approximately 8 weeks; (iv) a gradual decline in F.IX was accompanied by a transient asymptomatic elevation of liver transaminases that resolved without treatment. Further studies suggested that destruction of transduced hepatocytes by cell-mediated immunity targeting antigens of the AAV capsid caused both the decline in F.IX and the transient transaminitis. We conclude that rAAV-2 vectors can transduce human hepatocytes in vivo to result in therapeutically relevant levels of F.IX, but that future studies in humans may require immunomodulation to achieve long-term expression.

ISBT Code 128 implementation at a regional blood center

BACKGROUND

Application specifications for ISBT 128 bar code symbology and the International Council for Commonality in Blood Bank Automation (ICCBBA) were created in 1994. By June 2000, the FDA considered ISBT 128 a standard for uniform labeling of blood and blood components. Our blood center initiated a change process for ISBT 128 implementation and "went live" in 2003.

STUDY DESIGN AND METHODS

The intention to adopt ISBT 128 symbology with hospitals was actively communicated in October 2001. A Codabar-ISBT label cross-reference book was developed, FDA approval for the fullface label format in April 2002 was requested, and FDA approval was received in March 2003. In December 2002, donor identification labels and number sets were ordered, and an integration test plan was subsequently developed with departmental process flowcharts for each of the nine affected departments. Each step was tested, the labeling changes were approved in May 2003, training was completed in June 2003, and ISBT bar code symbology was implemented on July 1, 2003. A written survey was sent to hospital transfusion services in April 2004.

RESULTS

Implementation went smoothly except for an unanticipated high rate of "no-reads" on some analyzers in the testing lab. The hospitals spent an average of 18 hours preparing for changes, 14 hours on validation, 4 hours on documentation and procedure development, and 8 hours on training.

CONCLUSION

ISBT bar code symbology was successfully implemented. Hospital transfusion services made some adjustments and, overall, readily accepted the new bar code symbology.

Identification of factors that contribute to recombinant AAV2 particle aggregation and methods to prevent its occurrence during vector purification and formulation

Aggregation of recombinant AAV2 results in reduced yield during purification and may have deleterious effects on vector transduction efficiency, biodistribution and immunogenicity following in vivo administration. Studies to elucidate the mechanism of vector aggregation and methods to prevent its occurrence are reported. In excipient screening studies, the sugars sorbitol, sucrose, mannitol, trehalose, or glycerol at concentrations of up to 5% (w/v), or surfactants Tween 80 or Pluronic F68, did not prevent aggregation. Aggregation was prevented by the use of various salts at concentrations corresponding to solution ionic strengths of >200 mM. AAV2 vectors purified by double cesium chloride gradient centrifugation, cation-exchange chromatography, or combined chromatography and gradient centrifugation each demonstrated a similar requirement for ionic strength to prevent aggregation. AAV2 vectors concentrated to 6.7 x 10(13) vector genome (vg)/mL in neutral-buffered isotonic saline resulted in 59+/-6.0% recovery of nonaggregated material compared to 96+/-4.4% recovery in an isotonic formulation with elevated ionic strength. The latter showed no aggregation following storage or after 10 freeze-thaw cycles at -20 degrees C. AAV2 vectors stored for an extended period in an elevated ionic strength formulation retained a high infectivity titer (13 vg/infectious unit) and transduction efficiency. Nuclease digestion of purified AAV2 vectors reduced aggregation, implicating trace amounts of vector surface nucleic acids in interparticle binding.

Infectious Titer Assay for Adeno-Associated Virus Vectors with Sensitivity Sufficient to Detect Single Infectious Events

A highly sensitive assay for determination of infectious titers of recombinant adeno-associated virus (AAV) by limiting dilution analysis is described. This assay is capable of detecting single infectious events and can therefore provide an absolute rather than relative measure of infectivity. The assay utilizes a HeLa-derived AAV2 Rep/Cap-expressing cell line, D7-4, grown in 96-well plates and infected with replicate 10-fold serial dilutions of AAV2 vectors in the presence of adenovirus type 5. Forty-eight hours after infection, vector genome replication is determined by quantitative PCR (Q-PCR). A linear relationship between vector genome input and replicated copy number (slope = 2670 copies per vector genome) was determined, enabling detection of one infectious event per well by Q-PCR. The observed binomial distribution of the end-point data confirmed that single infectious events could be detected, and allowed calculation of infectious titers by the Kärber method. Analysis of an AAV2 reference vector, AAV-hFIX16, in 21 independent determinations gave an average ratio of AAV vector genomes (VG) to infectious units (IU) of 8.3 +/- 4.2 VG/IU, a value close to the theoretical limit. No significant differences in vector particle-to-infectious unit ratios were observed between vectors purified by column chromatography (9.3 +/- 5.0 VG/IU, n = 7) and cesium chloride gradient ultracentrifugation (6.4 +/- 3.2 VG/IU, n = 7).

Phenotypic correction of a mouse model of hemophilia A using AAV2 vectors encoding the heavy and light chains of FVIII

Using separate adeno-associated viral 2 (AAV2) vectors to deliver the heavy and light chains of factor VIII (FVIII) we have overcome the packaging limitations of AAV, achieving phenotypic correction of hemophilia A in mice. AAV vectors were constructed that use a liver-specific promoter and the cDNA sequences of either the human or canine heavy and light chains of FVIII. After intraportal vein injection of these vectors in hemophilia-A mice, therapeutic to superphysiologic levels of active FVIII were achieved in plasma in a dose-dependent manner. Phenotypic correction of the bleeding diathesis was demonstrated by survival of all treated mice after tail clipping. Biochemical analysis demonstrated lower levels of heavy-chain (25- to 100-fold) compared with light-chain protein in the plasma of treated animals. Differences in gene transfer and transcription did not account for the differences in protein expression. We hypothesize that improvements in FVIII activity could be achieved by improvements in FVIII heavy-chain expression. This work demonstrates that cotransduction of liver with AAV vectors expressing the heavy and light chains of FVIII corrects hemophilia A in vivo, providing an alternative approach to the use of a single vector. This strategy may potentially be useful for other large therapeutic proteins that contain functionally distinct domains.

