In vitro evidence that hyperglycemia stimulates tumor necrosis factor-alpha release in obese women with polycystic ovary syndrome.

Women with polycystic ovary syndrome (PCOS) are often insulin resistant and have chronic low-level inflammation. The purpose of this study was to determine the effects of hyperglycemia in vitro on tumor necrosis factor (TNF)-alpha release from mononuclear cells (MNC) in PCOS. Twelve reproductive-age women with PCOS (six lean, six obese) and 12 age-matched controls (six lean, six obese) were studied. Insulin sensitivity (IS(HOMA)) was estimated from fasting levels of glucose and insulin and percent truncal fat was determined by dual energy absorptiometry (DEXA). TNFalpha release was measured from MNC cultured under euglycemic and hyperglycemic conditions. IS(HOMA) was higher in obese women with PCOS than in lean women with PCOS (student's t-test; 73.7 +/- 14.8 vs 43.1 +/- 8.6, P < 0.05), but similar to that of obese controls. IS(HOMA) was positively correlated with percent truncal fat (r=0.57, P < 0.04). Obese women with PCOS exhibited an increase in the percent change in TNFalpha release from MNC in response to hyperglycemia compared with obese controls (10 mM, 649 +/- 208% vs 133 +/- 30%, P < 0.003; 15 mM, 799 +/- 347% vs 183 +/- 59%, P < 0.04). The TNFalpha response directly correlated with percent truncal fat (r=0.45, P < 0.03) and IS(HOMA) (r=0.40, P < 0.05) for the combined groups, and with plasma testosterone (r=0.60, P < 0.05) for women with PCOS. MNC of obese women with PCOS exhibit an increased TNFalpha response to in vitro physiologic hyperglycemia. MNC-derived TNFalpha release may contribute to insulin resistance and hyperandrogenism, particularly when the combination of PCOS and increased adiposity is present.

Trophoblast differentiation: progenitor cells, fusion and migration -- a workshop report.

Challenge lies ahead in unravelling the role played by trophoblast and its repertoire of expressed genes in normal human placental development, growth and pathology. Specific technical advances will clearly be required for characterisation of function. In particular, improvements in our repertoire of in vitro models are needed before many of the key questions can be answered. Recent advances in the study of human trophoblast differentiation are discussed.

Infarction of solid Hodgkin's tumors in mice by antibody-directed targeting of tissue factor to tumor vasculature.

We demonstrated previously that selective thrombosis of the blood vessels of solid tumors in mice can be achieved by targeting the extracellular domain of tissue factor by means of an antibody to an experimentally induced marker on tumor vascular endothelium. In the present study, we extend this finding to a naturally occurring marker of tumor vascular endothelium, vascular cell adhesion molecule-1 (VCAM-1). VCAM-1 is expressed by vascular endothelial cells in Hodgkin's disease and various solid tumors in mice and humans. It is absent from vascular endothelial cells in normal tissues in mice, with the exception of the heart and lungs, where it is present on venules. A monoclonal antibody to murine VCAM-1 was covalently linked to the extracellular domain of human tissue factor to create a "coaguligand." After i.v. administration to severe combined immunodeficient mice bearing human Hodgkin's tumors, the coaguligand localized selectively to VCAM-1-expressing vessels, caused thrombosis of those vessels, and retarded tumor growth. The coaguligand also localized to VCAM-1-expressing vessels in the heart and lungs of the mice but did not induce thrombosis in these sites. An immunohistochemical evaluation of the distribution of a monoclonal anti-phosphatidylserine (PS) antibody in the mice showed that the VCAM-1-expressing vessels in the tumor expressed PS, whereas the VCAM-1-expressing vessels in the heart and lungs lacked PS. The lack of thrombotic effect of the coaguligand on heart and lung vessels may be because PS is needed to provide the procoagulant surface upon which coagulation complexes can assemble. The requirement for coincident expression of the targeted marker and PS on tumor endothelium probably contributes to the selectivity of thrombotic action and the safety of coaguligands.

Detection of circulating immune complexes with a Raji cell enzyme immunoassay.

