Structural changes occurring during atresia in sheep ovarian follicles

The structural changes that characterize primary, secondary and tertiary atresia in sheep Graafian follicles have been studied by means of histological, histochemical and ultrastructural techniques. In primary atresia vacuoles representing swollen endoplasmic reticulum are prominent along the antral border together with disorganized granulosa cells containing pyknotic nuclei. Phagocytic cells, which increase in number as atresia progresses, were seen within the membrana granulosa and are considered to be transformed granulosa cells. Even in follicles classified as nonatretic, a few antral vacuoles and occasional pyknotic nuclei are present. During secondary atresia there is a large increase in the number of cells with pyknotic nuclei; many of these nuclei had been extruded and had fused to form the characteristic Feulgen-positive atretic bodies found along the edge of the antral cavity. These bodies usually have a diameter of up to 15 mum but occasionally reached as much as 400 mum. A second area of degeneration is frequently present in the membrana granulosa, two or three cell layers from the basal lamina, and it is at this level that exfoliation of granulosa cells occurs in tertiary atresia. In contrast to the membrana granulosa, there are during secondary atresia, only slight indications of degeneration in the cumulus. In tertiary atresia the membrana granulosa is highly disorganized; the atretic bodies are often fewer in number than at earlier stages. The basal lamina remains essentially intact. It is at this stage that the first clear signs of degeneration occur in the theca interna. Despite some disintegration of the cumulus, the integrity of the oocyte is maintained and its nucleus remains vesicular. Changes in the thecal microcirculation may plan a key role in atresia: adjacent to the basal lamina of non-atretic follicles, there is a well-developed capillary network which is significantly reduced as atresia progresses.     

Expression of androgen receptors and steroidogenic enzymes in relation to follicular growth and atresia following ovulation in pigs

Among the new antral follicles that develop after ovulation in pigs, the incidence of atresia, based on granulosa cell apoptosis, increases between Days 5 and 7 of the estrous cycle. The purpose of this study was to determine how follicular growth and atresia affected the expression of some key enzymes regulating follicular steroidogenesis and androgen receptor on Days 3, 5, and 7 after the onset of estrus. Ovaries were frozen in liquid propane for subsequent sectioning and immunohistochemical analysis. Ninety-six follicles were classified according to size as small (< 3 mm), medium (3-5 mm), or large (> 5 mm). Follicles in the active stages of the cell cycle were identified by the presence of the cell proliferation-associated nuclear antigen Ki-67 in granulosa cells. Follicles with apoptotic cells were identified by in situ 3'-end labeling of DNA. Staining intensity of antigens on sections was assigned a numeric value (0-3). Follicles assigned a value > 1 for 3'-end labeling in their granulosa cells were classified atretic. The percentage of atretic follicles increased (p < or = 0.05) from 5% on Days 3 and 5 to 41% on Day 7. Expression of Ki-67 in granulosa cells was more strongly (p < or = 0.05) associated with nonatretic follicles (98% expressing) than with atretic follicles (41% expressing). Aromatase cytochrome P450 (P450arom) was localized predominantly in granulosa cells of nonatretic follicles and was undetectable in atretic follicles. Androgen receptor in granulosa cells and expression of P450 17 alpha-hydroxylase/C17-20 lyase (P450c17) in theca interna were lower (p < or = 0.001) in atretic follicles than in nonatretic follicles. The expression of 3 beta-hydroxysteroid dehydrogenase (3 beta HSD) was localized to the theca interna and was unaffected by follicle atresia. In nonatretic small follicles, the expression of P450arom and P450c17 decreased (p < 0.01) between Days 3 and 7 while expression of Ki-67 was unchanged. In nonatretic follicles, increased follicle size was associated with a decrease (p < 0.01) in androgen receptor expression and increases (p < 0.01) in P450arom, P450c17, and 3 beta HSD expression. In conclusion, increased expression of steroidogenic enzymes was associated with follicular growth. Loss of P450arom expression in vivo is an early event in atresia and is followed by decreased cell proliferation, and decreased expression of androgen receptor and P450c17.     