Sustained phenotypic correction of canine hemophilia A using an adeno-associated viral vector

Gene therapy for hemophilia A requires efficient delivery of the factor VIII gene and sustained protein expression at circulating levels of at least 1% to 2% of normal. Adeno-associated viral type 2 (AAV2) vectors have a number of advantages over other viral vectors, including an excellent safety profile and persistent gene expression. However, a major disadvantage is their small packaging capacity, which has hampered their use in treating diseases such as hemophilia A, cystic fibrosis, and muscular dystrophy, which are caused by mutations in large genes. Here we demonstrate that this can be overcome by using small regulatory elements to drive expression of a B-domain-deleted form of FVIII. The use of this vector for hepatic gene transfer in a canine model of hemophilia A resulted in the sustained (> 14 months) expression of biologically active FVIII. FVIII activity levels of 2% to 4% were achieved. These levels correlated with a partial correction in the whole-blood clotting time and cuticle bleeding time. In addition, immunoprecipitation analysis demonstrated the expression of canine FVIII of the predicted size in the plasma of injected animals. These data support the use of AAV2 vectors in human clinical trials to treat hemophilia A patients.

Recombinant adeno-associated virus: formulation challenges and strategies for a gene therapy vector

Recombinant adeno-associated virus (AAV)-based vectors capable of expressing therapeutic gene products in vivo have shown significant promise for human gene therapy. One challenge facing the field is the development of vector formulations to achieve optimal vector safety, stability and efficacy. Formulation challenges for AAV vectors can be divided into those relating to maintaining vector activity during purification and storage, and those relating to efficient target tissue transduction in vivo. AAV vectors are potentially susceptible to loss of activity through aggregation, proteolysis and oxidation, as well as through non-specific binding to product contact materials used for vector purification and storage. These deleterious changes need to be thoroughly characterized, and the conditions and excipients to prevent them need to be identified. For in vivo administration, major vector formulation challenges include optimization of efficiency and specificity of target tissue transduction, and the ability to overcome host immune responses.

Quantification of adeno-associated virus particles and empty capsids by optical density measurement

We show here that UV absorbance of denatured adeno-associated virus (AAV) vector provides a simple, rapid, and direct method for quantifying vector genomes and capsid proteins in solution. We determined the molar extinction coefficients of capsid protein to be 3.72 x 10(6) M(-1) cm(-1) at 260 nm and 6.61 x 10(6) M(-1) cm(-1) at 280 nm. For recombinant AAV vectors, extinction coefficients can be calculated by including the predicted absorbance of the vector DNA. Since the amount of empty capsids in purified vector preparations lowers the A(260)/A(280) ratio in a predictable manner, the vector genome (vg) and capsid particle (cp) titers in purified AAV vector preparations can be calculated from the absorbance at 260 nm and the A(260)/A(280) ratio. To validate this method, the vg and cp titers calculated by UV absorbance were compared with titers determined by quantitative (Q)-PCR and capsid ELISA. The vg titers determined by absorbance agreed well with titers determined by Q-PCR. The cp/vg ratio determined by the A(260)/A(280) method also correlated well with those determined by AAV capsid ELISA in conjunction with Q-PCR. This new method provides a simple and rapid means to determine AAV vg titers and the ratio of empty to full particles in purified virus preparations.

A limited number of transducible hepatocytes restricts a wide-range linear vector dose response in recombinant adeno-associated virus-mediated liver transduction

Recombinant adeno-associated virus (rAAV) vectors are promising vehicles for achieving stable liver transduction in vivo. However, the mechanisms of liver transduction are not fully understood, and furthermore, the relationships between rAAV dose and levels of transgene expression, total number of hepatocytes transduced, and proportion of integrated vector genomes have not been well established. To begin to elucidate the liver transduction dose response with rAAV vectors, we injected mice with two different human factor IX or Escherichia coli lacZ-expressing AAV serotype 2-based vectors at doses ranging between 4.0 x 10(8) and 1.1 x 10(13) vector genomes (vg)/mouse, in three- to sixfold increments. A 2-log-range linear dose-response curve of transgene expression was obtained from 3.7 x 10(9) to 3.0 x 10(11) vg/mouse. Vector doses above 3.0 x 10(11) vg/mouse resulted in disproportionately smaller increases in both the number of transduced hepatocytes and levels of transgene expression, followed by saturation at doses above 1.8 x 10(12) vg/mouse. In contrast, a linear increase in the number of vector genomes per hepatocyte was observed up to 1.8 x 10(12) vg/mouse concomitantly with enhanced vector genome concatemerization, while the proportion of integrated vector genomes was independent of the vector dose. Thus, the mechanisms that restrict a wide-range linear dose response at high doses likely involve decreased functionality of vector genomes and restriction of transduction to fewer than 10% of total hepatocytes. Such information may be useful to determine appropriate vector doses for in vivo administration and provides further insights into the mechanisms of rAAV transduction in the liver.

