Increased incidence of antiphospholipid antibodies in left ventricular assist system recipients

BACKGROUND

Antiphospholipid antibodies are associated with thrombosis. Because thromboembolic complications are often observed in recipients of a left ventricular assist system, we questioned if antiphospholipid antibodies were present in these patients. We report results from 10 patients who received a Novacor left ventricular assist system.

METHODS

Serum samples were collected before left ventricular assist system placement and weekly thereafter until discharge after cardiac transplantation. Samples were tested for IgG, IgA, and IgM antiphosphatidylserine, anticardiolipin, and antiphosphatidylethanolamine using an enzyme-linked immunosorbent assay.

RESULTS

Development of phospholipid-binding plasma protein-dependent antiphospholipid antibodies was observed in 9 of the 10 patients. Before placement of the assist system, 3 patients had IgG antiphospholipid antibodies, and 9 were positive after placement. None had IgA antiphospholipid antibodies before placement, whereas 5 seroconverted for IgA after placement. One patient had IgM antiphospholipid antibodies before placement, and 1 additional patient became positive after placement. In patients with a preexisting antibody, increased titers and additional specificities developed subsequent to placement.

CONCLUSIONS

All but 1 patient showed development of phospholipid-binding plasma protein-dependent antiphospholipid antibodies after left ventricular assist system placement.

Prevalence and heterogeneity of antiphosphatidylethanolamine antibodies in patients with recurrent early pregnancy losses

OBJECTIVE

To describe the prevalence of antiphospholipid antibodies to both anionic and zwitterionic phospholipids in women with early recurrent pregnancy losses (RPLs).

DESIGN

Retrospective data analysis.

SETTING

Tokai University Hospital, Kanagawa, Japan.

PATIENT(S)

One hundred thirty-nine patients with unexplained early RPLs.

INTERVENTION(S)

None.

MAIN OUTCOME MEASURE(S)

Enzyme-linked immunosorbent assays were used to measure autoantibodies to phosphatidylethanolamine, cardiolipin, and phosphatidylserine.

RESULT(S)

Twenty-eight (20.1%), 17 (12.2%), and 2 (1.4%) patients of the 139 total patients were positive for immunoglobulin (Ig) G, IgM, and IgA antiphosphatidylethanolamine antibodies, respectively. Because 3 patients had two isotypes, 44 (31.7%) of the patients were positive for antiphosphatidylethanolamine antibodies. Six patients (4.3%) and 1 patient (0.7%) were positive for IgG and IgM antiphosphatidylserine antibodies, respectively. Seven patients (5%) were positive for beta2-glycoprotein I-independent anticardiolipin IgG, and 1 patient was positive for beta2-glycoprotein I-dependent anticardiolipin IgG. Two patients (1.4%) had lupus anticoagulant.

CONCLUSION(S)

Our data show a statistically stronger association between RPLs and antiphosphatidylethanolamine antibodies than between RPLs and antibodies to anionic phospholipids for early gestational losses. Our data suggest that antiphosphatidylethanolamine antibodies may be a risk factor in patients with early RPLs.

Development of optimal techniques for cryopreservation of human platelets. I. Platelet activation during cold storage (at 22 and 8 degrees C) and cryopreservation

