A human monoclonal antibody inhibiting partially factor VIII activity reduces thrombus growth in baboons.

BACKGROUND: The inhibitory activity of an anti-factor VIII (FVIII) antibody can be modulated through glycosylation of the antigen binding site, as has recently been described. This offers the opportunity to develop an optimized anticoagulant agent targeting partial FVIII inhibition. OBJECTIVES: We investigated in non-human primates the antithrombotic activity, pharmacokinetics,and pharmacodynamics of a human monoclonal antibody, Mab-LE2E9Q, inhibiting FVIII activity partially. METHODS: The ability of Mab-LE2E9Q to prevent thrombosis was evaluated in baboons after administration of 1.25 and 5 mg kg(-1) antibody or saline as a single intravenous (i.v.) bolus. Thrombus development was recorded in expansion ('venous') and in Dacron ('arterial') thrombosis chambers incorporated in an extracorporeal arteriovenous shunt implanted between the femoral vessels 1 h, 24 h and 7 days after the administration of Mab-LE2E9Q. RESULTS: Mab-LE2E9Q reduced thrombus growth to a similar extend 1 h, 1 day and 1 week after administration of the antibody. Ex vivo pharmacodynamic analysis indicated that the evaluation of the residual FVIII activity was strongly dependent on the type of FVIII assay and on the phospholipid concentration in the assay. No significant difference in bleedings was observed between animals treated with Mab-LE2E9Q or with saline. CONCLUSIONS: Understanding the role of glycosylation in FVIII inhibition by a human monoclonal antibody allowed selection of an antibody inhibiting only moderately FVIII activity while significantly reducing thrombus development in a baboon extracorporeal model. As that antibody did not increase the bleeding tendency, it may represent a novel type of a long-acting antithrombotic agent with an optimal safety/efficacy profile.

Regional control of chromatin organization by noncoding roX RNAs and the NURF remodeling complex in Drosophila melanogaster.

Dosage compensation in Drosophila is mediated by a histone-modifying complex that upregulates transcription of genes on the single male X chromosome. The male-specific lethal (MSL) complex contains at least five proteins and two noncoding roX (RNA on X) RNAs. The mechanism by which the MSL complex targets the X chromosome is not understood. Here we use a sensitized system to examine the function of roX genes on the X chromosome. In mutants that lack the NURF nucleosome remodeling complex, the male polytene X chromosome is severely distorted, appearing decondensed. This aberrant morphology is dependent on the MSL complex. Strikingly, roX mutations suppress the Nurf mutant phenotype regionally on the male X chromosome. Furthermore, a roX transgene induces disruption of local flanking autosomal chromatin in Nurf mutants. Taken together, these results demonstrate the potent capability of roX genes to organize large chromatin domains in cis, on the X chromosome. In addition to interacting functions at the level of chromosome morphology, we also find that NURF complex and MSL proteins have opposing effects on roX RNA transcription. Together, these results demonstrate the importance of a local balance between modifying activities that promote and antagonize chromatin compaction within defined chromatin domains in higher organisms.

Antinuclear antibody testing in obstetric patients.

OBJECTIVES: To assess possible associations between the presence of antinuclear antibodies (ANAs) and pregnancy outcome in order to determine the significance of this test in obstetric practice. METHODS: A case-control study was performed on 408 patients admitted to an obstetric high care unit and on whom ANA testing was consecutively performed. The study group consisted of 46 patients who tested positive for ANAs and a control group of 92 patients who tested negative for ANAs. In addition to demographic data, indications for admission and pregnancy outcome were compared between the two groups. RESULTS: Of the 46 patients with a positive ANA result, 28 had an antinuclear pattern, 13 an anticytoplasmic pattern and 5 an antinuclear and an anticytoplasmic pattern. No significant differences were observed between the two groups (ANA-positive and negative) with regard to demographic data, indication for admission, clinical and laboratory data, and pregnancy outcome. The patients were also tested for anticardiolipin antibodies, and significantly more patients with severe pre-eclampsia tested positive (24% versus 4.7%, p = 0.01). No difference in HIV status and presence of autoantibodies was found between the two groups. CONCLUSION: The presence of ANAs was not associated with adverse pregnancy outcome. Therefore routine patient testing for ANAs in an obstetric high-care unit is not recommended.