A comparison of the macrophyte cover and macroinvertebrate fauna at three sites on the River Kennet in the mid 1970s and late 1990s

In 1974-1976, baseline studies were carried out on the flora and macroinvertebrate fauna of the R. Kennet at two sites downstream of Marlborough (Savernake Upper and Lower) and at one site upstream of Hungerford (Littlecote). Simplified maps of each site, showing the cover of macrophytes, were obtained monthly between April 1974 and April/June 1976, and replicated quantitative samples of the macroinvertebrates were collected on the dominant macrophyte and on gravel in June 1974, and also in June and December 1975. As a consequence of two major droughts and increasing concern over water quality in the Upper Kennet in the 1990s, the studies recommenced in the summer of 1997 using the same sites and methodologies. Maps and macroinvertebrate samples were obtained in early July and December 1997 and in June of both 1998 and 1999. At the Savernake sites, mapping in summer 1997 confirmed what had been apparent for some years. That is, macrophyte cover (both Ranunculus and Schoenoplectus) was much lower than in the 1970s. In contrast, the site downstream at Littlecote retained a relatively high cover of Ranunculus, despite the drought. In late autumn 1997, phosphate stripping commenced at Marlborough Sewage Treatment Works, the drought ended and in addition, the spring of 1998 was unusually wet. Ranunculus recolonised both Savernake sites with remarkable speed by summer 1998 and retained this dominant position in 1999. Quantitative samples of macroinvertebrates collected on gravel and the dominant macrophyte at each of the three study sites indicated that there was no evidence of major loss of family richness between the 1970s and 1990s as a result of the low flows or enrichment. However, at Savernake (but not Littlecote) in summer 1997, the macroinvertebrate assemblage was affected by low flows and/or enrichment. This took the form of changes in the abundance of some families, with lentic forms being favoured in relation to some lotic families. Following the end of the drought, many macroinvertebrate families at Savernake showed a rapid response to the new conditions and the assemblages reverted to those expected in a fast-flowing cretaceous chalk stream. Continued monitoring through the next drought is advisable to provide a greater understanding of the interplay between water quality, the discharge regime, habitat quality (including macrophyte growth) and the response of the macroinvertebrate fauna.

Teratoma in desert grassland whiptail lizards (Cnemidophorus uniparens)

The clinical, gross necropsy, and histopathology findings in two unrelated desert grassland whiptail lizards (Cnemidophorus uniparens) with teratoma are described. The desert grassland whiptail is a parthenogenic lizard species with a polyploid chromosomal complement. The chromosome composition of the teratomas from these lizards was not determined.

Scaling-up production of recombinant AAV vectors for clinical applications

Recombinant adeno-associated virus (AAV)-based vectors capable of expressing therapeutic gene products in vivo have shown significant promise for human gene therapy. A major challenge for these applications is the development of processes to enable production of large quantities of AAV vectors and purification of material that is well characterized and appropriate for parenteral administration. Several cell culture systems have been developed for AAV vector production, and a limited number of these demonstrate the potential to generate AAV vectors at concentrations compatible with cost-effective large-scale production. Vector purification protocols, in particular those based on the use of scalable column chromatography, have concurrently been developed that demonstrate the potential to provide highly purified AAV vector preparations with high yield. These advances support the potential for AAV vectors as therapeutic agents for gene therapy.

Cholesterol granulomas in three meerkats (Suricata suricatta)

Cholesterol granulomas are uncommon pathologic lesions in animals, although they are important intracranial tumors in humans. This report describes cholesterol granulomas associated with multiple organ systems of three captive meerkats. In the most severe case, meerkat No. 1, the pathologic behavior of the cholesterol granuloma was unique in that it appeared to locally invade the cerebrum and calvarium, possibly contributing to neurological deficits observed antemortem. A review of other meerkat necropsies revealed incidental, asymptomatic cholesterol granulomas in organs of two other individuals, meerkat Nos. 2 and 3. Histologically, all lesions were composed of cholesterol clefts admixed with large, foamy macrophages containing hemosiderin, multinucleated giant cells, lymphocytes, plasma cells, and foci of mineralization. Hypercholesterolemia was documented in two of the three meerkats.

Characterization of platelet glycoproteins and platelet/endothelial cell antibodies in patients with thrombotic thrombocytopenic purpura

Platelets and sera from 12 patients with thrombotic thrombocytopenic purpura (TTP) and 12 healthy normal control subjects were examined. As determined by quantitative flow cytometry, prior to plasma exchange therapy platelet surface glycoprotein (GP) Ib levels were similar in TTP patients and normal controls (mean 20 188 and 20 226 molecules/platelet, respectively). Platelets from patients with TTP did, however, have significantly reduced levels of GPIIb/IIIa prior to plasmapheresis (mean 36 348 v 52 505 molecules/platelet in controls; P = 0.0004) and of GPIV (mean 13 321 v 26 212 molecules/platelet in controls; P = 0.0002). An increase in activated platelets, as determined by CD62 expression, was observed in 82% of patients. Increased platelet-associated immunoglobulins and/or complement was also seen in approximately 60% of the patients. In general, with return of platelet counts to normal levels following seven plasmaphereses, the above abnormalities were reversed, although often not to normal levels. Western blot analysis indicated the presence of antibodies reactive to platelet GPIV (88 kD) in 70% of pretreatment sera from patients with TTP; a similar band was observed in 80% of patient sera against microvascular endothelial cells. Immunofluorescence microscopic examination indicated the presence of antibody in pretreatment sera from patients with TTP to microvascular (73%) and large vessel (36%) endothelial cells. As measured by an indirect flow cytometric assay, pretreatment sera from 55% of patients with TTP were reactive with large vessel endothelial cells and 100% reacted with microvascular endothelial cells; reactivity was significantly greater against the microvascular endothelial cells (P = 0.0048) and was reduced following plasma exchange therapy. These results indicate abnormalities in platelet glycoprotein expression in TTP and suggest that anti-platelet and anti-endothelial cell antibodies play a role in the thrombocytopenia and vasculitis characteristic of this disorder.

