Non-Hodgkin's lymphomas of childhood: an analysis of the histology, staging, and response to treatment of 338 cases at a single institution.

Between 1962 and 1986, a total of 338 consecutive newly diagnosed children and adolescents with non-Hodgkin's lymphomas (NHLs) were evaluated and treated at St Jude Children's Research Hospital (SJCRH). Median follow-up is 6.6 years (range, 1.8 to 23 years). The patients ranged in age from 7 months to 21 years (median, 10 years), and 71% were males. All cases were staged (I to IV) by a clinical staging system. Eighteen percent were stage I, 21% stage II, 43% stage III, and 18% stage IV. Cases frankly leukemic at diagnosis (ie, greater than 25% marrow blasts) were excluded from the analysis. Pathologic material from all cases was reviewed and classified according to the Working Formulation. The histologic distribution of cases was as follows: 38.8% diffuse small non-cleaved cell (undifferentiated, Burkitt's and non-Burkitt's); 26.3% diffuse large-cell, mainly immunoblastic; 28.1% lymphoblastic; and 6.8% other. Treatment policy evolved over time to a stage- and histology-specific strategy for treatment assignment, and overall results significantly improved by era from 37% (+/- 5%) 2-year event-free survival (EFS) for patients treated before 1975 to 77% (+/- 4%) since 1978. By univariate and multivariate Cox regression analyses, the era of treatment (hence, the protocol-specific treatment itself), the stage, and the log of the initial serum lactic dehydrogenase (LDH) emerged as the most powerful prognostic indicators, while histology per se was not significantly related to outcome. For the 154 patients treated since 1978, the 2-year EFS by stage was 97% (+/- 3%) for stage I, 86% (+/- 6%) for stage II, 73% (+/- 6%) for stage III, and 47% (+/- 11%) for stage IV (P less than .0001). Compared with our previous experience, we conclude that the cure rate of childhood NHL has doubled in the last decade with modern management.

Treatment of localized primary non-Hodgkin's lymphoma of bone in children: a Pediatric Oncology Group study.

PURPOSE: The treatment of primary lymphoma of bone (PLB) in children has traditionally included radiotherapy to the primary site; more recently, it has included systemic chemotherapy. Because of concern about the untoward effects of treatment in a disease that is curable, we attempted to determine whether radiotherapy can be safely excluded from treatment. PATIENTS AND METHODS: The results of three consecutive Pediatric Oncology Group (POG) studies were examined to determine the impact on outcome of radiotherapy as adjunctive treatment in children and adolescents receiving chemotherapy for early-stage primary lymphoma of bone. RESULTS: From 1983 to 1997, 31 patients with localized PLB were entered onto POG studies of early-stage non-Hodgkin's lymphoma (NHL). Between 1983 and 1986, seven patients were treated with 8 months of chemotherapy with irradiation (XRT) of the primary site. After 1986, patients were treated without XRT; four received 8 months of chemotherapy, and 20 received 9 weeks of chemotherapy. Primary sites were the femur (nine), tibia (eight), mandible (five), mastoid (one), maxilla (one), zygomatic arch (one), rib (one), clavicle (one), scapula (one), ulna (one), talus (one), and calcaneous (one). Histologic classification revealed 21 cases of large cell lymphoma, five cases of lymphoblastic lymphoma, two cases of small, noncleaved-cell lymphoma, and three cases of NHL that could not be classified further. One patient relapsed at a distant site 22 months after completion of therapy. There have been no deaths. CONCLUSION: Localized PLB is curable in most children and adolescents with a 9-week chemotherapy regimen of modest intensity, and radiotherapy is an unnecessary adjunct.

Randomized comparison of pentostatin versus interferon alfa-2a in previously untreated patients with hairy cell leukemia: an intergroup study.