The Raji cell assay for circulating immune complexes (CIC) is frequently the method of choice when the detection of large, complement fixing complexes is desired. We have developed an enzyme-linked immunosorbent assay modification (Raji-ELISA) of the Raji cell technique which is easy to perform, uses commercially available reagents and is more convenient than the conventional Raji cell radioimmune assay (Raji-RIA). Fourteen samples were tested by both assays and a good correlation was observed (P = 0.05). Sera from patients with suspected immune complex associated diseases were tested in the Raji-ELISA. Unsensitized Raji cells gave a value of 1.25 micrograms of immune complex associated IgG/ml. Using 2 standard deviations as a cutoff to determine positivity, 2 of 32 healthy controls (6.2%) had elevated levels of circulating immune complexes. In our study population, 9 of 23 cancer patients (39.1%), 10 of 13 patients with autoimmune diseases (76.9%), 3 of 17 patients with positive rheumatoid factor titers (17.6%), 1 of 23 pregnant patients (4.3%), 1 of 5 preeclamptic patients (20%) and 9 of 30 other patients with suspected immune complex associated diseases (30%) had elevated levels of CIC.

Antiphospholipid antibody binding to bilayer-coated glass microspheres.

Thrombosis, recurrent fetal loss, and thrombocytopenia are clinical manifestations associated with circulating antibodies that recognize cardiolipin (CL)- or phosphatidylserine (PS)-dependent antigens. Enzyme-linked immunosorbent assays (ELISAs) are generally used to determine the presence and specificity of antiphospholipid antibodies (aPLs). However, the presentation of the phospholipid antigen in the ELISA assay is unknown. In this study, we determined the specificity of three mouse monoclonal aPLs for phospholipid bilayer membranes. These monoclonal aPLs had been characterized by ELISA to have different specificities for CL and PS and were designated BA3B5C4 (CL+/PS+), 3SB9b (CL-/PS+), and D11A4 (CL+/PS-). Bilayers composed of 0-100% PS or CL in phosphatidylcholine (PC) were formed on the surface of 1.6 microns diameter glass microspheres to permit analysis by flow cytometry. BA3B5C4 and 3SB9b bound specifically to both PS- and CL-containing bilayers, and binding increased with increasing percentage of anionic phospholipid. The threshold for PS-dependent binding was 20 mol% PS for both BA3B5C4 and 3SB9b. For CL-dependent binding, the threshold was below 25 mol% CL for both of these antibodies. Binding to PS-containing bilayers was tested as a function of ionic strength for BA3B5C4 and 3SB9b. The ionic strength dependence of the binding suggested that the intermolecular attractive forces between anti-PS antibodies and PS-containing bilayers are predominantly multiple weak electrostatic bonds. D11A4 bound only to bilayers composed of 100% PS and 100% PC, and this antibody did not bind to CL-containing bilayers. The binding specificities of these aPLs to bilayer membranes suggest that, in this system, the conformation of the epitope involving CL, and perhaps PS, is different from that expressed in the routine clinical ELISA. Two of the monoclonal antibodies reacted in this model system at the low levels of PS typically externalized in the plasma membranes of activated platelets, apoptopic lymphocytes, and senescent red blood cells: thus, these surfaces are plausible candidates for the site of pathologically relevant antibody interactions.

The role of human endogenous retroviruses in trophoblast differentiation and placental development.

A major portion of the human genome appears to be of retroviral origin. These endogenous retroviral elements are expressed in a variety of normal tissues and during disease states, such as autoimmune and malignant conditions. Recently, potential roles have been described for endogenous retroviral envelope proteins in normal differentiation of human villous cytotrophoblast into syncytiotrophoblast. This article provides a brief critical review of the current state of knowledge concerning the expression of the env regions of three endogenous retroviral elements: ERV-3, HERV-W, and HERV-FRD. A testable model of villous cytotrophoblast differentiation is constructed, in which a complementary expression of endogenous retroviral envelope proteins initiates hCG production, decreased cell proliferation, and intercellular fusion.

The cellular mechanism by which the human endogenous retrovirus ERV-3 env gene affects proliferation and differentiation in a human placental trophoblast model, BeWo.

The env region of the human endogenous retrovirus ERV-3 is expressed during differentiation of trophoblast and the choriocarcinoma BeWo. Stable transfectants with ERV-3 env exhibit most aspects of trophoblast differentiation, including inhibition of cell proliferation, changes in cell morphology, and increased production of beta-hCG mRNA. In this study, the cellular mechanism of induction of BeWo cell differentiation by ERV-3 env was investigated. In BeWo cells stably transfected with ERV-3 env, the production of beta-hCG mRNA and hCG protein was increased. Intracellular cAMP level was markedly increased over that of vector transfected cells. The effect on beta-hCG protein production was inhibited by H89, a protein kinase A (PKA) inhibitor, while protein kinase C (PKC) and protein tyrosine kinase (PTK) inhibitors had no effect. The expression of a major cell cycle promoter, cyclin B, was markedly reduced while expression of p21, a negative regulator of the cell cycle, was up-regulated. Inhibition of ERV-3 env induced hCG production with H89 had no significant effect on cell growth when compared with cells transfected with vector alone.

