Expression of the core promoter factors TATA box binding protein and TATA box binding protein-related factor 2 in Drosophila germ cells and their distinct functions in germline development.

In Drosophila, the expression of germline genes is initiated in primordial germ cells (PGCs) and is known to be associated with germline establishment. However, the transcriptional regulation of germline genes remains elusive. Previously, we found that the BTB/POZ-Zn-finger protein, Mamo, is necessary for the expression of the germline gene, vasa, in PGCs. Moreover, truncated Mamo lacking the BTB/POZ domain (MamoAF) is a potent vasa activator. In this study, we investigated the genetic interaction between MamoAF and specific transcriptional regulators to gain insight into the transcriptional regulation of germline development. We identified a general transcription factor, TATA box binding protein (TBP)-associated factor 3 (TAF3/BIP2), and a member of the TBP-like proteins, TBP-related factor 2 (TRF2), as new genetic modifiers of MamoAF. In contrast to TRF2, TBP was found to show no genetic interaction with MamoAF, suggesting that Trf2 has a selective function. Therefore, we focused on Trf2 expression and investigated its function in germ cells. We found that Trf2 mRNA, rather than Tbp mRNA, was preferentially expressed in PGCs during embryogenesis. Depletion of TRF2 in PGCs resulted in decreased mRNA expression of vasa. RNA interference-mediated knockdown showed that, while Trf2 is required for maintenance of germ cells, Tbp is needed for their differentiation during oogenesis. Therefore, these results suggest that Trf2 and Tbp expression is differentially regulated in germ cells and that these factors have distinct functions in Drosophila germline development.

Active vitamin D analogs, maxacalcitol and alfacalcidol, as maintenance therapy for mild secondary hyperparathyroidism in hemodialysis patients - a randomized study.

BACKGROUND: The present randomized study was designed to compare the efficacy between two active vitamin D analogs, alfacalcidol (ACD) and maxacalcitol (OCT), for the management of mild secondary hyperparathyroidism (SHPT) in dialysis patients. METHODS: SHPT in all 32 patients analyzed in the study was initially treated with OCT. Once patients' intact PTH levels decreased to the target range of 150 - 180 pg/mL, they were randomized either to switch to ACD at 0.5 µg/day (n = 14), or to remain on an effectively unchanged dose of OCT (n = 13). Phosphate, calcium, and intact PTH levels were measured every 2 weeks for 12 weeks and vitamin D doses were changed according to target ranges of phosphate (3.5 - 6.0 mg/dL), calcium (albuminadjusted calcium: 8.4 - 10.0 mg/dL), and intact parathyroid hormone (60 - 180 pg/mL). Achievement rates of the target ranges of the parameters were estimated. RESULTS: Baseline calcium levels in the OCT group were significantly higher than in the ACD group. Changes in achievement rates of target ranges of intact PTH and calcium during the study did not differ significantly between the vitamin D drugs. Changes in calcium levels in the OCT and ACD groups were similar during the study. Achievement rates of the target range of phosphate in both groups were also similar until 8 weeks, although the rate in the OCT group declined at 10 weeks. CONCLUSIONS: The efficacy and safety of OCT for the treatment of mild SHPT are similar to those of ACD in hemodialysis patients.

Binding of Drosophila maternal Mamo protein to chromatin and specific DNA sequences.

Alterations in chromatin structure dynamically occur during germline development in Drosophila and are essential for the production of functional gametes. We had previously reported that the maternal factor Mamo, which contains both a BTB/POZ domain and C2H2 zinc-finger domains and is enriched in primordial germ cells (PGCs), is required for the regulation of meiotic chromatin structure and the production of functional gametes. However, the molecular mechanisms by which Mamo regulates germline development remained unclear. To evaluate the molecular function of Mamo protein, we have investigated the binding of Mamo to chromatin and DNA sequences. Our data show that Mamo binds to chromatin and specific DNA sequences, particularly the polytene chromosomes of salivary gland cells. Overexpression of Mamo affected the organization of polytene chromosomes. Reduction in maternal Mamo levels impaired the formation of germline-specific chromatin structures in PGCs. Furthermore, we found that the zinc-finger domains of Mamo directly bind to specific DNA sequences. Our results suggest that Mamo plays a role in regulating chromatin structure in PGCs.

Effect of renal impairment on the pharmacokinetics of memantine.