PURPOSE: Therapy of hairy cell leukemia has markedly improved. Interferon alfa-2a and pentostatin are active agents. The National Cancer Institute organized an intergroup trial to compare these agents prospectively in untreated patients. METHODS: Patients were randomized to receive either interferon alfa-2a (3 x 10(6) U subcutaneously three times per week) or pentostatin (4 mg/m2 intravenously every 2 weeks). Patients who did not respond to initial treatment were crossed over. RESULTS: Of 356 patients on study, 313 were eligible. Among interferon patients, 17 of 159 (11%) achieved a confirmed complete remission and 60 of 159 (38%) had a confirmed complete or partial remission. Among pentostatin patients, 117 of 154 (76%) achieved a confirmed complete remission and 121 of 154 (79%) had a confirmed complete or partial remission. Additional patients achieved criteria for complete remission, but lacked confirmatory follow-up evaluation. Response rates were significantly higher (P < .0001) and relapse-free survival was significantly longer with pentostatin than interferon (P < .0001). The median follow-up duration is 57 months (range, 19 to 82). Myelosuppression was more frequent with pentostatin (P = .013). A multivariate logistic regression analysis of the confirmed complete remissions on pentostatin showed the following factors to be important for achieving a complete remission: high hemoglobin level (two-tailed P = .024), young age (P = .0085), and no or little splenomegaly (P = .0029). CONCLUSION: Both agents were well tolerated. Pentostatin produced higher response rates, and the responses were durable. Patient age and clinical status had an impact on outcome with pentostatin. Pentostatin is effective therapy for hairy cell leukemia.

B-cell lineage confers a favorable outcome among children and adolescents with large-cell lymphoma: a Pediatric Oncology Group study.

PURPOSE: The goal of this study was to assess the immunophenotype of uniformly treated cases of pediatric large-cell non-Hodgkin's lymphoma (NHL) to determine the prognostic importance of B-cell and T-cell lineages and of CD30 positivity. PATIENTS AND METHODS: Sixty-nine patients were analyzed by immunochemistry. All patients were classified histologically, staged in a uniform manner, and treated according to one of two protocols for localized (stage I and II) NHL or advanced (stage III and IV) large-cell NHL. Antibodies included anti-CD45, CD20, CD45Ra, MB-2 (not clustered), CD3, CD45Ro, CD43, CD15, CD30, and CD68. Statistical analysis used the exact conditional chi 2 and Kruskall-Wallace tests for clinical features and the log-rank test to evaluate event-free survival (EFS). RESULTS: Immunophenotypic results demonstrated 25 B-cell, 23 T-cell, and 21 indeterminate lineage. Twenty-seven patients expressed CD30 (17 T-cell and 10 indeterminate lineage), and of these, 22 showed histology of anaplastic large-cell lymphoma (ALCL). B-cell patients were older (P = .018) and showed more favorable survival than patients with T-cell or indeterminate lineage (96% EFS at 3 years, 96% v 67% and 74%, B v T and indeterminate lineage [P = .027]). B-cell lineage was seen more frequently in limited-stage patients, but was also associated with favorable survival when stratified for stage (P = .036). CD30 expression (P = .96) and ALCL histology (P = .90) did not show significant associations with survival. CONCLUSION: We conclude that among pediatric large-cell lymphomas, B-cell lineage is proportionately less frequent than in adults and CD30 antigen-expressing lymphomas are frequent among patients with T-cell and indeterminate lineage. B-cell phenotype tends to occur in older children and is associated with superior survival.

Large cell non-Hodgkin lymphoma of childhood: clinical characteristics and outcome.

Less is known about the clinical features and treatment outcome in pediatric large cell non-Hodgkin lymphoma (NHL) than the lymphoblastic and small noncleaved cell subtypes of NHL. To characterize presenting features and assess possible risk factors associated with this diagnosis, we analyzed data for 91 patients treated on a succession of multiagent regimens from 1975 to 1990. Five-year event-free survival (EFS) (+/- SE) was related to disease extent (St Jude system): stage I (n = 24), 95% +/- 5%; stage II (n = 20), 84% +/- 9%; stage III (n = 38), 50% +/- 10%; and stage IV (n = 9), 22% +/- 11%. Advanced stage disease, age < or = 5 years and serum LDH > 500 U/l were associated with poorer EFS in the univariate model (p < 0.001, 0.005, and 0.002, respectively). In the multivariate model, advanced stage and age retained prognostic significance (p = 0.001 and 0.02, respectively), but LDH did not. Among limited stage cases, age < or = 5 years was the only adverse risk feature (p = 0.016); treatment era (pre- vs. post-1979) was the only significant feature in patients with advanced disease (p = 0.004). Intrathoracic primaries were associated with a better outcome than other sites among the 38 stage III patients (p = 0.005). Only one of eight patients with bone marrow disease remains failure-free. The excellent results for limited stage pediatric large cell NHL permit consideration of treatment modifications to decrease toxicity; for cases with advanced disease, especially those with bone marrow involvement, novel therapeutic approaches are clearly needed.