Expression of endogenous retrovirus ERV-3 induces differentiation in BeWo, a choriocarcinoma model of human placental trophoblast.

Differentiation-related expression of endogenous retrovirus ERV-3 env in the normal human placental syncytiotrophoblast suggests a role in placental development. The choriocarcinoma cell line BeWo, a model of trophoblast differentiation, is maintained in an undifferentiated state and undergoes differentiation upon the addition of forskolin. The expression of ERV-3 env mRNA increased after 48 h forskolin treatment, concurrently with increased intercellular fusion and production of human chorionic gonadotropin (beta-hCG) mRNA, a hormonal differentiation marker for trophoblast. Over expression of ERV-3 env induced differentiation of BeWo characterized by decreased cell growth, differentiation-related morphologic changes, and induction of beta-hCG mRNA. These results support the first known role for the expression of an endogenous retrovirus in trophoblast differentiation.

Modulation of phosphatidylserine epitope expression by BeWo cells during forskolin treatment.

The adenylcyclase activator forskolin induces the human choriocarcinoma line, BeWo, to undergo differentiation and fusion within 48 to 72 h. Using three monoclonal antibodies that differentiate between the anionic phospholipids cardiolipin (CL) and phosphatidylserine (PS) and immunoperoxidase techniques we investigated the expression of PS by BeWo during 48 h of forskolin treatment. We observed that BeWo cells not exposed to forskolin express an epitope of pS that reacts strongly with monoclonal antibody BA3B5C4 (CL+/PS+), whereas following treatment with forskolin there is a decrease in reactivity with BA3B5C4 and a concurrent increased activity with a second PS-reactive monoclonal antibody, 3SB9b (CL-/PS+). A third monoclonal antibody, D11A4 (CL+/PS-), that reacted with all anionic phospholipids except PS did not bind to BeWo cells, whether forskolin treated or not. These observations support previous interpretations using human placenta that during cytotrophoblast differentiation two antigenic forms of PS are expressed. Based on the described relationship of PS with cellular fusion events in other systems and the association of naturally occurring antibodies against PS with pregnancy loss and intrauterine growth retardation in humans, we propose that altered expression of PS during normal placental development and in BeWo after exposure to forskolin may be critical in the cytotrophoblast differentiation process.

Phosphatidylserine efflux and intercellular fusion in a BeWo model of human villous cytotrophoblast.

Phosphatidylserine (PS) efflux characterizes cytotrophoblast apoptosis and differentiation. To evaluate whether PS externalization and intercellular fusion were secondary to apoptosis, BeWo cells were induced to differentiate by forskolin or undergo apoptosis by staurosporine. PS externalization was measured by FITC-annexin V binding, and intercellular fusion was quantified by counting nuclei in syncytial cells. During forskolin treatment, vanadate decreased PS efflux by 78.0 per cent from 68.0 [5.3] (mean [SD]) to 15.0 [8.8] Lum (x10(3)) (P<0.001), whereas Z-VAD-fmk had no effect (66.5 [7.3]). Vanadate decreased intercellular fusion from 78.1 per cent [4.1] fusion in uninhibited cultures to 23.4 per cent [2.5], compared with 10.0 per cent [1.7] in media alone. Z-VAD-fmk did not affect fusion (80.4 per cent [6.8]). Staurosporine induced PS efflux was not affected by vanadate (69.6 [5.5] Lum x10(3)), but was inhibited 87.8 per cent by Z-VAD-fmk; from 71.5 [6.2] to 8.7 [3.6] Lum (x10(3)) (P<0.001). Apoptosis was measured by the TUNEL and COMET assays, lamin B fragmentation, activation of procaspase 3, mitochondrial membrane potential, and release of mitochondrial cytochrome c and apoptosis inducing factor. There was no indication of apoptosis associated with differentiation. Thus, PS efflux and intercellular fusion occurred through a vanadate-sensitive mechanism that was independent of apoptosis.