Expression of steroidogenic enzyme and gonadotropin receptor genes in bovine follicles during ovarian follicular waves: a review

Ovarian follicular development in cattle is characterized by waves of growth during the prepubertal and postpartum periods and during estrous cycles. Each wave of follicular growth is characterized by recruitment of a cohort of follicles 4 to 5 mm in diameter. From the cohort, one follicle is selected for continued growth and becomes dominant. If luteolysis occurs during the growth phase of dominant follicles, final maturation and ovulation occurs. If luteolysis does not occur during the growing and maintenance phase of follicles, the fate is atresia. Changes in mRNA expression for the gonadotropin receptors (FSHr and LHr), key steroidogenic enzymes (cytochrome P450 side chain cleavage [P450scc], cytochrome P450 17alpha-hydroxylase-[P450c17], cytochrome P450 aromatase [P450arom], and 3beta-hydroxysteroid dehydrogenase [3beta-HSD]), and growth factors (IGF-I and -II) and their binding proteins (IGFBP) have been associated with different stages of follicular growth and atresia. In general, expression of mRNA for the gonadotropin receptors, steroidogenic enzymes, and steroidogenic acute regulatory protein (StAR) increase with progressive follicular development and is highest when dominant follicles approach maximum size. Expression of mRNA declines rapidly and becomes low or undetectable in atretic follicles. The IGF-I (granulosal cells) and IGF-II (thecal cells) are increased, whereas IGFBP-2 (granulosal cells) is reduced, in dominant follicles. Recruitment of a cohort of follicles is associated with initiation of expression of mRNA for P450scc and P450arom in granulosal cells. Selection of dominant follicles is associated with expression of mRNA for LHr and 3beta-HSD in granulosal cells. Thus, changes in gene expression likely are important to recruitment, selection, dominance, and atresia in ovarian follicles.     

Mechanical isolation and in vitro growth of preantral and small antral human follicles

OBJECTIVE: To develop a procedure for isolating small human follicles and to determine their growth requirements. DESIGN: Preantral and early antral follicles were isolated manually, allocated randomly to experimental groups, and cultured for a few weeks. SETTING: Patients giving informed consent in hospitals. PATIENT(S): Women undergoing laparotomy or oophorectomy. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Follicular size, E2, histology. RESULT(S): Human FSH (at a dose of 1.5 U/mL) induced antral growth of follicles, and the addition of human LH (2.5 ng/mL) to human FSH stimulated growth and antral development. Histologic studies showed that most of the early antral follicles did not contain an oocyte and already had begun to undergo atresia before culturing. Levels of E2 increased in the incubation medium as the follicles increased in size, but those levels were significantly greater when the follicles contained oocytes. CONCLUSION(S): It is possible to grow small human follicles after they have been isolated manually. To develop successfully, they require a low concentration of human LH in addition to human FSH. The rate of atresia between the preantral and early antral stages in vivo is very high; therefore, it is worthwhile to develop techniques for isolating and culturing the follicles before the antral stages.     

Testosterone production by isolated rabbit follicles in the presence of luteinizing hormone and inhibitors of steroidogenic enzymes

Rabbit ovarian follicles were incubated with luteinizing hormone (LH) and various inhibitors of steroidogenesis in order to determine what enzymes were necessary for testosterone production. Incubations were carried out in minimum Eagle's medium: normal rabbit serum: 95:5 with medium being changed every 15 min and stored at -15 degrees C until assayed for testosterone by radioimmunoassay. Inhibitors and LH were added at different times after the start of the incubations. Of the inhibitors tested only SU 10603, an inhibitor of the 17alpha-hydroxylase enzyme completely prevented the testosterone response to LH while aminoglutethimide (inhibitor of 20alpha-hydroxycholesterol dehydrogenase) and U 30870 (inhibitor of 3beta-hydroxysteroid dehydrogenase) only showed partial inhibition. These results suggest that cholesterol is an obligatory intermediate for the production of testosterone by rabbit ovarian follicles and that normal LH stimulated endocrine function can resume after treatment with inhibitors. The results are also in agreement with previous data using inhibitors of protein synthesis in the presence of LH.     

Intrafollicular insulin-like growth factor-binding protein levels in equine ovarian follicles during preovulatory maturation and regression