Induction of apoptosis in oral cancer cells by an anti-bcl-2 ribozyme delivered by an adenovirus vector.

Human oral cancer cells may have any of several genetic changes, but the role of the bcl-2 oncogene is relatively unexplored. To find out if this gene plays a significant role and whether it could act as a target for gene therapy of oral cancer, we have examined the effects of an anti-bcl-2 ribozyme on the phenotype of oral cancer cells. A hammer-head ribozyme was designed to cleave the bcl-2 transcript after nucleotide 279 and was confirmed to be effective against a synthetic bcl-2 transcript. A gene encoding the ribozyme was cloned into an adenovirus vector and transferred to the human oral cancer cell lines 686LN, 1483, and Tu183. Over a 6-day period, the growth of each cancer cell line was reduced, whereas growth of the fibroblast cell line FS7 was less inhibited. Inhibition of the oral cancer cells could be attributed to apoptosis, as indicated by the detection of histone-associated DNA fragments in an immunoassay. Northern blots showed no detectable reduction in the level of bcl-2 mRNA of Tu183 cells, but Western blots showed a reduction of Bcl-2 protein at 24 h after infection with the ribozyme-expressing adenovirus vector. The results imply that (a) expression of the bcl-2 oncogene is necessary for the survival of oral cancer cells, (b) the bcl-2 gene transcript presents a target for gene therapy by ribozymes, and (c) an adenovirus vector is a suitable method for transfection of the ribozyme-expressing gene.

Deletion of Stat3 enhances myeloid cell expansion and increases the severity of myeloproliferative neoplasms in Jak2V617F knock-in mice.

The JAK2V617F mutation commonly found in myeloproliferative neoplasms (MPNs) induces constitutive phosphorylation/activation of the signal transducer and activator of transcription 3 (Stat3). However, the contribution of Stat3 in MPN evoked by JAK2V617F remains unknown. To determine the role of Stat3 in JAK2V617F-induced MPN, we generated Stat3-deficient Jak2V617F-expressing mice. Whereas expression of Jak2V617F resulted in a polycythemia vera-like disease characterized by increased red blood cells (RBCs), hematocrit, neutrophils and platelets in the peripheral blood of Jak2V617F knock-in mice, deletion of Stat3 slightly reduced RBC and hematocrit parameters and modestly increased platelet numbers in Jak2V617F knock-in mice. Moreover, deletion of Stat3 significantly increased the neutrophil counts/percentages and markedly reduced the survival of mice expressing Jak2V617F. These phenotypic manifestations were reproduced upon bone marrow (BM) transplantation into wild-type animals. Flow cytometric analysis showed increased hematopoietic stem cell and granulocyte-macrophage progenitor populations in the BM and spleens of Stat3-deficient Jak2V617F mice. Stat3 deficiency also caused a marked expansion of Gr-1+/Mac-1+ myeloid cells in Jak2V617F knock-in mice. Histopathologic analysis revealed marked increase in granulocytes in the BM, spleens and livers of Stat3-deficient Jak2V617F-expressing mice. Together, these results suggest that deletion of Stat3 increases the severity of MPN induced by Jak2V617F.

Aromatase expression by astrocytes after brain injury: implications for local estrogen formation in brain repair.

Recent evidence indicates that 17beta-estradiol may have neuroprotective and neuroregenerative properties. Estradiol is formed locally in neural tissue from precursor androgens. The expression of aromatase, the enzyme that catalyses the conversion of androgens to estrogens, is restricted, under normal circumstances, to specific neuronal populations. These neurons are located in brain areas in which local estrogen formation may be involved in neuroendocrine control and in the modulation of reproductive or sex dimorphic behaviours. In this study the distribution of aromatase immunoreactivity has been assessed in the brain of mice and rats after a neurotoxic lesion induced by the systemic administration of kainic acid. This treatment resulted in the induction of aromatase expression by reactive glia in the hippocampus and in other brain areas that are affected by kainic acid. The reactive glia were identified as astrocytes by co-localization of aromatase with glial fibrillary acidic protein and by ultrastructural analysis. No immunoreactive astrocytes were detected in control animals. The same result, the de novo induction of aromatase expression in reactive astrocytes on the hippocampus, was observed after a penetrating brain injury. Furthermore, using a 3H2O assay, aromatase activity was found to increase significantly in the injured hippocampus. These findings indicate that although astrocytes do not normally express aromatase, the enzyme expression is induced in these glial cells by different forms of brain injury. The results suggest a role for local astroglial estrogen formation in brain repair.