The effect of renal impairment on the pharmacokinetics of a single oral dose of memantine (10 mg) was determined in Japanese subjects. Subjects were assigned to four groups based on baseline creatinine clearance (CL(CR)): normal renal function (> 80 mL/min, n = 6), and mild (50 to ≤ 80 mL/min, n = 6), moderate (30 to < 50 mL/min, n = 6), and severe renal impairment (5 to < 30 mL/min, n = 7). Mean memantine maximum plasma concentration (C(max)) was similar in the groups (12.66, 17.25, 15.75, and 15.83 ng/mL, respectively), as was mean time to C(max) (6.2, 5.2, 4.3, and 5.4 h, respectively). However, exposure to memantine determined from mean area under the plasma concentration-time curve was 1.62-, 1.97-, and 2.33-times higher in subjects with mild, moderate, and severe renal impairment, respectively, as compared to controls with normal renal function. Mean memantine plasma elimination half-life increased according to increasing renal impairment (61.15, 83.00, 100.13, and 124.31 h, respectively), while mean cumulative urinary recovery of unchanged memantine in 72 h after dosing decreased according to increasing renal impairment (33.68%, 33.47%, 23.60%, and 16.17%, respectively). These results are the same as those in the previous study on caucasian individuals, when compared per body weight. It is suggested that the dose of memantine should be halved in patients with renal impairment.

The brainstem is at high risk for recurrent noncardioembolic cerebral infarction in association with diabetes mellitus: a hospital-based study.

The goals of the study were to investigate the importance of brainstem infarction (BSI) in recurrent noncardioembolic ischemic stroke and to examine the relevant clinical background. Data were retrospectively reviewed for 655 consecutive patients with acute noncardioembolic infarction who were admitted to our hospital from January 2004 to August 2010. The patients were divided into first-stroke (n = 592) and recurrent-stroke (n = 63) groups. Acute infarcted lesions were explored on MRI, and clinical background factors including age, sex, smoking, atrial fibrillation, coronary heart disease, hypertension, hyperlipidemia and diabetes mellitus (DM) were assessed. The frequency of BSI in the recurrent-stroke group was significantly higher than that in first-stroke patients (30.2 vs. 14.9%, p = 0.0033). No other clinical background factors differed between the two groups. Only the frequency of DM differed significantly among four subgroups formed based on stroke recurrence and BSI (p < 0.0001): DM was present in 63.2% of recurrent-stroke patients with BSI, 54.5% of first-stroke patients with BSI, 27.4% of first-stroke patients without BSI, and 20.5% of recurrent-stroke patients without BSI. We conclude that the brainstem is at high risk for recurrent cerebral infarction in patients with DM.

Deep white matter hyperintensities, decreased serum low-density lipoprotein, and dilative large arteriopathy.

Deep white matter hyperintensities (DWMHs) seen on magnetic resonance imaging (MRI) are thought to reflect small-vessel diseases (SVDs) and may have a background that differs from that of stenotic large-vessel diseases. We assessed risk factors for DWMHs and investigated the association between DWMHs and dilative changes in the basilar artery (BA) on MRI in nonstroke patients. We reviewed clinical information and MRI findings for 149 outpatients aged 46-90 years, excluding those with a previous symptomatic cerebrovascular event. DWMHs were graded 0-3, and the maximal BA diameter and area were measured from the flow void on axial T2-weighted MRI to assess dilatation. We divided the patients into groups with and without DWMH grade 2 or 3, and compared clinical information and BA parameters in these groups. The two groups demonstrated significant differences in age, serum low-density lipoprotein (LDL) level, estimated glomerular filtration rate (eGFR), and BA parameters. An adjusted logistic regression analysis including BA diameter found that age (odds ratio [OR], 1.974 per 10 years; 95% confidence interval [CI], 1.030-1.112; P = .0006), LDL (OR, 0.811 per 10 mg/dL; 95% CI, 0.964-0.965; P = .0085), eGFR (OR, 0.835 per 10 mL/min/1.73 m(2); 95% CI, 0.967-0.998; P = .0229), and BA diameter (OR, 2.515 per 1 mm; 95% CI, 1.191-4.098; P = .0119) were independently associated with the presence of DWMHs. An analysis including the BA area yielded similar results. DWMHs are manifestations of SVDs and show a strong association with lower serum LDL level, lower eGFR, and BA dilatation.