Tramtrack69 interacts with the dMi-2 subunit of the Drosophila NuRD chromatin remodelling complex.

dMi-2, the ATPase subunit of the Drosophila nucleosome remodelling and histone deacetylation (dNuRD) complex, was identified in a two-hybrid screen as an interacting partner of the transcriptional repressor, Tramtrack69 (Ttk69). A short region of Ttk69 is sufficient to mediate this interaction. Ttk69, but not the Ttk88 isoform, co-purifies with the dNuRD complex isolated from embryo extracts. dMi-2 and Ttk69 co-immunoprecipitate from embryonic extracts, indicating that they can associate in vivo. Both dMi-2 and Ttk69 co-localize at a number of discrete sites on polytene chromosomes, showing that they bind common target loci. We also demonstrate that dMi-2 and Ttk interact genetically, indicating a functional interaction in vivo. We propose that Ttk69 represses some target genes by remodelling chromatin structure through the recruitment of the dNuRD complex.

Tramtrack controls glial number and identity in the Drosophila embryonic CNS.

Neurons and glia are often derived from common multipotent stem cells. In Drosophila, neural identity appears to be the default fate of these precursors. Stem cells that generate either neurons or glia transiently express neural stem cell-specific markers. Further development as glia requires the activation of glial-specific regulators. However, this must be accompanied by simultaneous repression of the alternate neural fate. I show that the Drosophila transcriptional repressor Tramtrack is a key repressor of neuronal fates. It is expressed at high levels in all mature glia of the embryonic central nervous system. Analysis of the temporal profile of Tramtrack expression in glia shows that it follows that of existing glial markers. When expressed ectopically before neural stem cell formation, Tramtrack represses the neural stem cell-specific genes asense and deadpan. Surprisingly, Tramtrack protein levels oscillate in a cell cycle-dependent manner in proliferating glia, with expression dropping before replication, but re-initiating after S phase. Overexpression of Tramtrack blocks glial development by inhibiting S-phase and repressing expression of the S-phase cyclin, cyclin E. Conversely, in tramtrack mutant embryos, glia are disrupted and undergo additional rounds of replication. I propose that Tramtrack ensures stable mature glial identity by both repressing neuroblast-specific genes and controlling glial cell proliferation.

Regulated nuclear export of the homeodomain transcription factor Prospero.

Subcellular distribution of the Prospero protein is dynamically regulated during Drosophila embryonic nervous system development. Prospero is first detected in neuroblasts where it becomes cortically localized and tethered by the adapter protein, Miranda. After division, Prospero enters the nucleus of daughter ganglion mother cells where it functions as a transcription factor. We have isolated a mutation that removes the C-terminal 30 amino acids from the highly conserved 100 amino acid Prospero domain. Molecular dissection of the homeo- and Prospero domains, and expression of chimeric Prospero proteins in mammalian and insect cultured cells indicates that Prospero contains a nuclear export signal that is masked by the Prospero domain. Nuclear export of Prospero, which is sensitive to the drug leptomycin B, is mediated by Exportin. Mutation of the nuclear export signal-mask in Drosophila embryos prevents Prospero nuclear localization in ganglion mother cells. We propose that a combination of cortical tethering and regulated nuclear export controls Prospero subcellular distribution and function in all higher eukaryotes.

Nutritional status of preschool children in informal settlement areas near Bloemfontein, South Africa.

OBJECTIVE: To determine the nutritional status and household resources of preschool children. DESIGN: A cross-sectional survey. SETTING: : Two informal settlement areas, Joe Slovo (JS) and JB Mafora (JBM) in Mangaung, near Bloemfontein, South Africa. SUBJECTS: Preschool children (<72 months) of a randomly selected sample of households in JS (experimental) (n = 162) and JBM (control) (n = 186) were included. Standard methods were used to obtain household and care-giver particulars, weight and height measurements, blood and stool samples, and 24-hour dietary recalls. RESULTS: Breast-feeding and dietary intake in the two areas were nearly similar; breast-feeding was continued for 12 months and longer. Although the children's total protein intake was sufficient, their energy intake was low. A low median intake of micronutrients prevailed, including iron, zinc, calcium, niacin, riboflavin, thiamine and vitamins C, B6, A and D. The prevalence of being underweight (JS = 19.8%; JBM = 18.8%), stunted (JS = 29%; JBM = 21. 5%) and wasted (JS = 6.5%; JBM = 3.7%) were fairly similar in both areas, as well as the prevalence of marginal vitamin A deficiency, anaemia, iron deficiency and parasite infestations. No significant associations could be found between household and nutritional status indicators, probably due to the small number of well-nourished children and the generally poor household situation of the participants. CONCLUSIONS: The generally poor nutritional status and environmental conditions emphasize the urgency of intervention for these children.