Influence of oestrogen on the initiation of nesting behaviour in female budgerigars.

Intact, breeding female budgerigars enter and occupy a nestbox 8-10 days before egg-laying. Ovariectomized budgerigars did not enter nestboxes. Oestradiol-17 beta monobenzoate (OB) (0-05 mg, i.m.) induced ovariectomized birds to enter nestboxes. Higher dosages (0-1 or 0-5 mg) did not facilitate this effect. The amount of time spent in nesting behaviour by all OB-treated groups did not differ from that shown by a group of intact females during the initial phase of the nesting sequence (first 4 days in the nestbox), but was significantly less than that of intact females undergoing later stages of nesting behaviour (4 days before and after laying the first egg). Oestradiol induced a increase in oviduct weight and size which was dose-dependent. With a low dose (0-05 mg OB), the oviducts were slightly larger than those of intact females which had been in the nestbox 1-2 days. With higher dose levels (0-1-0-5 mg OB), precursor albumen granules had formed in tubular glands of the magnum, a stage more typical of females which had been in the nestbox 4-6 days. The results indicate that low levels of oestrogen induce female budgerigars to enter nestboxes and initiate oviduct development. Subsequent phases of the nesting behaviour and oviduct development may be causally related to other hormones.

Effects of ovarian steroids and prolactin on the sequential development of nesting behaviour in female budgerigars.

As intact, breeding female budgerigars approach egg-laying, they spend an increasing amount of time in the nestbox and nest hollow. The brood patch area begins to defeather and becomes vascular and the oviduct increases in weight. Precursory albumen forms in the tubular glands of the oviduct. Oestradiol (OB) treatment in combination with prolactin (OB+PL) induced ovariectomized budgerigars to display nesting behaviour which did not differ from that shown by intact females in the 3 days immediately preceding egg-laying. In contrast, OB induced only the initial phase of the nesting sequence and the effects of OB in combination with progesterone (OB+PR) were intermediate between treatments with OB alone and OB+PL. Incubation of artificial eggs occurred only in the OB+PL group and the latency to display of the incubation posture was shorter in the OB+PL group than in the OB+PR group. No incubation posture was displayed by the OB-treated group. Oviduct development was not influenced by prolactin, but progesterone induced precocious development of tubular glands in the magnum region of the oviduct. Treatment with OB+PR induced uniform development of precursor albumen in the tubular glands. Development of the brood patch occurred with both OB+PL and OB+PR treatment. However, OB+PR resulted in defeathering which was advanced in relation to vascularity when compared with intact breeding females, whereas defeathering and vascularity of the OB+PL group did not differ from that of intact females at egg-laying. These results indicate that prolactin in combination with oestradiol was more effective than progesterone not only in inducing the later phases of nesting behaviour but also in initiating incubation behaviour and defeathering of the brood patch area.

Regulation of female brain aromatase activity during the reproductive cycle of the dove.

Oestrogen is formed in the female dove brain. The aim of this study was to determine whether (a) the catalytic properties of the brain aromatase are similar to the ovarian enzyme and (b) aromatase activity in the female brain changes during the reproductive cycle and is influenced by steroids and environmental stimuli. The results show that female preoptic aromatase has a higher substrate affinity than the enzyme in ovarian follicles (apparent Km: preoptic area, 7 nmol/l; ovarian follicles, 29 nmol/l), but a lower activity in the preoptic area (Vmax: preoptic area, 290 fmol/mg tissue per h; ovarian follicles, 843 fmol/mg tissue per h). In intact females with developing follicles, oestradiol-17 beta formation was higher in the posterior hypothalamus than the preoptic area. Females in a later stage of reproductive development (yolked follicles) had a different distribution of oestrogen formation with increased aromatase activity in the preoptic area. Preoptic and posterior hypothalamic aromatase activity of females paired with males for 10 days was positively correlated (r = 0.84, P = 0.0001; r = 0.75, P = 0.001 respectively) with ovarian development. Females with undeveloped ovaries which interacted with males had higher preoptic aromatase activity than visually isolated females with similar ovarian development, suggesting that behavioural stimuli have direct effects on brain aromatase activity which are independent of the ovary. Oestradiol benzoate treatment increased preoptic and posterior hypothalamic aromatase activity in intact and ovariectomized females, and testosterone propionate treatment increased anterior hypothalamic aromatase activity, but did not affect other areas, indicating that the distribution of induced aromatase activity is steroid-specific. Oestrogen treatment in ovariectomized or intact females did not replicate the maximal hypothalamic aromatase activity seen when the ovary contained yolked follicles. We conclude that brain aromatase activity is related directly to ovarian condition during the reproductive cycle of the female dove. As in the male, steroids have a role in the regulation of oestrogen formation in the female hypothalamus; behavioural stimuli are also likely to be involved in the control of the brain enzyme.