The profiles of insulin-like growth factor-binding proteins (IGFBPs) in follicular fluid have been characterized in a number of mammals (rats, pigs, sheep, cattle, humans) and are good indicators of follicular status. We studied the IGFBP profiles of equine serum and ovarian follicular fluid recovered at various stages of the follicular phase. The levels of IGFBPs were related to the morphology and the steroidogenic activity of the follicles. Follicular fluids were recovered by ultrasound-guided follicular aspiration. In the first experiment, the dominant follicles of 10 mares were partly punctured (aspiration of 0.5-2.2 ml of fluid) once at the early dominant stage (22-25 mm in diameter) and a second time at the preovulatory stage (PO), 34 h after induction of ovulation. Among these 10 PO follicles, 5 were classified as healthy whereas the other 5 were classified as hemorrhagic, as assessed by ultrasonic morphology and subsequent ovulation or not. In another group of mares (n = 5), the largest follicle was punctured once at the late dominant stage (33-35 mm in diameter) and then at the PO stage, 34 h after induction of ovulation. Serum was prepared at each puncture session. In the second experiment, follicular fluid was recovered from the dominant and contemporary cohort subordinate follicles (n = 5 mares). Samples were individually assayed for estradiol-17beta and progesterone content by RIA, and IGFBPs were studied by using Western ligand blotting and densitometry. Equine serum and follicular fluid displayed IGFBP at 42-44 kDa (likely corresponding to IGFBP-3), 28-32 kDa (likely corresponding to IGFBP-5), 24 kDa (likely corresponding to IGFBP-4), and 35 kDa, identified as IGFBP-2 by immunoblotting, plus one band at 120 kDa. IGFBP were clearly more abundant in serum than in fluid from healthy follicles. In the follicular fluid, 42-44-kDa IGFBP was the major binding protein, and its level was almost constant at the various physiological statuses studied. Follicular development of the dominant follicle in each mare was characterized by a decrease in intrafollicular IGFBP-2 and 28-32-kDa IGFBP levels before LH stimulation and by an increase in IGFBP-2 after LH stimulation. Follicular regression of large follicles, as well as subordinate ones, was characterized by a low level of intrafollicular estradiol-17beta and was associated with an increase in IGFBP-2, 24-kDa IGFBP, and 28-32-kDa IGFBP intrafollicular levels. Taking these results together, we have demonstrated clear correlations between the intrafollicular levels of estradiol-17beta and IGFBP-2 and 28-32-kDa IGFBP. Therefore, follicular growth and regression in the mare are associated with specific changes in IGFBP levels. These changes could be of crucial importance for follicular development in ovulation or atresia.     

Characteristics of prolonged dominant versus control follicles: follicle cell numbers, steroidogenic capabilities, and messenger ribonucleic acid for steroidogenic enzymes

Cattle with low (subluteal) levels of plasma progesterone develop a persistent dominant follicle; plasma estradiol and LH pulse frequency are elevated, and fertility subsequent to the ovulation of a prolonged dominant follicle is compromised. The hypotheses were 1) that prolonged dominant follicles produce more estradiol because they have theca and granulosa cells with an enhanced capacity to produce androgen and estradiol, respectively, and 2) that these changes in steroidogenic capacity are paralleled by concomitant changes in mRNA for the appropriate steroidogenic enzymes. Prolonged dominant follicles were induced by treating Holstein heifers with exogenous progesterone via an intravaginal controlled internal drug-release device (CIDR) from Day 14 to 28 of the cycle. Prolonged dominant follicles were collected just before (CIDRb, Day 28; n=4) or 24 h after (CIDRa, Day 29; n=4) CIDR removal, and their steroidogenic capacity was compared to that of growing, control dominant follicles obtained just before (CONTb, n=4) or 24 h after (CONTa, n=4) a luteolytic injection of prostaglandin F2alpha during the late luteal phase. After natural luteolysis, CIDR heifers maintained subluteal concentrations of progesterone (1-2 ng/ml) and had higher estradiol and LH pulse frequency than control heifers, as expected. In CIDR heifers, prolonged dominant follicles were present on the ovary for a longer time, reached a larger diameter, and had more granulosa cells and a larger mass of theca than dominant follicles from control heifers (p < 0.05). Concentrations of steroids in follicular fluid, estradiol secretion by granulosa cells in vitro, and levels of mRNA for steroidogenic enzymes in theca and granulosa cells provided no evidence for greater capacity of theca and granulosa cells of CIDR follicles to produce androgen and estradiol. In fact, follicular fluid estradiol and mRNA for P450 aromatase were higher after luteolysis than before in control animals (p < 0.05) but not after CIDR removal in treated animals. Therefore, the data do not support the hypotheses. Rather it is suggested that prolonged dominant follicles produce more estradiol because they have more granulosa cells and a larger mass of theca than control dominant follicles. In contrast, progesterone concentrations in the follicular fluid increased in CIDRa relative to CIDRb follicles (p < 0.05), a change that did not occur in control follicles; and granulosa cells from CIDRa follicles secreted more progesterone than granulosa cells from any other group. The increased capacity of CIDRa follicles to secrete progesterone suggests premature luteinization, which could contribute to decreased fertility in cattle that ovulate a prolonged dominant follicle.     