Aminophospholipid translocase activity in JEG-3; a choriocarcinoma model of cytotrophoblast differentiation.

The plasma membrane is characterized by a non-symmetrical distribution of phospholipids; the outer monolayer of the plasma membrane consists primarily of phosphatidylcholine (PC), and the aminophospholipids, phosphatidylserine (PS) and phosphatidylethanolamine (PE), preferentially reside in the inner monolayer. Asymmetry is maintained by a membrane associated ATP-dependent aminophospholipid translocase that preferentially relocates PS and PE from the outer to the inner monolayer. Although in most cells the translocase minimizes expression of PS on the outer surface, differentiating trophoblasts express increasing levels of surface PS. One possible explanation of prolonged PS externalization is that trophoblasts lack an effective aminophospholipid translocase. To test this hypothesis, fluorescent PC and PS analogues, NBD-PC and NBD-PS, were introduced into the plasma membrane of a choriocarcinoma model of trophoblast, JEG-3 cells. After incubation, the fluorescent lipid remaining on the outer monolayer was removed by incubation with fetal bovine serum. JEG-3 cells selectively translocated 80 per cent of the NBD-PS without significant translocation of NBD-PC. The process was significantly inhibited by N-ethylmaleimide (NEM) and vanadate. It is concluded that this model of trophoblast contains an active aminophospholipid translocase.

Preparation and functional characterization of villous cytotrophoblasts free of syncytial fragments.

Recent studies suggest that purified villous cytotrophoblasts are largely contaminated by mononucleated syncytial fragments and therefore unsuitable for studies of trophoblast differentiation. We assessed highly purified (>99.99 per cent) populations of villous trophoblasts for fragment contamination using the syncytial markers placental alkaline phosphatase (PLAP, by immunohistochemistry) and exteriorized phosphatidyl serine (ePS, by flow cytometric analysis). The preparations contained from 4-46 per cent syncytial fragments. However, we find that PLAP negative cells preferentially adhere to tissue culture surfaces and that all preparations were <2 per cent PLAP positive after routine plating and washing procedures. A second purification procedure eliminated dead (propidium iodide permeable) cells and separated viable syncytial fragments (ePS-positive) from viable cytotrophoblasts (ePS-negative) by two colour fluorescence activated cell sorting (FACS). Viable ePS-positive cells were ultrastructurally apoptotic, adhered poorly in culture and those that adhered rapidly underwent apoptosis. Viable ePS-negative cells contained large heterochromic nuclei and cytoplasmic structures, adhered strongly in culture and remained viable. The latter population (putative true villous CT) differentiated into syncytialized cells when cultured with EGF. We conclude that villous CT can be routinely purified, are viable in culture and can undergo syncytial fusion without extensive preformed syncytium.

Platelet-binding immunoglobulins in pregnancy-induced hypertension. II. Origin of circulating IgG and IgM antiplatelet antibodies in the umbilical cord serum.

IgG and IgM have been identified on the surface of maternal platelets in both autoimmune thrombocytopenia (ATP) and pregnancy-induced hypertension (PIH). IgG is also found on the umbilical cord platelets of patients with ATP and PIH, whereas IgM is only found on the umbilical cord platelets of patients with PIH. The possible maternal or fetal origins of these umbilical cord blood immunoglobulins were investigated by immunoblot analysis of antibodies in paired maternal and umbilical cord blood sera of ATP and PIH patients. Maternal sera contained IgG and IgM antibodies which reacted with several platelet proteins, however, a large amount of patient-to-patient variation was observed in the specific antigens that were identified. Analysis of paired maternal and umbilical cord sera from patients with ATP or PIH showed identical patterns of antigen specificity, which suggested that the IgG antibodies in the fetal circulation were of maternal origin. Circulating IgM antibodies were not observed in the umbilical cord sera of ATP patients. The umbilical cord sera of PIH patients, however, contained IgM antibodies that reacted against a variety of platelet antigens. In addition, most umbilical cord sera from PIH patients had identical patterns and relative intensities of reactivity, which differed from the patterns observed in the paired maternal sera. Antiplatelet IgM in the umbilical cord blood of PIH patients, therefore, appears to be a product of the fetal immune system.

Platelet-binding immunoglobulins in pregnancy-induced hypertension. I. Platelet-associated IgM on fetal platelets: evidence of a fetal autoimmune reaction?