Sex differences in plasma concentrations of steroids during the sensitive period for brain differentiation in the zebra finch.

Changes in plasma concentrations of sex steroids were examined in male and female zebra finch chicks during the sensitive period for differentiation of sexually dimorphic brain nuclei associated with the control of song. Using a chromatographic separation procedure and radioimmunoassay, androstenedione, testosterone and 5 alpha-dihydrotestosterone were detected in plasma at relatively high concentrations immediately after hatching. There were no sex differences in concentrations of these androgens. An oestrogen, oestradiol-17 beta, which is known to differentiate the song-control system, is raised specifically in the circulating plasma of male zebra finch chicks, and not in females. The surge in oestradiol, which occurs during the first week after hatching, coincides with the period when capacity for differentiation of the song system is maximal. Exposure of the male brain to oestradiol-17 beta could trigger neuronal differentiation.

Inhibition of hypothalamic aromatase activity by 5 Beta-dihydrotestosterone.

Abstract A variable amount of circulating testosterone that reaches brain cells is converted to biologically inactive 5beta-reduced metabolites, namely, 5beta-dihydrotestosterone (5beta-DHT) and 5beta-androstane-3alpha,17beta-diol (5beta,3alpha-diol). In avian species, the production of inactive 5beta-DHT and 5beta,3alpha-diol is highest during embryonic and post-hatching life. In the present study, we have investigated the possibility that 5beta-reduction may not only correspond to a steroid inactivation pathway, but that 5beta-reduced metabolites of testosterone may exert direct inhibitory effects on enzymatic pathways producing biologically active steroids. When added to hypothalamic homogen-ates prepared from adult male doves, 5beta-DHT but not 5beta,3alpha-diol inhibits the activity of the aromatase enzyme, which converts testosterone to 17beta-oestradiol. During the first days after hatching, when the production of 5beta-reduced metabolites is high, the hypothalamic aromatase is also inhibited by 5beta-DHT. We conclude that a high 5beta-reductase activity during sensitive periods for sexual differentiation may protect the avian brain from the differentiating effects of circulating androgens by inhibiting the production of oestrogen.

High-dose therapy and autologous stem cell transplant does not result in long-term disease-free survival in patients with recurrent chemotherapy-sensitive ALK-negative anaplastic large-cell lymphoma.

Primary systemic anaplastic lymphoma kinase (ALK)-negative anaplastic large-cell lymphoma (ALCL) has a poor prognosis. This study sought to determine if high-dose therapy and ASCT results in long-term disease-free survival (DFS) in patients with recurrent, chemotherapy-sensitive ALK-negative ALCL. All patients with non-Hodgkin's lymphoma (NHL) who underwent ASCT at Wake Forest University and Upstate Medical University from 1 January 1990 to 12 December 2002 were reviewed to determine if they had T-, B- or null-cell NHL that was CD30+/CD15-/ALK negative. In all, 16 patients were thus identified as having ALK-negative ALCL. All 16 patients underwent ASCT at the time of first relapse and form the basis of this report. Median age of the 16 patients was 51 years. There were 11 males and five females. International prognostic index scores in 12 patients at the time of relapse were: low 3, LI 6 and HI 3. Of 15 patients, 13 relapsed after ASCT; one patient was lost to follow-up. Median progression-free survival for the 15 patients was 12 weeks (3-212+ weeks). Of 15 patients, 10 have died; nine of recurrent disease. Median overall survival for the 15 evaluable patients was 72 weeks. In our experience, high-dose therapy and ASCT does not produce long-term DFS in patients with recurrent chemotherapy-sensitive ALK-negative ALCL.

Concurrent invasive thymoma and T-cell lymphoblastic leukemia and lymphoma. A case report with necropsy findings and literature review of thymoma and associated hematologic neoplasm.