Selection, dominance and atresia of follicles during the oestrous cycle of heifers

This study examined the correlation between measurement of follicle growth by ultrasound, and measurement of intrafollicular ratios of oestradiol and progesterone concentrations and the serum concentrations of FSH during selection, dominance and atresia or ovulation of dominant follicles in heifers. Heifers were ovariectomized on days 0 (before LH surge), 1 (after LH surge, preovulation), 1 (postovulation), 3, 6 and 12 of the oestrous cycle. Blood samples were collected at 4-6 h intervals. After ovariectomy all follicles > or = 5 mm were measured and follicular fluid was aspirated. Follicles were classified by size according to ultrasound (F1, largest; F2, second largest; F3, all remaining follicles > or = 5 mm) and by the ratio of oestradiol:progesterone concentrations. During the follicular phase, a single dominant oestrogen-active follicle increased in diameter while serum concentrations of LH increased and FSH decreased (P < 0.05). On day 1 (after LH surge, preovulation), serum LH and FSH decreased to pre-surge concentrations (P < 0.0001), while follicle size and intrafollicular progesterone concentration increased and oestradiol concentration decreased (P < 0.05). A dominant nonovulatory follicle, classified as oestrogen-active on days 1, 3 and 6 and oestrogen-inactive on day 12, increased in size from day 1 to day 7 and lost dominance during days 10-12, coincident with the growth of multiple oestrogen-active follicles. The serum FSH concentration increased transiently (P < 0.05) before each new wave of dominant follicular growth. The overall correlation of ultrasound measurements of follicle diameter with measures of follicle size after ovariectomy was high.(ABSTRACT TRUNCATED AT 250 WORDS)     

Localisation of insulin-like growth factor receptors in skin follicles of sheep (Ovis aries) and changes during an induced growth cycle

Pelage growth cycles are regulated by circulating prolactin in many mammals, but the intercellular mediators of this signaling are unknown. Binding sites for insulin-like growth factors (IGFs) were examined in sheep skin to show changes in distribution and abundance of IGF receptors associated with a prolactin stimulus and the subsequent hair follicle growth cycle. Follicle cycles were induced in New Zealand Wiltshire ewes by a surge in plasma prolactin following a 4-month period of prolactin suppression with bromocriptine. Eight treated and three control sheep were slaughtered at intervals over 43 days during the follicle growth cycle. At 12-20 days after the elevation of prolactin, wool follicles passed through brief catagen and telogen phases, followed by a return to anagen. IGF binding sites were localized in skin sections by incubation with 125I-IGF-I or 125I-IGF-II. Displacement with competitive binding inhibitors (unlabeled IGF-I, IGF-II, des(1-3)IGF-I, des(1-6)IGF-II, or insulin) and affinity cross-linking showed that these binding sites were predominantly IGF type 1 and type 2 (mannose-6-phosphate) receptors. The radioligands bound especially to follicle germinal cells and prekeratinocytes. Increases in specific binding of both radioligands were observed after the rise in prolactin, but prior to anatomical changes in follicles associated with cessation of growth. For IGF-I, highest binding density was observed during catagen in the germinal matrix and dermal papilla cells. For IGF-II, peak density occurred during late anagen/early catagen in the germinal matrix and during telogen in the dermal papilla. These cycle associated changes in receptor availability suggest that IGF receptors are involved in control of the wool growth.     

HIV-I induced destruction of neocortical extracellular matrix components in AIDS victims

Neurological dysfunction is not uncommon in patients suffering from acquired immunodeficiency syndrome (AIDS) and, when manifested, intimates involvement of the central nervous system. Here, the human immunodeficiency virus (HIV) infects preferentially microglial cells, which thereby release substances known to interfere with neuronal function. One class of agents set free in this manner are proteases; these degrade certain components within, and thereby undermine the integrity of, the extracellular matrix (ECM) compartment, which plays a vital role in cell-to-cell communication. We wished to ascertain whether the ECM compartment is indeed disrupted in the brains of AIDS victims. We examined the neocortical areas of 27 AIDS autopsy cases, including 9 with diagnosed HIV-encephalopathy (HIVE); 8 HIV-seronegative cases with various types of brain lesion, including viral infections, were also included in this study. HIV-antigens and DNA were identified by use of immunohistochemistry and in situ hybridization, and ECM components by lectin staining and immunohistochemistry. Of the 27 AIDS cases examined, each of the 9 with HIVE was completely devoid of labeled ECM components; 8 of the 18 without HIVE had incurred substantial losses, and only 2 manifested a normal complement of constituents within this compartment. With respect to stratal and topographic variations, layers II and III were less affected than layers V to VII, as was the frontal cortex relative to other areas. These findings confirmed our expectations of the brain's ECM undergoing degradation following HIV infection, and these changes may well underlie the neurological disturbances manifested in AIDS patients.     