Occurrence and clinicotopographical correlates of brainstem infarction in patients with diabetes mellitus.

BACKGROUND: The goal of the study was to clarify the association between diabetes mellitus (DM) and brainstem infarctions (BSIs) and to investigate the clinicotopographic characteristics of BSIs in patients with diabetes. METHODS: Data were retrospectively reviewed for 1026 consecutive patients admitted to our hospital because of acute cerebral infarctions from January 2004 to August 2010. Acute symptomatic BSIs were explored on radiologic images and classified into multiple infarctions with BSIs, multifocal BSIs, and monofocal BSIs. Isolated BSIs were further classified based on the vertical distribution into midbrain, pontine, and medullary infarctions, and on the horizontal distribution into anterior-dominant, posterior-dominant, and anterior/posterior BSIs. Neurologic symptoms of BSIs and clinical background were compared between DM and non-DM patients. RESULTS: The prevalence of BSIs was 2.6-fold higher (P < .0001) in DM patients. Logistic regression analysis including age, sex, smoking, previous stroke, atrial fibrillation, other cardiac diseases, hypertension, hyperlipidemia, and DM showed that DM was independently associated with BSIs (odds ratio [OR] 2.814; 95% confidence interval [CI] 1.936-4.090; P < .0001). Compared with non-DM patients, DM patients showed more frequent monofocal BSIs (P < .0001) and multifocal BSIs (P = .0296). Monofocal BSIs (n = 114) more frequently involved the pons (P < .0001) and medulla (P = .0212). Anterior-dominant BSIs (P < .0001) were more common in DM patients than in non-DM patients. Symptoms of BSIs included more frequent motor paresis (P = .0180) and less frequent diplopia (P = .0298) in DM patients than in non-DM patients. CONCLUSIONS: DM is important in the development of BSIs, and the associated clinical characteristics include more frequent motor paresis and less frequent diplopia.

Cerebral microbleeds and dilative remodeling of the basilar artery: a magnetic resonance imaging study.

This study was conducted to examine the relationship between cerebral microbleeds (CMBs), one of the manifestations of small-vessel diseases (SVDs), and basilar artery (BA) dilatation on magnetic resonance imaging (MRI). Clinical information and MRI images were reviewed for 149 outpatients aged 46-90 years, excluding those who had a previous symptomatic cerebrovascular event. CMBs were evaluated on T2∗-weighted MRI, and BA diameters were measured as the maximal width of the flow void on axial T2-weighted MRI to assess dilatation. Patients were divided into 2 groups, with CMBs and without CMBs, and clinical information and BA diameters were compared between the groups. Regression analyses of the data also were performed. The 2 groups had significant differences in mean blood pressure (MBP), low-density lipoprotein (LDL) and uricemic acid levels, and BA diameter. Adjusted logistic regression analysis showed that MBP (odds ratio [OR], 1.059 per 1 mm Hg; 95% confidence interval [CI], 1.019-1.101; P = .0035), LDL (OR, 0.976 per 1 mg/dL; 95% CI, 0.960-0.994; P = .0072), and BA diameter (OR, 3.266 per 1 mm; 95% CI, 1.504-7.103; P = .0028) each had an independent association with the presence of CMB. Adjusted multiple regression analysis showed that only BA diameter (ß coefficient, 0.240; 95% CI, 0.775-3.734; P = .0031) was independently associated with the number of CMBs. Our data indicate that CMB, a manifestation of SVD, shows a strong association with BA dilatation.

Novel markers of left ventricular hypertrophy in uremia.

AIMS: Left ventricular hypertrophy (LVH) is the most frequent cardiac complication in chronic renal disease. Previous studies implicate elevated serum phosphorus as a risk factor for LVH. METHODS: We treated 5/6 nephrectomized rats with enalapril or enalapril + sevelamer carbonate for 4 months to determine if sevelamer carbonate had an additional beneficial effect on the development of LVH and uremia-induced left ventricle (LV) remodeling. RESULTS: Uremia increased LV weight and cardiomyocyte size. Enalapril and enalapril + sevelamer blunted the increase in left ventricular weight. Only enalapril + sevelamer diminished the increase in cardiomyocyte size. Uremia increased cyclin D2 and PCNA and decreased p27 protein expression in the heart. Enalapril + sevelamer diminished the decrease in p27 expression caused by uremia. Uremia increased Ki67-positive and phosphohistone H(3)-positive interstitial cells. This was not seen in cardiomyocytes. Multivariable regression analysis showed that increased phosphorus was an independent risk factor for both increased LV weight and cardiomyocyte size. CONCLUSIONS: These data suggest left ventricular remodeling consists of cardiomyocyte hypertrophy and interstitial cell proliferation, but not cardiomyocyte proliferation. p27 and cyclin D2 may play important roles in the development of LVH. In addition, phosphorus can be an independent risk factor for the development of LVH.