Using the current blood bank storage conditions at 22 degrees C, the viability and function of human platelets can be maintained for only 5 days. This does not allow for the necessary and extensive banking of platelets needed to treat patients afflicted with thrombocytopenia, a side effect of many invasive surgeries such as cardiopulmonary bypass or bone marrow transplantation. The development of optimal techniques for long-term cryopreservation and banking of human platelets would provide the ability to greatly extend the viable life of the platelet and would fulfill an increasing and urgent need in many clinical applications. To determine the optimal techniques for platelet preservation, the expression of an activation marker, phosphatidylserine, on the platelet membrane during storage at 22 and 8 degrees C as well as during the different freezing preservation processes was examined using flow cytometry and annexin V binding assay. Human platelets were identified by both CD41 and light scatter in flow cytometry. In cryopreservation experiments, effects of the following factors on platelet activation were evaluated: (a) cryoprotective agents (CPAs) type: dimethyl sulfoxide (Me2SO), ethylene glycol (EG), and propylene glycol (PG), (b) CPA concentration ranging from 0 to 3 M, and (c) ending temperatures of a slow cooling process at -1 degrees C/min. Our results demonstrated that (a) approximately 50% of platelets were activated on days 7 and 16 at 22 and 8 degrees C, respectively; (b) platelets were not significantly activated after 30-min exposure to 1 M Me2SO, EG, and PG at 22 degrees C, respectively, and (c) there was a significant difference in cryoprotective efficacy among these three CPAs in preventing platelets from cryoinjury. After being cooled to -10 degrees C, 74% of the cryopreserved platelets survived (nonactivated) in 1 M Me2SO solution, while in 1 M EG and 1 M PG solutions, 62 and 42% of the platelets survived, respectively. Using the information that Me2SO consistently yields higher percentages of nonactivated platelets and does not seem to be cytotoxic to platelets for 30-min exposure time, this was found to be the optimal cryoprotective agent for platelets. In addition, significant Me2SO toxicity to platelets was not noted until Me2SO concentrations exceeded 2 M. Finally, a concentration of 1 M Me2SO proved to be the most effective at all cryopreservation ending temperatures tested (-10, -30, -60, and -196 degrees C). In conclusion, under the present experimental conditions, a storage temperature of 8 degrees C appeared to be much better than 22 degrees C. Although the potential chemical toxicity of 1 M Me2SO, EG, or PG is negligible, 1 M Me2SO was found to be optimum for cryopreservation of human platelets. PG has the least cryoprotective function for low-temperature platelet survival.

Platelet membrane early activation markers during prolonged storage

The relationship between platelet aging and early markers of membrane activation have not been defined clearly. Activation markers expressed during prolonged storage are similar if not identical to those that appear after exposure to thrombin. Using flow cytometry, we investigated platelet membrane expression of CD62P, CD63, and annexin V binding (i.e., loss of membrane asymmetry) in platelets stored for up to 11 days under standard blood banking conditions. We compared five apheresis platelets to two random donor platelet concentrates, and to one pooled platelet preparation from six single platelet concentrates before and after exposure to thrombin. CD62P, CD63 expression, and annexin V binding increased during storage albeit with different kinetics. The differential increments observed between resting and thrombin (1 unit/ml) activated platelets showed an inverse correlation to storage time for CD62P, CD63, and annexin V binding, which was identical to published survival curves. A difference between apheresis platelets and platelet concentrates was observed only on day 1. Our data indicate that the in vitro platelet reserve activity to thrombin activation mirrors that of radiolabeled platelet survival in vivo and that platelet cross-sectional residual life span can explain their diminished capacity to respond to thrombin as a function of viability.

Uteroplacental vascular involvement in recurrent spontaneous abortion

Spontaneous abortion is common in human pregnancy. Recent advances in pregnancy immunology and vascular biology are reviewed with emphasis upon the events associated with recurrent fetal losses. Certain treatment options used to alleviate or prevent some miscarriages are presented and discussed.