Measurement of plasma volume using 99Tc(m)-labelled DMP-HSA.

Red cell volume (RCV) and plasma volume (PV) measurements are performed routinely in nuclear medicine departments to diagnose a number of haematological disorders. Currently, 125I-HSA is used as a plasma tracer and 99Tcm-labelled red cells to determine red cell volume. 125I-HSA is not always readily available, leading to inconvenience for patients and medical practitioners. Due to the availability of 99Tcm in nuclear medicine departments, the use of albumin labelled with 99Tcm was investigated. A new 99Tcm-human serum albumin labelling kit (99Tcm-DMP-HSA) was developed by Verbeke and supplied for use in this study. The main aim of the study was to investigate the use of 99Tcm-DMP-HSA for PV determination. Secondly, the feasibility to determine red cell and plasma volume simultaneously using 99Tcm as radionuclide in both instances was investigated. Fourteen healthy volunteers were enrolled in the dual-phase study. During the first study, 99Tcm-DMP-HSA was used as tracer to calculate PV (PV1a) after intravenous administration. Subsequently, 99Tcm-labelled red cells were administered and the PV (PV1b) and RCV (RCV1) were calculated. The second study was repeated within 2 weeks using the conventional method. 125I-HSA and 99Tcm-labelled red cells were administered simultaneously. The PV (PV2) and RCV (RCV2) were calculated. We found that the redistribution of 99Tcm-DMP-HSA is faster than that of 125I-HSA; therefore, the plasma counts obtained at different times were back-extrapolated to time zero for plasma volume calculations. The mean values for the different calculated PVs were 2964+/-470 ml for PV1a, 3006+/-623 ml for PV1b and 3001+/-530 ml for PV2, the reference PV. The confidence intervals indicate no significant differences between plasma volumes PV1a and PV2 and plasma volumes PV1a and PV1b. The mean calculated RCV1 was 2130+/-322 ml; that of RCV2 was 2128+/-353 ml. The difference between RCV1 and RCV2 was not significant. Our results indicate that 99Tcm-DMP-HSA could be used for plasma volume calculation. Red cell and plasma volumes can be calculated simultaneously using 99Tcm as radionuclide in both cases.

Determination of the dosage of recombinant hirudin to inhibit arterial thrombosis in baboons.

Recombinant hirudin, a potent and direct inhibitor of thrombin, effectively inhibits platelet-dependent thrombosis. Our aim was to establish the plasma concentration at which r-hirudin expresses its optimal antithrombotic effect. We measured the extent of inhibition of (111)In-labeled platelet deposition onto 0.6 cm(2) segments of Dacron vascular grafts. These grafts were incorporated as extension segments into exteriorized permanent femoral arteriovenous shunts in baboons. In six control studies a mean of 1.99 +/- 0.26 x 10(9) platelets were deposited at the end of 120 min. In the treatment studies, a thrombus was allowed to form for 10 min in six animals. Treatment for 30 min with r-hirudin at dosages of 140, 70, and 35 microgram/kg/min, but not 14 microgram/kg/min, dose dependently interrupted platelet deposition. The relationship between the percent inhibition of platelet deposition caused by r-hirudin and the plasma concentration of hirudin was exponential (i.e., % Inhibition = 95(1-e(0.23 x [r-hirudin])) (R(2) = 0.76). From this, we estimated that 50% inhibition of platelet deposition will occur at a plasma concentration of approximately 3.3 microgram r-hirudin/mL and 80% at 8.1 microgram/mL. The relationship between the inhibition of platelet deposition and the plasma concentration of hirudin makes it possible to estimate the dose of hirudin that will result in a given level of inhibition of platelet deposition.

The effect of monoclonal anti-human-platelet antibodies on platelet kinetics in a baboon model: IgG subclass dependency.

We assessed the in vivo effect of six intact anti-human antiplatelet antibodies of two major IgG subclasses on platelet kinetics in baboons. Five of the six antibodies tested caused thrombocytopenia of varying degree when injected at a precalculated threshold value. An agglutinating IgG1 antibody (MA-8L4A12) caused a long-lasting, mild thrombocytopenia with a predominant uptake of radiolabelled platelets in the spleen, while the four IgG2 antibodies tested (MA-13G8E1, MA-2M5A6, MA-21K2E8 and MA-22M10) caused a severe, transient thrombocytopenia with uptake of platelets in the liver. Two of the IgG2 antibodies (MA-13G8E1 and MA-2M5A6) caused platelet activation and aggregation in vitro, whilst the other two did not elicit a platelet aggregation response. The platelet survival time was shortened with all five of the thrombocytopenia-inducing antibodies, while only one antibody (MA-2M5A6) had a significant effect on the bleeding time. This study indicates that the IgG subclasss may be a determining factor in the outcome of platelet sequestration in immune-induced thrombocytopenia.