Involvement of the basilar artery in diabetes mellitus: an MRI study of brainstem infarctions.

The relationships among diabetes mellitus (DM), brainstem infarctions (BSIs) and involvement of the basilar artery (BA) were investigated in 254 patients with acute cerebral infarctions detected on magnetic resonance (MR) imaging. Radiological findings included lesion topography and size (mm(2)) of BSIs on MR images, and the extent of BA stenosis measured by MR angiography. Adjusted logistic regression analyses showed that DM (OR 4.018; p = 0.0006) and BA stenosis (OR 1.003 per 1 mm; p < 0.0001) had an independent association with the incidence of BSIs, but the lesion size of the BSIs was only associated with BA stenosis (ß coefficient 0.280; p < 0.0001). Diabetic patients showed significantly more frequent isolated pontine infarctions and a lesser degree of BA stenosis (p < 0.005) compared to non-diabetic patients. Preferential involvement of the pons and smaller vessels may be characteristics of diabetic patients.

Ultrapure dialysate influences serum myeloperoxidase levels and lipid metabolism.

BACKGROUND: Ultrapure dialysate (UD) might contribute to improvements in the morbidity and mortality of hemodialysis (HD) patients. However, it is unclear whether increasing dialysate purity affects chronic inflammation, oxidative stress, and lipid abnormalities. METHODS: In a prospective cohort study, 126 patients undergoing maintenance HD using conventional dialysate (CD) with one endotoxin cut filter were assigned to either continuation of the same HD or HD using UD (more purified dialysate). At baseline and 6 months we measured lipids, high-sensitive (hs)CRP, oxidative LDL-cholesterol, and myeloperoxidase. RESULTS: Serum myeloperoxidase and hsCRP levels in the UD group were significantly decreased at 6 months compared with the CD group. Multivariate analysis showed that decreases in non-HDL-cholesterol and ApoB at 6 months were independently correlated with changes in myeloperoxidase. CONCLUSION: Endotoxin-free UD can improve the chronic inflammatory status, oxidative stress, and lipid abnormalities, suggesting a possible contribution to reduced cardiovascular disease risk and ultimately to lowered mortality in HD patients.

Intracranial dilative arteriopathy is associated with chronic kidney disease and small vessel diseases in the elderly.

OBJECTIVE: We sought to determine the responsible factors and clinical significance of dilatation of the internal carotid artery (ICA) and basilar artery (BA). METHODS: ICA and BA diameters were measured using magnetic resonance imaging (MRI) and their association with age, sex, conventional atherosclerotic factors, and estimated glomerular filtration rate (eGFR) were evaluated in 175 outpatients aged 40 to 89 years. The arterial diameters were measured as the maximal width of the flow void on T2-weighted MRI around the brainstem. The relationship between the arterial diameters and deep white matter hyperintensities (DWMHs) on MRI graded from 0 to 3 was also examined. Comparisons were performed between groups with diameters smaller and larger than the mean value for younger (40-69 years) and older (70-89 years) patients, and multiple regression analysis was conducted. RESULTS: Age differed significantly between the larger- and smaller-diameter groups in younger patients, but not in older patients. The larger-diameter group of older patients had a significantly lower eGFR (P = .0002 for ICA, P = .0035 for BA) and a higher DWMH grade (P = .0021 for ICA, P < .0001 for BA) compared to the smaller-diameter group. In multiple regression analysis adjusted for age and sex, a lower eGFR was an independent factor associated with larger arterial diameters (P = .0002 for ICA, P = .0021 for BA). CONCLUSION: Kidney dysfunction is an independent factor that is related to ICA and BA dilatation, which is also associated with DWMHs that reflect small vessel diseases in older patients.

Maternal RNAs encoding transcription factors for germline-specific gene expression in Drosophila embryos.