The pre-conditioning incidence of antiphospholipid antibodies is not significantly increased in patients with bone marrow transplant-related organ dysfunction

Hepatic dysfunction resulting from hepatic veno-occlusive disease (VOD) is a common complication of bone marrow transplantation (BMT). Some investigators believe that hepatic dysfunction, along with pulmonary and central nervous system (CNS) dysfunction, is part of a systemic disorder called multiple organ dysfunction syndrome (MODS). Endothelial damage by pretransplant chemo-radiation and activation of hemostasis are considered early events in the development of hepatic VOD. The pathological mechanism leading to fibrous obliteration of hepatic vessels may also take place in pulmonary and CNS vessels. Since antiphospholipid antibodies (aPA) are associated with venous and arterial thrombosis, which can lead to vessel occlusion, we asked if the incidence of aPA before conditioning was greater in patients who developed MODS following BMT. Samples drawn before pretransplant chemo-radiation from 57 patients who subsequently developed MODS and 55 control patients who did not develop MODS were studied blindly for aPA by ELISA. The number of aPA-positive patients who developed MODS (10/57), compared to the number of aPA-positive patient controls who did not develop MODS (7/55) was not statistically significant (P = 0.48). Our data indicate that the incidence of aPA before conditioning was not greater in patients who developed MODS, including hepatic VOD, following BMT.

Compliance and use behaviour, an issue in injectable as well as oral contraceptive use? A study of injectable and oral contraceptive use in Johannesburg

This study examines the compliance, use behaviour and knowledge of method of women using injectable and oral contraceptives in two clinic sites in the Johannesburg area, South Africa. An interviewer administered questionnaire was used to collect information in the clients' home language. A total of 400 women were interviewed in the clinics. The mean age of clients was 26.2 years (range 13 to 43 years). Of the clients not wanting to get pregnant, 30.4 per cent of injectable users and 18.4 per cent of oral contraceptive (OC) users had stopped using their method temporarily before returning to the same method (called the nonuse segment) and had not used any other form of contraception during this time. Almost one third of injectable users (31.2 per cent) had been late for their next injection at least once. Although nearly all women using injectables had experienced some menstrual disturbances, over one third (38.5 per cent) had not been informed by the providers about the possibility of these changes. Many women gave the disruption of their menstrual cycle as the reason for the nonuse segment. The majority of OC users lacked information on how to use their method correctly. Nearly all women expressed an interest in obtaining more information on their current method and other available methods. This study shows that compliance is an issue in injectable as well as OC users.

Caesarean section rates in South Africa: evidence of bias among different 'population groups'

OBJECTIVE

The aim of this study was to assess the effect of 'population group' classification, as a specific instance of 'racial' categorization, on caesarean section rates in South Africa.

DESIGN

Information on 'population group' classification ('Black, 'Coloured', 'Indian', or 'White', as defined under apartheid legislation) and place of delivery, together with basic obstetric, sociodemographic and perinatal data, were extracted from the birth notification forms of 5456 children who made up the birth cohort of the Birth to Ten longitudinal study. This cohort included all births that occurred to mothers resident in Soweto-Johannesburg during a 7-week period in 1990.

RESULTS

After accounting for differences in maternal age, gravidity, birth weight and gestational age at delivery, the rate of caesarean sections at private facilities was more than twice that at public facilities. Although there were significant differences in the utilisation of private facilities by women from different 'population groups', there was an independent effect of 'population group' classification on caesarean section rates: caesarean section rates among women classified as 'White' and 'Coloured' were significantly higher (95% confidence intervals for odds ratios: 1.40-2.42 and 1.05-1.81, respectively) than among women classified as 'Black'.

CONCLUSION

'Population group' differences in caesarean section rates among South African women are not explained by differences in demographic risk factors for assisted delivery, nor by differences in access to private health care. Instead, the differences in section rates may reflect the effect of bias in clinical decision-making, and/or differences among women from different 'population groups' in their attitude towards assisted delivery, and their capacity to negotiate with clinicians.