Stimulation of extracellular matrix components in the normal brain by invading glioma cells

Malignant gliomas are characterized by an extensive invasion of tumor cells into the normal brain parenchyma. A substantial amount of data indicates that cell movement in general is regulated by specific interactions between extracellular matrix components and specific cell-surface receptors. In the present work, multicellular spheroids from 4 human glioma cell lines (U-373Mg, A-172Mg, U-251Mg and HF-66) were confronted with normal rat brain cell aggregates in vitro, which resulted in a progressive invasion of tumor cells into the brain aggregates. The co-cultures were then sectioned and immuno-stained for specific extracellular matrix components (laminin, fibronectin and collagen type IV) and for specific cell-surface receptors which bind to these components (integrins beta1, beta4, alpha3, alpha6). In addition, flow-cytometric measurements and Northern blot analyses showed expression of several different integrins within the cell lines. The alpha3 subunit was expressed strongly in all cell lines. Whereas the beta1 subunit was expressed weakly in exponentially growing monolayer cultures, it showed a pronounced expression in multicellular spheroids, indicating that the integrin expression may vary depending on the micro-environment within a tumor. Furthermore, normal brain tissue was able to produce laminin when confronted with the glioma cells, which also was observed for fibronectin and collagen type IV. The relevance of our observations to the in vivo situation was investigated further by immuno-staining 5 human glioma biopsy samples for laminin. In some areas of the tumors, specific deposits of laminin were observed. In conclusion, we have shown that normal brain tissue has the ability to produce extracellular matrix components, such as laminin, collagen type IV and fibronectin, when confronted with invading glioma cells. Our results show that the glioma cells express specific integrins which can interact with these extracellular matrix components. Such interactions may facilitate tumor cell migration and invasion.     

Protein interactions during assembly of the enamel organic extracellular matrix

This paper describes a simple and sensitive high-performance liquid chromatographic (HPLC) method for the detection of human globin chains in blood and bloodstains. The method involves direct injection of the filtered samples of dilute hemolysates or bloodstain extracts onto a microbore C4 reversed-phase column (2.1 mm I.D.) with UV detection at 220 nm. Microbore HPLC offers a significant improvement in sensitivity with little loss of the resolution of globin chains and only small variations in the determination of gamma chain composition. The detection limit of hemoglobin (Hb) was 0.1 microgram, which is equivalent to about 1 nl of fresh whole blood. Umbilical cord blood could be differentiated from adult blood in stains that were up to twenty weeks old, by the presence of gamma globin chains. The present method will be useful for detection of abnormal Hbs and for the determination of gamma chain composition in clinical laboratories, as well as in the practice of forensic science for the analysis of minute amounts of blood and bloodstains.     

Quantitation of nexus junctions in the granulosa cell layer of rabbit ovarian follicles during ovulation

Electron microscopy was used in a semi-quantitative study to determine changes in the abundance and size of surface nexuses and changes in the abundance of interiorized nexuses in growing and mature ovarian follicles during the ovulatory process. Mature follicles contain larger granulosa cells than follicles in the early stage of antral formation. Also, the granulosa cells of mature follicles have a slightly greater number of surface nexuses (without a change in nexus length), and more interiorized nexuses, compared to immature follicles. As a mature follicle approaches rupture, there is an appreciable decrease in the number of surface nexuses per granulosa cell. There is also a slight reduction in the number of interiorized nexuses at this time. It is concluded that this decrease in both surface nexuses and interiorized nexuses may be a consequence of ovulatory changes during which the rate of granulosa cell division is greater than the rate of formation of new nexuses. Additionally, the disruption to cell-to-cell cohesion during the ovulatory process appears to be independent of the interiorization of surface nexuses.     