Covalent modification of the transcriptional repressor tramtrack by the ubiquitin-related protein Smt3 in Drosophila flies.

The ubiquitin-related SUMO-1 modifier can be covalently attached to a variety of proteins. To date, four substrates have been characterized in mammalian cells: RanGAP1, IkappaBalpha, and the two nuclear body-associated PML and Sp100 proteins. SUMO-1 modification has been shown to be involved in protein localization and/or stabilization and to require the activity of specialized E1-activating and E2 Ubc9-conjugating enzymes. SUMO-1 homologues have been identified in various species and belong to the so-called Smt3 family of proteins. Here we have characterized the Drosophila homologues of mammalian SUMO-1 and Ubc9 (termed dSmt3 and dUbc9, respectively). We show that dUbc9 is the conjugating enzyme for dSmt3 and that dSmt3 can covalently modify a number of proteins in Drosophila cells in addition to the human PML substrate. The dSmt3 transcript and protein are maternally deposited in embryos, where the protein accumulates predominantly in nuclei. Similar to its human counterpart, dSmt3 protein is observed in a punctate nuclear pattern. We demonstrate that Tramtrack 69 (Ttk69), a repressor of neuronal differentiation, is a bona fide in vivo substrate for dSmt3 conjugation. Finally, we show that both the modified and unmodified forms of Ttk69 can bind to a Ttk69 binding site in vitro. Moreover, dSmt3 and Ttk69 proteins colocalize on polytene chromosomes, indicating that the dSmt3-conjugated Ttk69 species can bind at sites of Ttk69 action in vivo. Altogether, these data indicate a high conservation of the Smt3 conjugation pathway and further suggest that this mechanism may play a role in the transcriptional regulation of cell differentiation in Drosophila flies.

Sites of elimination and pharmacokinetics of recombinant [131I]lepirudin in baboons.

Lepirudin has a short half-life, and only 50-60% of the intravenously administered dose is excreted by the kidneys. The fate of the remainder is unknown. We designed a study to determine the fate of this lepirudin. In each of six baboons, [131I]lepirudin was given intravenously as a bolus or infused over 30 min, 24 h apart. The in vivo redistribution of [131I]lepirudin was determined and quantified by scintillation camera imaging. In all studies, the half-life of [131I]lepirudin, as determined from the disappearance of radioactivity, was 21 +/- 3 min. The half-life determined from the disappearance of lepirudin, measured by the Ecarin Clotting Time (ECT) method, was similar at 23 +/- 8 min. Results obtained with the labeled lepirudin are therefore comparable with those obtained using the plasma concentration of lepirudin. When lepirudin was administered as a bolus, the half-life was 18 +/- 4 min, and lepirudin was cleared from the plasma at a rate of 42 +/- 12 mL/min and by the kidneys at 23 +/- 2 mL/min. Following infusion over 30 min, the half-life and total and renal clearances were not significantly different. In both studies, between 50 and 60% of the administered lepirudin was excreted by the kidney. Studies on sacrificed baboons showed that appreciable amounts of lepirudin were present in the bile, indicating the liver as a contributor to the elimination of lepirudin.

r-Hirudin inhibits platelet-dependent thrombosis during cardiopulmonary bypass in baboons.