Thymoma is associated with benign and neoplastic diseases. The authors report the concurrence of invasive thymoma and T-lymphoblastic leukemia/lymphoma in a 95-year-old man. The hematologic malignancy was suspected terminally, whereas the thymoma was discovered at necropsy. The lymphoblastic leukemia/lymphoma had a clonally rearranged T-cell receptor beta-chain gene and a mature thymocyte immunophenotype. No retroviral gene sequences (human immunodeficiency virus 1 and 2, and human T-cell leukemia virus 1 and 2) were identified by polymerase chain reaction and hybridization analysis. The association of thymoma with hematologic neoplasm is reviewed.

Clinical significance of histology and immunophenotype in childhood diffuse large cell lymphoma.

To correlate the histologic subtype of diffuse large cell (DLC) lymphoma with immunophenotype, clinical features, and treatment outcome, 88 consecutively diagnosed children with this disease were studied. Of these cases, 42 (48%) were immunoblastic (IB), polymorphous subtype; 17 (19%) IB, plasmacytoid; 8 (9%) IB, clear cell; 6 (7%) IB, not otherwise specified; and 15 (17%) DLC-follicular center cell (DLC-FCC) type. Of 34 cases successfully phenotyped from paraffin sections, 13 were T cell and 9 were B cell; of the remaining cases, 8 were suggestive of T-cell lineage, 3 of B-cell lineage, and 1 of histiocytic differentiation. Although histologic subtype did not correlate with clinical features or outcome, it did correlate with immunophenotype among those cases for which lineage could be unequivocally assigned (5 of 18 IB vs. 4 of 4 DLC cases were B cell; P = 0.02) Immunophenotype was also correlated with stage of disease (11 of 13 T-cell vs. 3 of 9 B-cell cases had stage III-IV disease; P = 0.03). (Stage III includes all primary thoracic tumors; stage IV includes all with central nervous system and/or bone marrow involvement.) Significant prognostic features were clinical stage and era (thus type) of therapy (P less than 0.001). The authors conclude that most cases of large cell non-Hodgkin's lymphoma in children are of IB morphologic type, most frequently of T-cell lineage, and those with T-cell phenotype appeared to have more advanced disease.

Sexual dimorphism in the developmental regulation of brain aromatase.

Steroid sex hormones have an organizational role in gender-specific brain development. Aromatase (cytochrome P450AR), converting testosterone (T) to estradiol-17 beta (E2) is a key enzyme in brain development and the regulation of aromatase determines the availability of E2 effective for neural differentiation. Gender differences in brain development and behaviour are likely to be influenced by E2 acting during sensitive periods. This differentiating action has been demonstrated in rodent and avian species, but also probably occurs in primates including humans. In rodents, E2 is formed in various hypothalamic areas of the brain during fetal and postnatal development. The question considered here is whether hypothalamic aromatase activity is gender-specific during sensitive phases of behavioural and brain development, and when these sensitive phases occur. In vitro preoptic and limbic aromatase activity has been measured in two strains of wild mice, genetically selected for behavioural aggression based on attack latency, and in the BALB/c mouse. Short attack latency males show a different developmental pattern of aromatase activity in hypothalamus and amygdala to long attack latency males. Using primary brain cell cultures of the BALB/c mouse, sex differences in hypothalamic aromatase activity during both early embryonic and later perinatal development can be demonstrated, with higher E2 formation in males. The sex dimorphism are brain region specific, since no differences between male and female are detectable in cultured cortical cells. Immunoreactive staining with a polyclonal aromatase antibody identifies a neuronal rather than an astroglial localization of the enzyme. T increases fetal brain aromatase activity and numbers of aromatase-immunoreactive hypothalamic neuronal cell bodies. T appears to influence the growth of hypothalamic neurons containing aromatase. Differentiation of sexually dimorphic brain mechanisms may involve maturation of a gender-specific network of estrogen-forming neurons which are steroid-sensitive in early development.

Sex differences in the regulation of embryonic brain aromatase.