Pregnancy-induced hypertension (PIH) can be complicated by maternal or fetal thrombocytopenia, or both. In order to investigate possible immunologic causes of these thrombocytopenias, platelet-associated IgG (PAIgG) and IgM (PAIgM) were measured in mothers with PIH and in their infants and compared with those from patients with autoimmune thrombocytopenic purpura (ATP), a known immunodestructive platelet disorder. Many PIH patients (33.3%) and most ATP patients (68.1%) had elevated levels of maternal PAIgG. In both diseases, the amount of PAIgG was directly proportional with the degree of thrombocytopenia (r = 0.446 in PIH and r = 0.668 for ATP). But in neither disease did the degree of maternal thrombocytopenia correlate with the degree of neonatal thrombocytopenia (r = 0.153 for PIH and r = 0.175 for ATP). Umbilical cord samples from PIH patients contained PAIgG (53.3%) and PAIgM (53.8%), whereas the umbilical cord samples from ATP patients had elevated amounts of PAIgG but not PAIgM. PAIgM in the umbilical cord blood could not be accounted for by IgM rheumatoid factors, IgM-containing immune complexes, or non-specific adsorption because of elevated total IgM levels. The umbilical cord blood PAIgM was probably not of maternal origin because it was observed even when the maternal blood contained no PAIgM and maternal IgM is not normally transported transplacentally. Therefore, the PAIgM appears to be of fetal origin. These results suggest that both maternal and fetal immunologic mechanisms may be involved in PIH-induced thrombocytopenia; if so, this is one of the first reported examples of a possible fetal autoimmune response.

Monoclonal antiphospholipid antibody reactivity against human placental trophoblast.

Naturally occurring antibodies against the negatively charged phospholipids cardiolipin (CL) and phosphatidylserine (PS) have been associated with recurrent pregnancy loss. One prevalent hypothesis proposes that antiphospholipid antibody (aPL) mediated pathophysiology is through increased placental thrombosis. In this study we investigated the reactivity of three mouse monoclonal aPLs with term and 26 week human placental preparations. Each monoclonal antibody reacted differently with CL and PS; 3SB9b reacted with PS (CL-/PS+), D11A4 reacted with CL (CL+/PS-) and BA3B5C4 reacted with both CL and PS (CL+/PS+). 3SB9b reacted strongly with the syncytiotrophoblastic layer of both formalin fixed and frozen placental tissue. Sporadic reactivity was observed against the cytotrophoblastic layer. BA3B5C4 reacted strongly and specifically with cytotrophoblastic cells. D11A4 had only weak reactivity in the subtrophoblastic stromal region of the placenta in frozen sections. aPL staining was also observed against extravillous cytotrophoblast. BA3B5C4 stained cytoplasmic structures, whereas 3SB9b stained the plasma membrane region with little cytoplasmic staining. These data suggest that the trophoblastic layer is reactive with aPLs and may potentially be directly damaged through mechanisms unrelated to thrombosis. In addition, the trophoblastic layer directly in contact with the maternal circulation is most reactive with aPLs that are PS+ rather than CL+. The differential reactivity of 3SB9b and BA3B5C4 suggests that the antigenic conformation involving PS on the cytotrophoblast is altered concurrent with fusion into the syncytium.

Expression of endogenous HIV-1 crossreactive antigens within normal human extravillous trophoblast cells.

Expression of intact endogenous retroviruses by normal placental villous trophoblast and immuno-crossreactivity of villous trophoblast with anti-retroviral antisera have been documented. The nature and/or potential function of these particles/proteins has not yet been fully defined. We previously reported that monoclonal antibodies directed against HIV-1 envelope and gag proteins react with normal human villous trophoblast. In this study, we report that extravillous trophoblast (EVT) from second- and third-trimester tissue are also cross-reactive with anti-HIV-1 gp120/160 and p17/18 antibodies. We document a differential expression of such cross-reactive epitopes between mononuclear EVT and placental bed giant cells. Mononuclear EVT principally displayed reactivity throughout the cytoplasm with little or no difference between cells, whereas placental bed giant cells displayed distinct localization of labeling to limited areas of cytoplasm. This pattern of reactivity apparently correlates with trophoblast morphological differentiation and with our earlier observations concerning villous trophoblast. These data illustrate that retrovirus-associated epitopes are expressed by trophoblast throughout the normal human placenta and that this distribution is related to morphologic differentiation of these cells.