Bioassays and biological stability

The development of the concept of <<Well Characterized Biotechnology Products>>, made possible through the power and detail afforded by modern biochemical and biophysical techniques, has resulted in questions being raised about the value of the more variable bioassay. However bioassays, particularly in a routine stability monitoring situation, continue to be useful and often provide unique information not obtained by other techniques. They have advantages in that they are relatively easy to perform and generally have better sensitivity than most structural techniques. They can be used to monitor for previously undetected changes - particularly those associated with conformational alterations, and can be used to monitor the effective combination of all individual changes.

Annexin II enhances cytomegalovirus binding and fusion to phospholipid membranes

A number of studies have suggested that the anionic phospholipid (anPL)-binding protein annexin II may play a role in cytomegalovirus (CMV) infection. Since annexin II has been shown to mediate aggregation and fusion of certain membranes, we investigated whether these properties could be exploited by CMV directly. The experiments showed that purified annexin II, but not the homologous protein annexin V (AnV), can mediate the binding of 35S-CMV (strain AD169) to anPL-coated microtiter wells. This association required Ca2+, could be titrated by varying either annexin II (apparent Kd = 4 x 10(-)8 M) or 35S-CMV, was inhibited by unlabeled CMV, and was observed for the heterotetrameric or monomeric form of annexin II. In experiments utilizing the fluorescence dequenching of octadecyl rhodamine incorporated into the CMV envelope, annexin II was furthermore found to enhance the rate of virus-anPL vesicle fusion. The observed fusion was dependent on the concentration of annexin II, Ca2+, and anPL and was mediated principally by the heterotetramer. Interestingly, AnV was observed to inhibit the effects of annexin II on CMV fusion but not binding to anPL, which indicates that annexin II enhances these processes by distinct mechanisms. The results presented here provide the first direct evidence that annexin II has the capacity to bridge CMV to a phospholipid membrane and to enhance virus-membrane fusion. These observations furthermore suggest that AnV may regulate the fusogenic function of annexin II.

Characterization of HIV-1-specific antibodies and HIV-1-crossreactive antibodies to platelets in HIV-1-infected haemophiliac patients

Sera from HIV-1-infected haemophiliacs were examined for human immunodeficiency virus (HIV) specific antibodies and for platelet crossreactive antibodies. Using HIV sepharose 4B affinity columns for serum absorption, antibodies against various HIV antigens, including HIV lysate. HIV-p24 and HIV-gp120, were eluted either by low or by high pH buffer. The eluates were examined by ELISA for HIV specificity and by flow cytometry for platelet crossreactivity. Two types of HIV antibodies could be eluted, i.e. acid-sensitive and alkaline-sensitive antibodies. HIV antibodies were obtained in 26/29 acid eluates and in 25/29 of the alkaline eluates from HIV-lysate columns; 96% (25/26) of the acid-eluted antibodies were HIV-specific but 48% (12/ 25) of the alkaline-eluted antibodies also showed crossreactivity to platelets. Of the 20 alkaline-eluted HIV-p24 antibodies, 40% (8/20) reacted specifically with HIV-p24 and 60% (12/20) were platelet crossreactive. In contrast, of the alkaline-eluted HIV-gp120 antibodies (n=17), 88% (15/17) were HIV gp120-specific and only 12% (2/17) were platelet crossreactive. Western blot analysis of platelets demonstrated that the anti-p24 antibodies recognized three bands with approximate molecular weights of 72000 to 95000. 69% of the serum antiplatelet antibodies showed platelet glycoprotein IIbIIIa specificity. Anti-HIV antibodies could be eluted from platelets. Hence, platelet crossreactive antibodies in HIV infection are primarily alkaline-sensitive and are associated predominantly with HIV p24 antibody; these antibodies may play a role in the immune thrombocytopenia of HIV-infected haemophiliacs.

Virus-associated idiopathic thrombocytopenic purpura

Two-thirds of children with acute idiopathic thrombocytopenic purpura (ITP) have a history of an infectious illness a few days to a few weeks before the onset of thrombocytopenia. In a subset of affected children, identification of a specific virus can be made, such as varicella zoster virus, rubella, Epstein-Barr virus, influenza, or human immunodeficiency type 1 virus, indicating an etiological role for preceding viral infection in these children with ITP. While inhibition of thrombopoiesis has been established to play a role in thrombocytopenia associated with infection with some viruses, it does not appear to play a major role in the etiology of most typical ITP cases. Rather, enhanced clearance of platelets by the reticuloendothelial system is considered to be, at least in part, responsible for the thrombocytopenia which occurs during the viremic phase of acute virus infection or which develops days to weeks following the virus illness. Molecular mimicry between viral antigens and host proteins has been implicated in a number of autoimmune phenomena, and may be involved in the enhanced platelet clearance in virus-associated ITP.