Altered deposition of extracellular matrix components in the skeletal muscle and lymph node of the MDX dystrophic mouse

1. MDX mice derived from a colony of C57BL/10ScSn mice develop an X-linked recessive muscular dystrophy, thus providing an adequate model to study the pathogenesis of muscular dystrophy. 2. Skeletal myofibers of MDX mutant mice were heterogeneous, with disorganization of myofilaments and the absence of immunolabelling for dystrophin with monoclonal antibody DY4/6D3. 3. Marked deposition of reticulin, collagenic fiber (types, I, IV) and laminin (LN) were consistently present mostly around lesioned and necrotic myofibers associated with an intense inflammatory reaction, whereas strong immunolabelling for TIII-C, TIV-C and FN was often associated with regenerated fibers. 4. During the onset (3 weeks of postnatal life) of disease and height of myonecrosis (5-6 weeks of postnatal life), popliteal lymph nodes showed dense argyrophilic meshwork, intense immunolabelling for collagens types I and IV, FN, LN and enlargement of the hili which were packed with mononuclear cells. Such alterations, albeit less intense, were still observed in MDX mice with 20 weeks of postnatal life. 5. The results support the view that ECM components might be influencing the migration of inflammatory cells and the process of myonecrosis in the skeletal muscle of MDX dystrophic mice.     

Immunohistochemical localization of epidermal growth factor receptors, epidermal-growth-factor-like and transforming-growth-factor-alpha-like peptides in chicken ovarian follicles

The purpose of this study was to determine the presence of epidermal growth factor receptor and its potential ligands epidermal growth factor (EGF) and transforming growth factor alpha (TGF-alpha) in the tissues of the maturing follicles in the ovary of laying ISA-Brown hens using peptide-specific immunohistochemical methods. Cryostat sections, 6-8 microns thick, were made from fresh-frozen tissues of F1-F4 (largest to fourth largest) and large white follicles and they were immunostained for epidermal growth factor receptor, epidermal growth factor or transforming growth factor alpha using specific polyclonal antibodies. The EGF receptor and both ligands were detected in the granulosa, theca interna and theca externa layers of the follicles. The EGF receptor was localized both in the plasma membrane and cytoplasm of all cell types. EGF was predominantly cytosolic, whereas TGF-alpha was found in the plasma membranes and perinuclear areas of all cell types. The concentration of the receptor and both ligands decreased with follicular maturation. This observation is consistent with our previous observation that the response to EGF and TGF-alpha decreases as follicles mature, and thus provides further evidence that the receptor or the ligands may have a regulatory role in avian ovarian function.     

Detection of apoptosis in human and rat ovarian follicles

OBJECTIVE: The purpose of this study was to determine whether cells acquired from individual human preovulatory follicles undergo apoptosis (physiologic cell death) and, if so, to correlate the degree of apoptosis with characteristics of the follicles or the oocytes derived from the follicles. METHODS: We devised a sensitive nonradioactive method for detecting apoptotic DNA fragmentation in small numbers of cells derived from rat atretic follicles and follicular aspirates of patients undergoing assisted reproductive technologies. RESULTS: Using this method, apoptotic DNA was detected in rat atretic follicles, with optimal detection at 10-100 ng. Furthermore, apoptotic DNA was detected in some, but not all individual human follicular aspirates from several patients, and was found in follicles that produced oocytes that fertilized and developed into embryos. CONCLUSION: Apoptosis occurs in cells from human ovarian preovulatory follicles and may be a normal physiologic process of the follicle during luteinization.     

CD44 plays a role in adhesive interactions between glioma cells and extracellular matrix components

To evaluate how individuals infected with the human immunodeficiency virus (HIV) became aware of their infection, when they first suspected they were infected with HIV and factors associated with suspecting HIV infection, we surveyed 227 patients at an urban outpatient HIV clinic. Though nearly all patients acknowledged risk factors for HIV, 60% reported that they did not suspect that they were infected until they received a positive HIV antibody test result. Non-white patients were less likely to suspect HIV infection prior to testing than white subjects (p < 0.03). Subjects not suspecting infection more often received HIV testing through a screening program or during a medical encounter (p = 0.02) and were less likely to be told by others that they might be infected (p = 0.001) than patients suspecting infection prior to testing. Forty-eight percent of subjects who suspected HIV infection prior to testing waited one year or more before obtaining their HIV antibody test. Interventions to reduce faulty personal HIV risk perception are needed to promote earlier HIV diagnosis.