DQB1*0202 and the new DQB1*0203 allele: a fourth pair of DQB1 alleles differing only at codon 57

Amino acid 57 of DQ beta chains is of functional importance as it influences peptide binding, is part of B and T cell epitopes, and is associated with susceptibility and resistance to insulin-dependent diabetes mellitus and humoral immunodeficiencies. Polymorphism of codon 57 is conserved in primates and in HLA class II B genes implying that balancing selection operates on this residue. Previously, three DQB1 allele pairs have been described, that only differ at residue 57. In an African-American Black individual with the HLA phenotype A23.30;B58,63;Cw6;DR18,12;DR52;DQ5,2, we found a fourth example of this dimorphism: the new DQB1*0203 allele, that was identical to DQB1*0202 except for codon 57, which encodes aspartic acid and alanine respectively in the two alleles. The class II haplotype carrying the new allele was deduced to be DRB1*0302,DRB3*0101,DQA1*05011,DQB1*0203.

The reliability of hand-written and computerised records of birth data collected at Baragwanath hospital in Soweto

This study examined the reliability of hand written and computerised records of birth data collected during the Birth to Ten study at Baragwanath Hospital in Soweto. The reliability of record-keeping in hand-written obstetric and neonatal files was assessed by comparing duplicate records of six different variables abstracted from six different sections in these files. The reliability of computerised record keeping was assessed by comparing the original hand-written record of each variable with records contained in the hospital's computerised database. These data sets displayed similar levels of reliability which suggests that similar errors occurred when data were transcribed from one section of the files to the next, and from these files to the computerised database. In both sets of records reliability was highest for the categorical variable infant sex, and for those continuous variables (such as maternal age and gravidity) recorded with unambiguous units. Reliability was lower for continuous variables that could be recorded with different levels of precision (such as birth weight), those that were occasionally measured more than once, and those that could be measured using more than one measurement technique (such as gestational age). Reducing the number of times records are transcribed, categorising continuous variables, and standardising the techniques used for measuring and recording variables would improve the reliability of both hand-written and computerised data sets.

Identification of structural mutations in the fifth domain of apolipoprotein H (beta 2-glycoprotein I) which affect phospholipid binding

Apolipoprotein H (apoH), also known as beta 2-glycoprotein-I, is considered to be a cofactor for the binding of certain antiphospholipid autoantibodies to negatively charged phospholipids. Genetically determined structural abnormalities in the lipid binding domain(s) of apoH can affect its ability to bind lipid and consequently the production of the autoantibodies. In this study we have identified two common structural mutations at codons 316 and 306 in the fifth domain of apoH which rendered apoH unable to bind to negatively charged phosphatidylserine (PS). The missense mutation at codon 316 (TGG --> TCG) replaces Trp316 with Ser316 and disrupts the integrity of four highly conserved hydrophobic amino acids sequence at positions 313-316, which is a potential protein-lipid hydrophobic interaction site. The missense mutation at codon 306 (TGC --> GGC) involves the substitution of Cys306 by Gly306 which causes the disruption of a disulfide bond between Cys281 and Cys306 and affects the normal configuration of the fifth domain of apoH that appears to be critical for clustering positively charged amino acids along with four hydrophobic amino acids sequence. ApoH from the two homozygotes (Ser316/Ser316) and all seven compound heterozygotes (Ser316/Gly306) failed to bind to PS; all heterozygotes at one or the other sites and wild type showed normal PS binding. These data indicate that the fifth domain of apoH harbors the lipid binding region. An estimated 2 million Caucasians in the United States, who are compound heterozygotes for the two mutations, may be precluded from producing apoH-dependent antiphospholipid autoantibodies.

Phospholipid binding plasma proteins required for antiphospholipid antibody detection--an overview

PROBLEM

Antibodies to phospholipid antigens (aPA) are associated with thrombosis thrombocytopenia and recurrent pregnancy loss. Contemporary data show many aPA target phospholipid-binding plasma proteins and not phospholipids. The purpose of this overview is to describe several phospholipid-binding proteins and provide data to demonstrate how the interaction between phospholipids and phospholipid binding proteins results in expression of neo-autoantigenic epitopes.

METHOD

Review of existing data.