Coagulation initiated on herpesviruses

Herpesviruses have been previously correlated to vascular disease and shown to cause thrombogenic and atherogenic changes to host cells. Herein we show that even in the absence of cells, purified cytomegalovirus (CMV) and herpes simplex virus type 1 (HSV-1) and type 2 (HSV-2) can initiate thrombin production. Functional assays demonstrated that purified HSV-1 and HSV-2 provide the necessary phospholipid (proPL) for assembling the coagulation factors Xa and Va into prothrombinase, which is responsible for generating thrombin. These observations are consistent with our earlier studies involving CMV. The presence of proPL on all three herpesviruses was confirmed directly by flow cytometry and electron microscopy by using annexin V and factor Va, respectively, as proPL-specific probes. Of equal importance, we found that CMV, HSV-1, and HSV-2 were also able to facilitate factor Xa generation from the inactive precursor factor X, but only when factor VII/VIIa and Ca2+ were present. Monoclonal antibodies specific for tissue factor (TF), the coagulation initiator, inhibited this factor X activation and, furthermore, enabled identification of TF antigen on each virus type by flow cytometry and electron microscopy. Collectively, these data show that CMV, HSV-1, and HSV-2 can initiate the generation of thrombin by having essential proPL and TF activities on their surface. Unlike the normal cellular source, the viral activity is constitutive and, therefore, not restricted to sites of vascular injury. Thus cell-independent thrombin production may be the earliest event in vascular pathology mediated by herpesviruses.

The connectionist simulation of aphasic naming

The simulation of language disorders using interactive activation (IA) networks and connectionist systems is discussed. An existing IA account of aphasic naming is described, in which two network parameters (decay rate and connection strength) are varied to fit the error production of an aphasic patient. Fairly similar results can be obtained through modification of additional parameters, including the so-called "shared weight increase factor" linking lexical and semantic units. This leads us to consider simulation of aphasic naming using connectionist networks which do not require explicit variation of network parameters. A modular connectionist architecture is presented, in which semantic-lexical and phonological knowledge are instantiated using self-organizing Kohonen maps, while connections between them are implemented using Hebbian networks; a linear connectionist network (Madaline) is used to simulate nonword repetition. The Hebbian connections are lesioned in order to reproduce the patient's naming errors.

Flow cytometric analysis of platelets from children with the Wiskott-Aldrich syndrome reveals defects in platelet development, activation and structure

The pathophysiology of platelet dysfunction in the Wiskott-Aldrich immune deficiency syndrome (WAS) remains unclear. Using flow cytometry, we have characterized the functional properties of platelets from 10 children with WAS. Patients with WAS had thrombocytopenia, small platelets, increased platelet-associated IgG and reduced platelet-dense granule content. Levels of reticulated 'young' platelets were normal in the WAS patients. Although the mean numbers of platelet glycoprotein (GP) Ib, GPIIbIIIa and GPIV molecules per platelet appeared lower in WAS patients than in healthy controls, analysis of similar-sized platelets revealed the mean number of GPIb molecules per platelet to be comparable in patients and normal controls. Surface GPIIbIIIa and GPIV expression was, however, significantly lower on the WAS platelets than on normal platelets. Compared with normal platelets, WAS platelets showed a reduced ability to modulate GPIIbIIIa expression following thrombin stimulation. In addition, thrombin- and ADP-induced expression of CD62P and CD63 was defective in WAS platelets. Phallacidin staining of the WAS platelets revealed less F-actin content than in normal platelets. Together, these data suggest that the reduced platelet number and function in WAS reflects, at least in part, a defect in bone marrow production as well as an intrinsic platelet abnormality.

Characterization of platelet-reactive antibodies in children with varicella-associated acute immune thrombocytopenic purpura (ITP)

Biochemical analyses were performed on blood samples obtained from two children (P1, P2) who presented with acute immune thrombocytopenic purpura (ITP) following a recent varicella zoster virus (VZV) infection. Patient sera had antibodies that were reactive with normal blood-group O platelets as measured by flow-cytometric assay. Western blot analysis of electrophoretically separated normal blood-group O platelets under reducing and non-reducing conditions demonstrated that these sera were reactive with platelet antigens of approximately 50 and approximately 110 kD, respectively. These 50/110 kD antigens were not reactive with seven sera from acute ITP patients whose illness was not preceded by VZV infection, with serum from a patient with a prior history of VZV and no thrombocytopenia, nor with normal healthy control sera. VZV antibodies (IgG and IgM), isolated from patient sera by affinity chromatography using immobilized purified VZV glycoproteins, were found to bind to gel-filtered autologous platelets and with normal blood-group O platelets, as analysed by flow cytometry. No binding was observed using antibodies similarly prepared from healthy volunteer sera. To investigate their ability to sensitize platelets to complement activation, affinity-purified VZV antibodies were incubated with platelets and then with purified complement components C1 and 125 I-labelled C4. Platelets reacted with VZV-specific antibodies from the two patients and showed increases of 2.3-2.4-fold of platelet-surface deposition of 125 I-C4b, compared to controls. These data provide evidence that virus-specific antibodies occurring in children with varicella-associated acute ITP cross-react with normal platelet antigens, and may contribute to platelet clearance.

Applications of flow cytometry in the analysis of blood leukocytes

Flow cytometric analysis of blood leukocytes is currently used for both routine clinical measurements as well as for cutting edge research applications. This technology has enabled rapid and accurate determination of leukocyte antigens and quantitative analysis of leukocyte subsets, tests of leukocyte function, determination of the presence of antineutrophil and antilymphocyte antibodies in plasma and on cells, measurement of CD34+ hematpoietic stem cells in peripheral blood and bone marrow samples, measurement of apoptosis, and detection of virus-infected leukocytes. This review will focus on the use of the flow cytometer for investigations of blood leukocytes in transfusion medicine.

Analysis of platelets by flow cytometry

Transfusion medicine has become a multi-disciplinary field with many recent technical developments and the flow cytometer has had a significant impact in transfusion medicine, especially at the level of platelet immunobiology. Many routine tests in platelet immunology are now performed by flow cytometry laboratories, including assessment of platelet-associated allo- and auto-antibodies and complement components. Platelet analysis by flow cytometry has been applied to detection of platelet antigens, platelet surface-bound proteins, platelet activation, measurement of reticulated platelets, intracellular calcium studies, and the measurement of platelet microparticles in vivo and in vitro. This review will focus on the use of the flow cytometer in these applications in investigations of platelet immunology.