In early Drosophila embryos, germ plasm is localized to the posterior pole region and is partitioned into the germline progenitors, known as pole cells. Germ plasm contains the maternal factors required for germline development. It has been proposed that germline-specific gene expression is initiated by the function of maternal factors that are enriched in the germ plasm. However, such factors have remained elusive. Here, we describe a genome-wide survey of maternal transcripts that encode for transcription factors and are enriched in the germ plasm. We isolated pole cells from blastodermal embryos by fluorescence-activated cell sorting (FACS) and then used these isolated cells in a microarray analysis. Among the 835 genes in the Gene Ontology (GO) category transcription regulator activity listed in FlyBase, 68 were found to be predominantly expressed in pole cells as compared to whole embryos. As the early pole cells are known to be transcriptionally quiescent, the listed transcripts are predicted to be maternal in origin. Our in situ hybridization analysis revealed that 27 of the 68 transcripts were enriched in the germ plasm. Among the 27 transcripts, six were found to be required for germline-specific gene expression of vasa and/or nanos by knockdown experiments using RNA interference (RNAi). The identified transcripts encode a transcriptional activator (ovo), components of the transcriptional initiation complexes (Trf2, bip2 and Tif-IA), and the Ccr4-Not complex (CG31716 and l(2)NC136). Our study demonstrates that germ plasm contains maternal transcripts encoding transcriptional regulators for germline-specific gene expression in pole cells.

A truncated form of a transcription factor Mamo activates vasa in Drosophila embryos.

Expression of the vasa gene is associated with germline establishment. Therefore, identification of vasa activator(s) should provide insights into germline development. However, the genes sufficient for vasa activation remain unknown. Previously, we showed that the BTB/POZ-Zn-finger protein Mamo is necessary for vasa expression in Drosophila. Here, we show that the truncated Mamo lacking the BTB/POZ domain (MamoAF) is a potent vasa activator. Overexpression of MamoAF was sufficient to induce vasa expression in both primordial germ cells and brain. Indeed, Mamo mRNA encoding a truncated Mamo isoform, which is similar to MamoAF, was predominantly expressed in primordial germ cells. The results of our genetic and biochemical studies showed that MamoAF, together with CBP, epigenetically activates vasa expression. Furthermore, MamoAF and the germline transcriptional activator OvoB exhibited synergy in activating vasa transcription. We propose that a Mamo-mediated network of epigenetic and transcriptional regulators activates vasa expression.

MAMO, a maternal BTB/POZ-Zn-finger protein enriched in germline progenitors is required for the production of functional eggs in Drosophila.

A hallmark of germline cells throughout the animal kingdom is their ability to execute meiosis. However, despite its prime importance, little is known about how germline progenitors acquire this ability. In Drosophila, the primordial germ cells (PGCs) are characterized by the inheritance of germ plasm, which contains maternal factors that have sufficient ability to direct germline development. Here, we show that a novel maternal factor, MAMO, is autonomously required in PGCs to produce functional gametes. MAMO protein which contains both a BTB/POZ (Broad Complex, Tramtrack, Bric-a-brac/Pox virus and Zinc finger) domain and C(2)H(2) zinc finger motifs is enriched in PGCs during embryogenesis. The PGCs with reduced maternal MAMO activity are able to undergo oogenesis, but fail to execute meiosis properly. In the resulting oocytes, meiosis-specific chromosomal configurations are impaired. We additionally show that the decondensation of fertilized sperm nuclei is also affected in the eggs. We propose that maternal MAMO activates downstream genes to promote specialized morphological changes of both female meiotic chromosomes and the sperm nucleus, which are critical in zygote formation.

Asymptomatic large T2 high-signal pontine lesions that are different from ischemic rarefaction.