Oestrogen formed from androgen by aromatization plays a critical role in the sexual differentiation of the male brain and behaviour. A question which has still to be answered is what regulates the gender-specific changes in aromatase activity forming oestrogen during sensitive periods of brain growth. Using a primary cell culture technique and sexed embryos, we have shown that in the fetal mouse brain, oestrogen formation in the male is neuronal rather than glial and aromatase activity is regionally localized, being higher in the hypothalamus than in the cortex. The aromatase activity measured from cells in culture has the same enzyme binding affinity (apparent Km approximately 40 nM) as intact brain samples. Neurones developing in the embryonic male brain (embryonic day (ED) 15) contain higher aromatase activity (Vmax, 895 fmol/h/mg protein) than the female (Vmax, 604). Although a sex difference exists at early stages of embryonic development (ED 13), the embryonic aromatase system is regulated by steroids later in fetal development. The developing aromatase-containing neuroblasts probably form processes which connect to other aromatase neurones. Immunoreactive staining with an aromatase polyclonal antibody identifies an increase in numbers of aromatase-immunoreactive hypothalamic neuronal cell bodies following testosterone treatment. Testosterone treatment also causes both stimulation of neurite growth and branching as well as functional maturation of aromatase neurones. In particular, there is an increase in aromatase activity per neurone as well as a dramatic increase in the number of neurones expressing the enzyme. Both the functional and morphological changes depend on androgen receptor stimulation for several days in vitro. This conclusion is supported by colocalization studies which reveal a high number of fetal hypothalamic aromatase neurones co-expressing androgen receptor. We conclude that testosterone influences the growth of male hypothalamic neurones containing aromatase at a sensitive period of brain development. Endogenous steroid inhibitors of aromatase, probably formed within the neuroglia, also play a role in the control of oestrogen production. An endogenous 5alpha-reduced metabolite of testosterone, 5alpha-androstanedione, is almost as potent in inhibiting neuronal hypothalamic aromatase activity (Ki = 23 nM) as the synthetic non-steroidal inhibitors such as the imidazole, fadrozole, and the triazoles, arimidex and letrozole. It is clear that the oestrogen-forming capacity of the male hypothalamus has the special characteristics and plasticity of regulation which could affect brain differentiation at specific steroid-sensitive stages in ontogeny.

Regulation of sex-specific formation of oestrogen in brain development: endogenous inhibitors of aromatase.

Brain sexual differentiation occurs during the steroid-sensitive phases in early development, and is affected particularly by exposure to oestrogens formed in the brain by aromatisation of androgen. The organisational effects of oestrogen result in male-specific neuronal morphology, control of reproductive behaviour, and patterns of gonadotrophin secretion. A question which still has to be resolved is what determines changes in aromatase activity effective for the differentiation of sexually dimorphic brain development during sensitive periods of growth. In the mouse, a sex difference exists at early stages of embryonic development in aromatase-containing neurones of the hypothalamus. The embryonic aromatase system is regulated later in foetal development by androgens. Testosterone treatment increases the numbers of aromatase-immunoreactive hypothalamic neuronal cell bodies. Kinetic evidence from studies on the avian brain suggest that endogenous steroid inhibitors of aromatase, probably formed within neuroglia, also have a role in the control of oestrogen production. Inhibitory kinetic constant determination of endogenous androgenic metabolites formed in the brain showed that preoptic aromatase is potently inhibited by 5 alpha-androstanedione(K(i)=6nM) and less strongly by 5 beta-dihydrotestosterone (K(i)=350nM). Regulation by steroidal and possibly non-steroidal inhibitors may contribute to the special characteristics and plasticity in aromatase activity which develops at certain stages in ontogeny.

Steroid metabolising enzymes in the determination of brain gender.

The neurotrophic effects of oestrogen formed in the brain are important in brain sexual differentiation of the central nervous system and behaviour. Aromatase, converting testosterone to oestradiol-17beta, is a key enzyme involved in brain development. In primary cell cultures of foetal hypothalamus, we have found that male neurones consistently have higher aromatase activity than in the female. Using a specific antibody to the mouse aromatase, immunoreactivity was localized in the neural soma and neurites in hypothalamic cultures. Additionally more male foetal hypothalamus neurones express aromatase than in the female. Testosterone increases aromatase activity in parallel with a greater number of aromatase-immunoreactive neurones. Testosterone also increases soma size, neurite length, and branching of cultured hypothalamic neurones. The neuronal aromatase activity appears to be sensitive to the inductive effects of androgen only during the later stages of foetal development. Endogenous inhibitors of the aromatase are also likely to have a regulatory role. This work suggests that regulation of a network of aromatase neurones, sensitive to the hormonal environment of the hypothalamus, may determine when oestrogens are available for neurotrophic effects underlying brain differentiation.