Host cellular annexin II is associated with cytomegalovirus particles isolated from cultured human fibroblasts

A significant amount of host cellular annexin II was found to be associated with human cytomegalovirus isolated from cultured human fibroblasts (approximately 1,160 molecules per virion). This composition was established by four different analytical approaches that included (i) Western blot (immunoblot) analysis of gradient-purified virions with a monoclonal antibody specific for annexin II, (ii) peptide mapping and sequence analysis of virus-associated proteins and proteins dissociated from virus following EDTA treatment, (iii) electron microscopic immunocytochemistry of gradient-purified virions, and (iv) labeling of virus-associated proteins by lactoperoxidase-catalyzed radioiodination. These results indicated that annexin II was primarily localized to the viral surface, where it bound in a divalent cation-dependent manner. In functional experiments, a rabbit antiserum raised against annexin II inhibited cytomegalovirus plaque formation in human foreskin fibroblast monolayers in a concentration-dependent manner. Cumulatively, these studies demonstrate an association of host annexin II with cytomegalovirus particles and provide evidence for the involvement of this cellular protein in virus infectivity.

An investigation into the use of hypertext as a user interface to taxonomic keys

In an attempt to overcome some of the drawbacks of existing taxonomic keys, we have experimented with a computerized version of a key, changing the interface to the key, but leaving the information contained in the key essentially unchanged. This is in contrast to most previous work on the use of computers in taxonomy, where work has concentrated on alternative identification techniques and techniques for manipulating taxonomic data. Hopkin's (1991) Key to the Woodlice of Britain and Ireland was converted (including the diagrams and colour photographs), into a hypertext document using the Unix Guide hypertext system. An experiment was conducted where test subjects with a wide range of experience of taxonomic keys and computing skills were asked to identify woodlouse specimens using the paper and hypertext forms of the keys. The results showed that test subjects were more likely to obtain an identification with the hypertext version of the key, they found it easier to use and preferred using it to the paper version of the key. However, there were indications that test subjects were more likely to obtain a misidentification using the hypertext key than the paper version of the key. Further work is needed to discover the possible causes of the increased frequency of misidentifications, and ways in which the usability of hypertext keys can be further enhanced.

C1q binding properties of monomer and polymer forms of mouse IgM mu-chain variants. Pro544Gly and Pro434Ala

The effect of replacing proline with alanine at position 434 in the C mu 3 domain (P434A) and with glycine at position 544 in the C mu 4 domain (P544G) of the mu-chain of mouse IgM has been studied. The P434A substitution results in the loss of measurable complement-mediated cytolytic activity (CML) and a decrease in the association rate constant at low ionic strength (mu = 0.06), that results in a diminished Ka for C1q binding to P434A IgM bound to haptenated cells (0.4 x 10(9) M-1). Binding of C1(qr2s2) could not be detected. In contrast, replacement of proline at 544 had no measurable effect on the cytolytic or C1q/C1 binding properties of the polymeric molecule, supporting the view that the C mu 3 domain is important in C1q binding and CML. The secreted monomeric subunit of P544G was not able to mediate CML. Also, whereas hapten-bound P544G polymer bound C1q with a functional affinity of 1.5 x 10(9) M-1 at low ionic strength (mu = 0.06), similar to that observed with wild-type polymer (1.7 x 10(9) M-1) and wild-type IgG monomer (4.7 x 10(9) M-1), no C1q binding was detected with the P544G IgM monomer. This could not be attributed to differences in glycosylation. Inasmuch as the P544G mutation per se had no effect on the C1q binding properties of the polymer, we conclude that unlike IgG, aggregation does not sufficiently enhance the avidity of IgM monomer to enable it to activate complement. Augmentation of the site must occur during polymerization or when the IgM binds to Ag.

C1q binding properties of monomer and polymer forms of mouse IgM mu-chain variants. Pro544Gly and Pro434Ala

The effect of replacing proline with alanine at position 434 in the C mu 3 domain (P434A) and with glycine at position 544 in the C mu 4 domain (P544G) of the mu-chain of mouse IgM has been studied. The P434A substitution results in the loss of measurable complement-mediated cytolytic activity (CML) and a decrease in the association rate constant at low ionic strength (mu = 0.06), that results in a diminished Ka for C1q binding to P434A IgM bound to haptenated cells (0.4 x 10(9) M-1). Binding of C1(qr2s2) could not be detected. In contrast, replacement of proline at 544 had no measurable effect on the cytolytic or C1q/C1 binding properties of the polymeric molecule, supporting the view that the C mu 3 domain is important in C1q binding and CML. The secreted monomeric subunit of P544G was not able to mediate CML. Also, whereas hapten-bound P544G polymer bound C1q with a functional affinity of 1.5 x 10(9) M-1 at low ionic strength (mu = 0.06), similar to that observed with wild-type polymer (1.7 x 10(9) M-1) and wild-type IgG monomer (4.7 x 10(9) M-1), no C1q binding was detected with the P544G IgM monomer. This could not be attributed to differences in glycosylation. Inasmuch as the P544G mutation per se had no effect on the C1q binding properties of the polymer, we conclude that unlike IgG, aggregation does not sufficiently enhance the avidity of IgM monomer to enable it to activate complement. Augmentation of the site must occur during polymerization or when the IgM binds to Ag.