BACKGROUND: Systemic anticoagulation is required during cardiopulmonary bypass (CPB) to inhibit the activation of platelets, the coagulation system and ultimately thrombus formation. Unfractionated heparin is most commonly used, but it is neither entirely safe nor completely effective. The use of protamine sulphate to reverse the anticoagulant effect further complicates the use of heparin. The clinical need for a heparin substitute is therefore obvious. We evaluated the efficacy of r-Hirudin, a potent and specific inhibitor of thrombin, as anticoagulant in a baboon model of cardiopulmonary bypass. METHODS: Ten baboons, divided into two groups of five each, were used. The one group received 0.7 mg/kg r-Hirudin as a bolus before CPB was started, followed by a constant infusion of 1.4 mg/kg/hr for the 90 min of CPB. The other group received a bolus of 2.5 mg/kg heparin before the start of CPB, followed by maintenance dosages to maintain the activated clotting time (ACT) >400 sec. RESULTS: Adequate anticoagulation was obtained with both anticoagulants. Haemodilution due to priming the extracorporeal system with Ringer's lactate and appropriately anticoagulated donor blood, was equivalent in both groups. During CPB with heparin, but not with hirudin, there was a significant increase in the number of circulating platelet aggregates, thrombin-antithrombin (TAT) complexes and 111In-labelled platelet accumulation in the oxygenator. After the initial decrease in platelet count due to haemodilution, it further decreased significantly during CPB with heparin but remained relatively constant when r-Hirudin was used. CONCLUSIONS: Our results strongly suggest that r-Hirudin is superior to heparin especially with respect to its inhibitory effect on platelet dependent thrombogenesis caused by the biomembranes of the oxygenator.

Haematology outreach clinics in the Free State and northern Cape.

OBJECTIVE: Evaluation of haematology outreach clinics in the Northern Cape and Free State. DESIGN: Retrospective analysis of records from March 1994 to February 1996. SETTING: Central South Africa is sparsely populated. Consultants from Bloemfontein held outpatient clinics in hospitals (with laboratories) in Bethlehem, Kimberley and Kroonstad. SUBJECTS: 117 patients with suspected haematological disease. MAIN OUTCOME MEASURES: Input measures (population, number of clinics and costs), process measures (patient numbers, patients per clinic, new consultations per clinic, patients' domicile, how they were referred, types of diagnoses and number of patients with non-haematological disorders) and output measures (attrition, changes in attendance and savings). MAIN RESULTS: The 84 clinics that were held, with 636 consultations, did not cost the State anything. Only 6% of the 117 patients had no haematological problem. Sixty-eight per cent had chronic haematological neoplasms. In Kimberley most of the patients came from Kimberley Hospital, while most of the patients at the other clinics were referred via Bloemfontein. There was only a 10% attrition rate and only one-third of patients were referred to Bloemfontein. We saved paying patients an estimated R21,260 in transport costs, while saving the State R172,992 by seeing patients at secondary, instead of tertiary, hospitals. CONCLUSIONS: It is cheaper to send a doctor to an outreach clinic than to refer patients to a central facility, provided there is enough work for a doctor at the clinic. It costs the State much less for patients to be seen at a secondary than a tertiary hospital. Positive spin-offs include academic stimulation of doctors and laboratories in the periphery, with more appropriate referrals to teaching hospitals. Weaknesses include poor availability of expensive drugs at the clinics and lack of standardised records. By commuting to outreach clinics, specialists can greatly reduce health expenditure and spread it from tertiary to lower levels. At the same time more patients have access to their services.

Haemostatic profile of the San (Bushmen) relocated to Schmidtsdrif.

OBJECTIVE: To document the routine haemostatic variables of a group of San relocated from Namibia to South Africa. DESIGN: Cross-sectional study done in two stages. SETTING: Schmidtsdrif military camp in late 1990 and early 1991. SUBJECTS: Healthy adult San volunteers: 31 males and 54 females from the Vasakela and Barakwena groups in 1990; 135 males from the Vasakela group in 1991. The subjects were all soldiers or their dependants. MAIN OUTCOME MEASURES: The following tests were performed: activated partial thromboplastin time, prothrombin time, thrombin time, fibrinogen and coagulation factors V, VII, VIII, IX, X, XI and XII. The results were compared with a Western population reference group (N = 50). MAIN RESULTS: Almost all the haemostatic variables were statistically significantly lower than those of the reference group. The mean derived fibrinogen concentration in the plasma in the first stage of the study (1990) was significantly higher, but this reverted to normal during the second stage (1991), perhaps reflecting a general improvement in health. CONCLUSIONS: Even though the San are one of the best studied groups of indigenous people, this is the first published report on their haemostatic condition. The generally lower levels of haemostatic variables may reflect the lower prevalence of cardiovascular disease in the San. The population needs to be followed up as they westernise.

Plasma volume expansion with Haemaccel does not impair haemostasis during reduction mammaplasty.