RESULTS

Illustrations of how certain plasma proteins beta 2 glycoprotein I, prothrombin, high and low molecular weight kininogens interact with the anionic phospholipids cardiolipin and phosphatidylserine and the zwitterionic phospholipid, phosphatidylethanolamine are shown and discussed. A model of aPA mediated thrombosis is presented.

CONCLUSIONS

Some aPA recognize phospholipids directly, however, the majority and many which correlate with pathology target phospholipid binding proteins. Published data indicate that aPA represent a constellation of antibodies with multiple specificities. Insight into mechanisms responsible for aPA-associated thrombosis should provide a basis for treatment.

Analyses of HLA-DQ alleles in recurrent spontaneous abortion (RSA) couples

PROBLEM

The influence of HLA sharing on pregnancy outcome is controversial. In renal transplantation, HLA-DQB1 donor-recipient mismatches have been shown beneficial for long-term transplant success. Since pregnancy is defined as Nature's allograft, we investigated the relevance of HLA-DQ mismatching in normal reproducing couples compared to couples experiencing RSA.

METHOD

Unexplained RSA couples referred to our laboratory for immunological testing were classified by immunological findings and obstetrical history. Primary RSA couples shared > or = 2 HLA-A, B, or DR antigens, had no cytotoxic anti-paternal antibodies, and no gestation beyond 20 weeks. Secondary RSA couples had cytotoxic anti-paternal antibodies and RSA after a live birth. HLA-DQA1 and DQB1 alleles were identified by PCR-SSP.

RESULTS

No differences in DQA1 and DQB1 mismatch were observed among RSA patients and controls. DQA1-DQB1 haplotype mismatches were not different among the three groups of couples.

CONCLUSIONS

In contrast to renal transplant, HLA-DQ incompatibility did not differ among RSA couples compared with successful reproducing couples.

Autoantibodies to kininogen-phosphatidylethanolamine complexes augment thrombin-induced platelet aggregation

Autoantibodies to the zwitterionic phospholipid (PL), phosphatidylethanolamine (PE), have been described in patients with thrombotic disease. We have reported that certain anti-PE antibodies (aPE) are not specific for PE, but are directed to PE-binding plasma proteins, high molecular weight kininogen (HK) and low molecular weight kininogen (LK). Kininogens bind to platelets and inhibit thrombin-induced platelet aggregation. This inhibition is specific for thrombin because kininogens do not inhibit platelet aggregation induced by adenosine diphosphate (ADP), collagen or calcium ionophore. To date, a platelet kininogen receptor has not been described. We recently reported that purified kininogens bind to purified PE in vitro. This opens the possibility that kininogens can bind to platelets by virtue of exposed PE in the platelet membrane. We thus questioned if aPE can recognize platelet bound kininogens and negate their antithrombotic property. Our experiments support this possibility by demonstrating that exogenously added kininogen-dependent IgG aPE markedly increased thrombin-induced platelet aggregation in vitro but did not alter ADP-induced aggregation. In contrast, kininogen independent IgG aPE which recognized PE per se did not augment thrombin-induced platelet aggregation. These data support a hypothesis that kininogen dependent aPE may cause thrombosis in vivo due to disruption of the normal antithrombotic effects of kininogen.

Utilization of intravenous immunoglobulin to ameliorate alloantibodies in a highly sensitized patient with a cardiac assist device awaiting heart transplantation. Fluorescence-activated cell sorter analysis

Surgery surrounding the use of mechanical assistance in cardiac transplant candidates often leads to multiple blood/platelet transfusions and subsequent development of alloantibodies. This is a case report of a 50-year-old male patient who had received blood transfusions during coronary bypass grafting 9 years earlier. He presented in acute and chronic heart failure and, despite therapy, became moribund with multisystem organ failure. His ejection fraction was 10%. A Novacor ventricular assist device was implanted on May 19, 1995 (day 0). The patient received 44 U of blood and 20 U of platelets. Although his percent reactive antibodies (PRA) were negative before surgery by fluorescence-activated cell sorter analysis, the PRA 3 days after implantation of the ventricular assist device was 80%; it increased to 100% by day 7. In an attempt to decrease the PRA, intravenous immunoglobulin was given at 3-week intervals. The PRA became negative and the patient received a donor heart that was negative by fluorescence-activated cell sorter cross-match on day 64. On days 69-72, a dramatic increase in alloantibody activity was promptly reversed with additional intravenous immunoglobulin. Currently at posttransplant month 12, the patient shows no humoral, cellular, or vascular evidence of rejection.