Characterization of antigens expressed in normal baboon trophoblast and cross-reactive with HIV/SIV antibodies.

Electron microscopic studies have revealed the presence of endogenous retroviral (ERV) particles in normal primate placental tissues. These particles have ultrastructural similarities to type C retroviral particles and are mainly associated with the trophoblast. In normal human placental tissues, they have antigenic similarity with exogenous retroviruses, such as the human immunodeficiency virus (HIV), and may have a role to play in the regulation of cellular gene expression, syncytiotrophoblast formation or pregnancy-related immunosuppression. In this study, a panel of antibodies (polyclonal and monoclonal antibodies) against viral proteins (anti-HIV and anti-SIV) and endogenous retroviral (ERV) proteins were assessed by immunohistochemistry and immunoblotting, for their cross-reactivity with ERV particles isolated from normal baboon placental tissues. The antibodies (anti-HERV-K RT, anti-ERV3 env, anti-HIV-1 p17, anti-HIV-2 gp120) reacted positively with the syncytiotrophoblast and each antibody recognized one or two proteins of molecular weights (MW) 38, 58 or 64 kDa present in the baboon placental villous tissues and SIV-infected molt-4 Cl8 cells, but not in uninfected cells. The results of this study confirm the specific expression of retroviral cross-reactive antigens in normal baboon placental tissues and suggest placental cellular proteins may have antigenic similarity with those recognized by anti-HIV/SIV antibodies. The role of these retroviral-related proteins expressed at the maternal-fetal interface remain unclear.

Immunologic aspects of recurrent abortion and fetal death.

Although mechanisms that prevent rejection of the conceptus are incompletely understood, recent evidence suggests that maternal immunologic aberrations may cause repeated abortions. Autoimmune conditions associated with antiphospholipid antibodies sometimes produce vascular abnormalities in the decidua and placenta; successful pregnancies can be achieved in most of these women by treatment with corticosteroids and low-dose aspirin. Abnormal maternal immune responses to paternal or trophoblast alloantigens, with insufficient production of blocking antibodies or suppressor cells, have also been implicated. Immunization of these patients with paternal or third-party leukocytes has resulted in a number of live births. However, the mechanism of action, effectiveness, and safety of all treatment regimens remain controversial. Present recommendations for the evaluation of recurrent-abortion patients include diagnosis and appropriate treatment of traditional nonimmunologic and more recently recognized autoimmune factors. The remaining patients with no detectable cause for repetitive pregnancy loss are candidates for referral to research centers for further immunologic evaluation and experimental immunotherapy.

The association of antiphospholipid antibodies with severe preeclampsia.

Over a 3-year period, we studied 43 women who presented with severe preeclampsia prior to 34 weeks' gestation. Seven (16%) had significant levels of antiphospholipid antibodies, whereas none of the normotensive controls of similar gestational age had antiphospholipid antibodies (P less than .001). Three of the seven women with antiphospholipid antibodies suffered the following complications during the peripartum period: 1) cerebral infarction and episodes of transient monocular blindness; 2) pulmonary embolism, deep venous thrombosis, and an autoimmune flare in the postpartum period; and 3) transient monocular blindness and amnesia after delivery. Our experience suggests that antiphospholipid antibodies are found in a substantial proportion of cases of early-onset severe preeclampsia and have important clinical implications. We suggest that patients with early-onset severe preeclampsia be screened for antiphospholipid antibodies; if antibodies are detected, these women should be considered for prophylactic anticoagulation therapy.

A new postpartum syndrome associated with antiphospholipid antibodies.

Three women with antiphospholipid antibodies and a postpartum syndrome of pleuropulmonary disease, fever, and cardiac manifestations are presented. Each patient had either lupus anticoagulant or anticardiolipin antibodies or both, but did not have antinuclear antibodies or fulfill the criteria for the diagnosis of systemic lupus erythematosus. No infection or embolus was detected that could explain the pulmonary findings. All three patients had electrocardiographic abnormalities, and one patient developed a cardiomyopathy with extensive immunoglobulin G (IgG), IgM, IgA, and C3 deposition in the myocardium. In addition to the reported association between antiphospholipid antibodies and fetal loss, fetal growth retardation, and preeclampsia, we suggest that patients with antiphospholipid antibodies are at risk for a previously unreported and serious autoimmune postpartum syndrome.