Prothrombinase assembly on an enveloped virus: evidence that the cytomegalovirus surface contains procoagulant phospholipid

In contrast to quiescent cells, we currently report that purified cytomegalovirus (CMV), strain AD169, constitutively expresses phosphatidylserine (PS)-like procoagulant activity. Initial evidence for this came from the observation of a CMV-dependent decrease in factor Xa clotting times. In a purified system, the assembly of a functional complex between factor Xa and the cofactor Va to form prothrombinase was found to be dependent on the addition of CMV. The corresponding dense bodies (DB) and noninfectious enveloped particles had similar activity. Quantification of the total virion and DB phospholipid, and comparison of prothrombin conversion rates to experiments conducted using known concentrations of PS-containing vesicles showed that 8.5% and 7.2% of the CMV and DB phospholipid, respectively is procoagulant. Direct binding studies of 125I-labeled factor Xa, active site-blocked factor Xa, or the zymogen factor X, with virions or DB showed a single class of Ca(2+)-dependent sites with dissociation constants in the order of 10(-7) MOL/l. Immunogold electron microscopy confirmed the specificity of the CMV/factor Xa association. Cumulatively, these data suggest that the CMV surface contains the necessary procoagulant phospholipid for coagulation enzyme complex assembly. This may enable CMV (and possibly other enveloped viruses) to bypass an important physiologic regulatory mechanism for the production of thrombin.

An endothelial cell-surface form of annexin II binds human cytomegalovirus

Human cytomegalovirus was shown to bind to human umbilical vein endothelial cells in a specific, saturable and calcium-dependent manner (Kd = 7.9 pM (4 degrees C), 6469 virus binding sites/cell). Affinity adsorption of detergent-prepared lysates of surface-radiolabeled endothelial cells to virions resulted in the identification of cell-derived proteins of approximate M(r) 36,000 and 32,000 that bound cytomegalovirus. Protein sequencing of peptides obtained by cyanogen bromide cleavage demonstrated that the 36 kDa protein corresponded to human annexin II, and the 32 kDa protein was likely a degradation product. Purified annexin II was demonstrated to bind directly to virions (Kd = 57 nM, 688 annexin II binding sites/virion). These results provide evidence that an endothelial cell-surface form of annexin II acts as a receptor for cytomegalovirus, and indicate a previously undescribed role for annexin II.

Identification and partial characterization of a low affinity metal-binding site in the light chain of tetanus toxin

Tetanus toxin was shown to contain a metal-binding site for zinc and copper. Equilibrium dialysis binding experiments using 65Zn indicated an association constant of 9-15 microM, with one zinc-binding site/toxin molecule. The zinc-binding site was localized to the toxin light chain as determined by binding of 65Zn to the light chain but not to the heavy chain after separation by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transfer to Immobilon membranes. Copper was an efficient inhibitor of 65Zn binding to tetanus toxin and caused two peptide bond cleavages in the toxin light chain in the presence of ascorbate. These metal-catalyzed oxidative cleavages were inhibited by the presence of zinc. Partial characterization of metal-catalyzed oxidative modifications of a peptide based on a putative metal-binding site (HELIH) in the toxin light chain was used to map the metal-binding site in the protein.

Formation of covalent C3b-tetanus toxin complexes: a tool for the in vitro study of antigen presentation

A novel method is described for the formation and purification of covalent complexes between the complement component C3b and an antigen (tetanus toxin, TT), using purified proteins in fluid phase. C3b is generated in situ by tryptic cleavage of C3 after co-precipitation of C3 and TT in the presence of polyethylene glycol. Various parameters were analysed to optimize complex formation; under conditions which minimized the formation of covalent C3b multimers, 30% and 8% respectively of C3b and TT were incorporated into covalent one-to-one complexes which were purified using gel filtration chromatography. The linkage was localized between the alpha' chain of C3b and either the H or L chain of TT; it required the in situ formation of C3b and was partially destroyed by 1 M hydroxylamine. Spontaneous dissociation of the complex could be partly avoided by HgCl2, a thiol reagent which inhibits the esterase-like activity of bound C3b. These findings suggest the involvement of the reactive carbonyl of nascent C3b with hydroxyl groups of TT. Such C3b-TT complexes provide a defined tool to analyse the influence of antigen-bound C3b on antigen addressing and intracellular processing by antigen-presenting cells.

Proteolytic fragmentation of tetanus toxin by subcellular fractions of JY, a B lymphoblastoid cell line

Proteolysis of 125I-labelled tetanus toxin by subcellular fractions from an Epstein-Barr-virus-transformed B lymphoblastoid cell line, JY, was investigated. Fractions enriched in lysosomes and plasma membranes cleaved the toxin molecule at several sites, with a pH optimum of 5.5. N-Terminal sequence analysis of Mr-81,000, -45,000 and -35,000 proteolytic fragments indicated cleavage of the Asp-460-Leu-461, Asp-872-Glu-873 and Ile-1013-Thr-1014 peptide bonds, all sites located within the heavy chain of the toxin molecule. Additional sites near the C-terminus of the heavy chain, giving rise to low-Mr peptides, were implicated. The toxin light chain was more resistant to proteolysis. A similar pattern of fragmentation was observed with tetanus toxin biosynthetically radiolabelled with 14C-labelled amino acids, showing that the proteolysis was not an artifact caused by iodination. The proteolytic activity was inhibited by the serine proteinase inhibitor di-isopropyl phosphorofluoridate, thiol-blocking proteinase inhibitors N-ethylmaleimide and iodoacetamide, and by EDTA. These results represent a preliminary characterization of the processing in vitro of tetanus toxin by an antigen-presenting cell line.