Antiphosphatidylethanolamine antibodies as the only antiphospholipid antibodies detected by ELISA. II. Kininogen reactivity

OBJECTIVE

To study the requirement for serum and for low (LMWK) and high molecular weight kininogen (HMWK) and/or HMWK binding proteins to detect antiphosphatidylethanolamine antibodies (aPE) in ELISA.

METHODS

Eighteen patients with aPE (9 IgG and 13 IgM) as the only antiphospholipid antibody (aPL) detected by ELISA were assigned to 4 groups: thromboembolic episodes (TEE) (Group I, n = 6); livedo reticularis (LR) without TEE, (Group II, n = 4); both LR and thrombosis (Group III, n = 4); and systemic lupus erythematosus (SLE) or primary antiphospholipid syndrome (APS) (Group IV, n = 4). All sera were analyzed in ELISA with and without bovine serum and with a purified chromatographic fraction containing LMWK, HMWK, and HMWK binding proteins.

RESULTS

Eleven aPE were serum dependent: mostly IgG (7/9) and some IgM (4/13). Among the 11 serum dependent aPE, all the 7 IgG and 2 IgM were kininogen reactive. Some serum independent IgM were better detected in the absence than in the presence of serum in the ELISA.

CONCLUSION

In the 18 patients, kininogens and/or HMWK binding proteins served as a "cofactor" significantly more often for aPE IgG than for aPE IgM (p = 0.007). Kininogen dependent aPE Ig were observed more often in patients with LR with or without TEE (6/8) than in those with SLE or primary APS (0/4) but this difference merely tended to significance (p = 0.06). In 2 patients, one with TEE, the other with primary APS, the IgM aPE was dependent on a serum "cofactor" that was not kininogen.

Detection of IGA anti-OKT3 antibodies in OKT3-treated transplant recipients

Murine OKT3 monoclonal antibodies function as an immunosuppressant drug for organ transplant recipients. A contraindication to retreatment may develop, however, because a high proportion of OKT3-treated patients form anti-OKT3 antibodies. Previous data have shown that only antiidiotypic IgG antibodies can negate the beneficial effect of the drug. Eighty-two OKT3-treated transplanted patients were tested by ELISA for IgG, IgK and IgA anti-OKT3 antibodies and compared with 200 controls. The anti-OKT3 antibody-positive sera were screened additionally by flow cytometry for the presence of antiidiotypic activity by measuring Ortho OKT3-FITC activity on a CD3-positive cell line, Jurkat, before and after incubation with serial dilutions of patient and control sera. Forty-four of 82 patients developed antibodies to OKT3, 20 manifested IgG, 20 produced both IgG and IgA, and 4 IgA only. We never detected IgM anti-OKT3. Of the 44 anti-OKT3-positive patient sera, 25 showed antiidiotypic specificity. Two IgG/IgA anti-OKT3-positive patient sera were IgG-depleted by Protein G. Both continued to exhibit antiidiotypic IgA activity. IgA anti-OKT3 was associated with low serum OKT3 levels and lack of ability of OKT3 to lower total CD3 cell numbers to therapeutic levels. This is the first report of IgA anti-OKT3 antibody in transplant recipients. Isotype IgA anti-OKT3 was observed in 54% of the patients whose sera tested positive for anti-OM by ELISA. The IgG/ IgA anti-OKT3-positive patient sera tested continued to exhibit antiidiotypic OKT3 reactivity when depleted of IgG. We urge that OKT3-treated patients be monitored routinely for IgA anti-OKT3 antibodies to avoid the expense and potential complications of retreatment with this drug in sensitized patients.