OBJECTIVES: An in vivo study under well-controlled conditions was undertaken to determine the effect of Haemaccel, a colloidal plasma volume expander, on normal haemostasis. METHODOLOGY: Twenty patients, who were admitted for reduction mammaplasty, were included in this study. A standardised anaesthesia protocol was followed with all patients. Ten patients received 500 ml Haemaccel and 10 controls received 1,500 ml Ringer's lactate, a crystalloid solution. The solutions were administered intravenously during surgery over a period of 30-40 minutes. Standardised clinical observations and haematological tests were done at the following time intervals: after anaesthesia but before infusion of the plasma substitute, immediately after infusion was completed, and 20, 40 and 60 minutes after infusion. RESULTS: The blood pressure, pulse rate and O2 saturation levels were not influenced by the treatment given. Haemodilution was similar for the two patient groups. The platelet count and plasma levels of fibrinogen decreased in parallel with haemodilution. Thereafter the platelet count gradually increased to pre-infusion counts at 60 minutes. The prothrombin time (PT), activated partial thromboplastin time (aPTT), thrombin time (TT) and platelet aggregation in response to adenosine diphosphate (ADP) and collagen were not affected by the plasma volume expander given. Arachidonic acid-induced aggregation decreased significantly after Ringer's lactate was given but did not change when Haemaccel was given. The bleeding time was prolonged slightly, but not significantly, from 7.4 +/- 1.6 minutes to 8.8 +/- 1.6 minutes with Ringer's lactate and from 6.9 +/- 2.0 to 9.7 +/- 3.7 minutes with Haemaccel. CONCLUSIONS: We could not find any scientific evidence that Haemaccel affects haemostasis; neither does it increase bleeding relative to Ringer's lactate.

Transient interruption of arterial thrombosis by inhibition of factor Xa results in long-term antithrombotic effects in baboons.

Recombinant tick anticoagulant peptide (r-TAP) is a potent and specific inhibitor of activated coagulation factor X which effectively interrupts in vivo arterial thrombosis during treatment. It is, however, uncertain if it also affects thrombosis after treatment is stopped. This was tested in a baboon model of arterial thrombosis where platelet deposition onto Dacron vascular graft segments, inserted as extensions into permanent femoral arteriovenous shunts, was measured. The baboons were intravenously treated with 10 micrograms/kg/min (low dose, aPTT = 39 +/- 1 s) and 25 micrograms/kg/min (high dose, aPTT = 58 +/- 2 s) r-TAP for two hours. During treatment the r-TAP inhibited thrombin formation and dose-dependently interrupted platelet deposition onto the graft segment. This effect lasted for up to two hours after treatment with the low dose. Following treatment with the high dose, the graft segments were kept in place for 53 h. After treatment was stopped, platelets again deposited, but at a much lower rate than in control studies. Maximum deposition was approximately 38% lower than in the control studies. Total platelet deposition over 55 h, calculated as the area under the deposition curve, was approximately 40% (p < 0.05) less than in the control studies. A significant shortening in the mean platelet life span and an approximately 15-fold increase in thrombin-antithrombin III complexes during the first 31 h indicated that the thrombus surface remained thrombogenic and that the effect of r-TAP was transient. We have shown that 2 h of treatment with a full antithrombotic dose of r-TAP markedly reduced both the rate of platelet deposition after treatment was stopped and the total number of platelets deposited over 55 h. This was in spite of the finding that the antithrombotic effect of r-TAP was transient.

Immunoglobulin G sub-class concentrations in South African adults: ethnic differences and reference ranges.

Serum samples from healthy adult volunteers (n = 149) were selected at random from disputed paternity cases, laboratory staff and volunteers attending clinical trials. Total immunoglobulin G (IgG) and IgG sub-class (IgGSc) concentrations were determined by a radial immunodiffusion technique (RID). Standard statistical analyses were used to determine differences between groups. Reference ranges of IgGSc concentrations were calculated on the resultant groups of data. Total IgG and IgGSc concentrations in men and women of the same racial group were similar, except for IgG4 which was slightly higher in white males than in white females (median: 0.36 g/L vs 0.20 g/L respectively). IgGSc concentrations were higher in blacks than in whites (median values: IgG: 17.1 vs 12.1 g/L; IgG1: 11.1 vs 7.6 g/L; IgG2: 4.3 vs 3.2 g/L; IgG3: 1.2 vs 0.90 g/L respectively) with the exception of IgG4 which was similar in both groups (median: 0.29 g/L). It would appear that IgGSc values differ among the ethnic groups. Ethnicity must therefore be considered when calculating reference ranges. The reference ranges for the IgG sub-classes in the two ethnic groups are intended for use in our laboratory and in others in South Africa that use the RID technique.