C1 binding by mouse IgM. The effect of abnormal glycosylation at position 402 resulting from a serine to asparagine exchange at residue 406 of the mu-chain

We have previously shown that IgM-Asn406, a mutant IgM which has asparagine in place of the serine which is normally found at position 406, also has an abnormally glycosylated mu-chain and is defective in complement-dependent cytolysis. Here we show by analyzing cyanogen bromide fragments from normal and mutant mu-chains that the site of abnormal glycosylation is at the neighboring position, Asn402. The cytolytic defect was shown to be due to impaired C1 binding. At physiological ionic strength, the C1 binding defect was estimated to be 12-fold, which correlates well with the measured defect in cytolytic activity; also, the severity of the defect in C1 binding by the mutant protein decreases with decreasing ionic strength. Kinetic studies showed that the difference in affinities is due to a proportional difference in the association rate for C1q. By comparing IgM made in the presence and absence of deoxymannojirimycin, we show further that the defect in cytolytic activity derives mostly from the abnormal oligosaccharide.

Some unusual features of mycobacteriosis in the cichlid fish Oreochromis mossambicus

Adult, aquarium-reared Oreochromis mossambicus (Tilapia mossambica), which were naturally infected with Mycobacterium marinum, displayed non-healing skin ulcers and other clinical signs considered to be typical of piscine mycobacteriosis. However, in addition, they frequently had melanotic foci in the skin and spleen, due to the presence of pigment cells surrounding the cutaneous and splenic inflammation. Such melanotic foci have never been reported in response to mycobacteriosis. All fish also have variable numbers of melanomacrophages which appeared to replace pancreatic acini. The relationship of the pancreatic melanomacrophages to the pathogenesis of mycobacteriosis is uncertain, but such lesions may have contributed to the chronic cachexia associated with this case.

Chronic inflammatory cells with epithelial cell characteristics in teleost fishes

Certain cells that participate in the chronic inflammatory response of teleost fishes have many features typical of epithelioid cells of mammals. Such features include high metabolic activity, frequent phagolysosomes, and cytoplasmic interdigitations between adjacent cells; however, the epithelioid granulomas formed in response to certain diseases in teleost fishes also have several features associated with epithelial cells. Cases of ulcerative mycosis or acid-fast bacterial infection in Atlantic menhaden (Brevoortia tyrannus), fungal infection in silver perch (Bairdiella chrysoura), and mycobacteriosis in Mozambique tilapia (Oreochromis mossambicus) had epithelioid cells that were joined together by well-formed desmosomes with tonofilaments. "Mature granulomas" of the ulcerative mycosis-infected menhaden stained positively for cytokeratin, a cytoskeletal protein that is considered to be highly specific for epithelial cells. The consistent presence of these heretofore unrecognized epithelial features suggest that they may be characteristic of certain types of cells participating in piscine chronic inflammation.

A mitochondrial intergenic mutation affecting processing of specific yeast mitochondrial transcripts

The mutation in the temperature-conditional mit- mutant h56, mapped previously to the var1 gene region of Saccharomyces cerevisiae mitochondrial DNA, results in a specific inhibition of var1 protein synthesis in cells incubated at the non-permissive temperature, 36 degrees C (1). We have now characterized the mutation present in mutant h56 by DNA sequencing and found it to be an A to T transversion located 109 nucleotides upstream of the var1 reading frame. Two spontaneous revertants of mutant h56 restore the parental strain sequence at residue -109, confirming that this single base change within the 5'-untranslated region of the var1 mRNA is responsible for defective synthesis of the var1 protein. A comparison of var1 transcripts in the parental and mutant strains has shown that the mutation specifically blocks formation of var1 mRNA at 36 degrees C and leads to accumulation of precursor transcripts. Expression of the oli1 gene, co-transcribed with the var1 gene in primary transcripts, is not affected. It is concluded that the mutation in mutant h56 alters the secondary structure of the precursor RNA, inhibiting an endonucleolytic cleavage required to generate the 5' end of var1 mRNA.

Fibrous dysplasia of the spine

Fibrous dysplasia affecting the spine is unusual. A further 11 cases are reported and the radiological features are described. The complications and difficulties with the diagnosis of this condition are discussed.

Spinal cord compression caused by dual pathology: a close shave with Ockham's razor

A case of spinal cord compression in an oncology patient is presented. The compression was caused by minimal expansion of a vertebral body involved by a metastatic deposit impinging on a previously asymptomatic lipomatous spinal cord tumour. Nuclear magnetic resonance imaging clearly demonstrated both the vertebral metastasis and the intramedullary and extramedullary components of the lipomatous tumour in a single noninvasive investigation.

C1 binding by murine IgM. The effect of a Pro-to-Ser exchange at residue 436 of the mu-chain

We have examined a defect in complement activation in a mutant trinitrophenyl-binding pentameric murine monoclonal IgM which has serine replacing the proline normally found at position 436 in the protein. The mutant protein showed equivalent hapten binding but a 100-fold decreased ability to initiate complement-dependent lysis of trinitrophenyl-coupled erythrocytes at physiological ionic strength (mu = 0.15). C4b deposition mediated by the mutant protein was impaired to a similar degree. C1 bound by the mutant protein showed C1s to C1-s conversion, suggesting normal activation. When measured at reduced ionic strength (mu = 0.06), the C1 and C1q binding affinity of the mutant protein was approximately one-half that of the wild type. However, the C1 binding affinity of the mutant protein showed a greater dependence upon ionic strength such that at physiological ionic strength we estimate a 50-fold lower C1 binding affinity for the mutant molecule. Kinetic studies suggested that this difference in affinity was largely attributable to differences in association rates. In addition, a fixed proportion of the mutant molecules showed no C1 binding. We conclude that the defect in complement activation occurs at the level of C1 binding. Our data support a role for the C mu 3 domain (residues 340-440) in C1 binding by IgM.