Alteration of the aPA ELISA by UV exposure of polystyrene microtiter plates

Interlaboratory inconsistencies in antiphospholipid antibody (aPA) solid phase assays have prompted controversy in clinical laboratory testing for aPA. We found that the aPA ELISA can be influenced by the type of microtiter plate utilized and by the conditions in which the plates are stored. By exposing 96-well, flat-bottom polystyrene microtiter plates to short wave UV light (254 nm), the aPA ELISA signal decreased in a UV dose-dependent manner. No effect was seen with long wave UV light (366 nm). These results were independent of the antibody isotype under study or the phospholipid (PL) antigen used: anionic phosphatidylserine (PS) and cardiolipin (CL), or zwitterionic phosphatidylethanolamine (PE). Purified human beta 2-glycoprotein I (beta 2 GPI), a known cofactor for anionic PL, and rabbit anti-beta 2 GPI antisera were used to demonstrate that beta 2 GPI bound equally to UV treated and untreated microtiter plates. In contrast, recognition of beta 2 GPI on an anionic PL surface was decreased on UV treated plates, suggesting that UV exposure alters the lipid binding properties of the microliter plate. To determine whether UV exposure inhibited PL binding directly or caused a change in the way the PL was bound, the amount of PL bound to UV treated and untreated plates was measured by using fluorescent labeled PS and a fluorimeter. PS binding was decreased by 53% in UV treated wells as compared to untreated wells. These data show that short wave UV exposure reduces PL binding to polystyrene microtiter plates, thereby reducing the amount of beta 2 GPI bound to PL coated ELISA plates. Thus by using UV exposed microtiter plates, decreased or false-negative a PA ELISA results may be obtained for aPA positive plasmas.

Autoantibodies to phosphatidylethanolamine (PE) recognize a kininogen-PE complex

Demonstration of autoimmune antiphospholipid antibodies (aPA) to negatively charged phospholipids (PL) in an enzyme-linked immunosorbent assay (ELISA) requires the presence of certain phospholipid-binding plasma proteins, eg, beta 2-glycoprotein I. We found a requirement for plasma against the electrically neutral or zwitterionic phospholipid, phosphatidylethanolamine (PE). Two of these PE-binding plasma proteins were identified as high molecular weight kininogen (HMWK) and low molecular weight kininogen (LMWK). We studied anti-PE antibody (aPE) seropositive plasma from 13 patients with SLE and/or recurrent spontaneous abortions by using partially purified kininogens and kininogen binding proteins from adult bovine serum isolated by carboxymethyl (CM)-papain affinity chromatography. Eleven of 13 sera recognized a kininogen-PE complex and/or a kininogen-binding protein-kininogen-PE complex. Some aPE-positive patient sera were shown to recognize highly purified HMWK and LMWK by ELISA only when the kininogens were presented on a PE substrate. These aPE sera did not recognize PE, HMWK, or LMWK when they were presented independently as the sole antigens on the ELISA plates. Other aPE-positive sera that did not react with PE-bound HMWK or LMWK reacted with the CM-papain column eluate when it was bound to PE, which suggests that these aPE recognize factor XI or prekallikrein, which normally bind to HMWK. The aPE ELISA reactivity of two patient sera were inhibited by preincubation of the CM-papain column eluate in the ELISA plate. These data show that most aPE are not specific for PE but require the presence of certain PL-binding plasma proteins that are kininogens or proteins in complex with kininogens. Our studies indicate that aPE bind to different plasma proteins than those implicated in anionic PL, aPA ELISA reactivity.

Antiphospholipid antibodies in heart transplant recipients

Antiphospholipid antibodies (aPA) are found in patients with systemic lupus erythematosus, aPA syndrome, myocardial infarction, and stroke. The presence of aPA may predict recurrent events in certain victims of heart attack and stroke. Blood samples from 105 cardiac transplant recipients (81 men, 24 women) were tested by enzyme-linked immunosorbent assay (ELISA) for the presence of IgG, IgM, and IgA aPA to phosphatidylserine (PS), cardiolipin (CL), and phosphatidylethanolamine (PE). Patients' ages ranged from 17 to 70 years (mean 51 years). Collection times ranged from Day 1 to 9 years post transplant (mean 36 months). All patients received triple immunosuppressive therapy. We report our aPA ELISA results in multiples of the normal median (MoM) of the OD405 values calculated for 252 healthy individuals. A positive MoM is greater than the MoM that encompasses 95% of the controls; for example, above 3 MoM is considered positive for IgG anti-CL, IgA anti-PS, and CL. Above 4 MoM is positive for IgG anti-PS and PE and IgM anti-PS and CL. Thirty-nine patients had IgG anti-PS (range 4.1-14.8 MoM), 63 had IgG anti-CL (3.1-9.4 MoM), 7 had IgM anti-PS (4.1-12.1 MoM), 1 had IgM anti-CL (14 MoM), 47 had IgA anti-PS (3.1-13.1 MoM), and 58 had IgA anti-CL (3.1-11.5 MoM). In our patient population, the incidence of IgG and/or IgA aPA was significantly higher (p < 0.001) than IgM. Few patients showed specificity for either PS, CL, or PE, and many were positive with more than one antibody isotype. Because aPA were evaluated in these patients, we investigated pretransplant serum samples which are available from 79 of the 105 recipients, and found aPA in 52 of 79 (66%) patients before transplantation. Longitudinal studies were done in three patients: two had increasing IgA aPA, beginning on Days 13 and 26 post transplant, whereas the third patient showed an increased aPA on Day 8 but a decrease on Day 23. Studies are in progress to determine whether a correlation exists between the presence of aPA, immunocytochemical (biopsy) findings, and clinical outcome.

Heterogeneity of the apolipoprotein H*3 allele and its role in affecting the binding of apolipoprotein H (beta 2-glycoprotein I) to anionic phospholipids

Apolipoprotein H (apoH, protein; APOH, gene) has been implicated as a necessary cofactor for the binding of certain autoimmune antiphospholipid antibodies to anionic phospholipids. APOH exhibits genetically determined structural polymorphism with the occurrence of four alleles. Recently three IgG1k monoclonal antibodies (mAb) to human apoH, designated 3G9, 1B4, and 3D11, have been produced. The mAb 3D11 does not recognize apoH bound to anionic phospholipids in contrast to mAb 3G9 and 1B4, which recognize free and phospholipid-bound apoH. In this investigation we have determined the reactivity of the three mAb with four APOH allele products and the binding ability of these allele products with anionic phospholipids. The mAb 3G9 and 1B4, like the polyclonal anti-apoH, were equally reactive with all four allelic products, but the 3D11 recognized only the APOH*3 allele product. In the 159 APOH*3 carriers tested from five ethnic groups, the reactivity of mAb 3D11 was observed with all the Chinese but none of the African blacks. For the U.S. whites and Polynesians 89% and 75%, respectively, of the APOH*3 allele products were recognized by 3D11, while 87% of the U.S. blacks with this allele had no 3D11 reactivity. These data show that the APOH*3 allele, originally identified as a single entity by the polyclonal anti-apoH, is heterogeneous with at least one distinct variation based on mAb 3D11 reactivity. Our data also demonstrate that the apoH from certain homozygous APOH*3 individuals is unable to bind to anionic phospholipids. Such ethnic-specific apoH variations could play a significant role in the binding properties of autoimmune antiphospholipid antibodies to anionic phospholipids.