BACKGROUND: We encountered asymptomatic large T2 high-signal pontine lesion (PL) with peachlike configurations on magnetic resonance imaging. The objective of this study was to determine the pathogenic factors for such PLs and to differentiate PL from ischemic rarefaction. METHODS: We compared the clinical backgrounds of 37 patients with PL and 100 patients without PL, including conventional atherosclerotic risk factors, chronic kidney disease (CKD) defined as an estimated glomerular filtration rate less than 60 mL/min/1.73 m(3), and the degree of supratentorial deep white matter hyperintensities (DWMH). The chronologic alternations in PL were also reviewed. RESULTS: Patients with PL had a much higher incidence of diabetes mellitus (DM) and CKD as compared with patients without PL: 54% versus 15% for DM (P < .001) and 57% versus 29% for CKD (P < .05). PL did not necessarily accompany DWMH and its degree did not correlate with PL. Further analyses showed that patients with PL without DWMH had a much higher incidence of DM and CKD as compared with patients with DWMH without PL: 85% versus 6% for DM (P < .0001) and 85% versus 50% for CKD (P < .05). Chronologically, PLs expanded, shrunk, or fluctuated in size in 5 patients with DM and CKD. CONCLUSIONS: We demonstrated that DM and CKD are important for PL and that the backgrounds of PL were inconsistent with DWMH. These results and the chronologic alternations in PL suggested that metabolic factors other than ischemic rarefaction played important roles in the development of PL.

Thrombotic thrombocytopenic purpura-like thrombocytopenia closely related to uremia: report of two cases.

Idiopathic thrombotic thrombocytopenic purpura (TTP) was suspected in two cases with end-stage renal disease (ESRD) before initiating dialysis based on both the clinical course and the initial laboratory findings. After three plasma exchanges and four sessions of hemodialysis, the thrombocytopenia improved to the normal range. However, the level of haptoglobin remained low after the recovery of platelet count. The low activity of ADAMTS13 (a disintegrin-like and metalloprotease with thrombospodin type 1 repeats) was not observed and the inhibitor was negative in these two cases. Therefore, thrombocytopenia was not attributable to TTP and other factors might therefore be involved in its pathogenesis. Some investigators have reported that a uremic state may possibly be involved in thrombocytopenia in both animals and humans. In the present cases, no apparent cause other than uremia was found because, after initiating hemodialysis, the thrombocytopenia rapidly improved. In conclusion, TTP-like thrombocytopenia can be caused by uremia and the influence of uremia should be considered in the pathogenesis of ESRD patients with TTP-like manifestations.

Expression of meiotic genes in the germline progenitors of Drosophila embryos.

Meiosis is one of the fundamental characteristics of germ cells. In Drosophila, genetic screens have identified many genes required for meiotic division. However, it remains elusive as to when and how these meiotic genes are activated during germline development. To obtain insights into their regulatory mechanisms, we examined the expression of 38 meiotic genes in the germline progenitors, pole cells, during embryogenesis. We found that the transcripts of 12 meiotic genes were enriched in pole cells within the embryonic gonads. Among them, bag of marbles (bam), benign gonial cell neoplasia (bgcn), deadhead (dhd), matotopetli (topi) and twine (twe) were activated only in pole cells within the gonads, whereas the transcripts from grapes (grp), Kinesin-like protein at 3A (Klp3A), pavarotti (pav), lesswright (lwr), mei-P26, Topoisomerase 2 (Top2) and out at first (oaf) were distributed ubiquitously in early embryos and then became restricted to pole cells and to a subset of somatic tissues at later embryonic stages. The remaining meiotic genes were either expressed ubiquitously in the embryos (15 genes) or were undetectable in pole cells within the gonads (11 genes). These observations suggest that pole cells have already acquired the potential to express several meiotic genes. Our data will thus provide a useful basis for analyzing how the germline acquires a potential to execute meiosis.

H3K36 Trimethylation-Mediated Epigenetic Regulation is Activated by Bam and Promotes Germ Cell Differentiation During Early Oogenesis in Drosophila.

Epigenetic silencing is critical for maintaining germline stem cells in Drosophila ovaries. However, it remains unclear how the differentiation factor, Bag-of-marbles (Bam), counteracts transcriptional silencing. We found that the trimethylation of lysine 36 on histone H3 (H3K36me3), a modification that is associated with gene activation, is enhanced in Bam-expressing cells. H3K36me3 levels were reduced in flies deficient in Bam. Inactivation of the Set2 methyltransferase, which confers the H3K36me3 modification, in germline cells markedly reduced H3K36me3 and impaired differentiation. Genetic analyses revealed that Set2 acts downstream of Bam. Furthermore, orb expression, which is required for germ cell differentiation, was activated by Set2, probably through direct H3K36me3 modification of the orb locus. Our data indicate that H3K36me3-mediated epigenetic regulation is activated by bam, and that this modification facilitates germ cell differentiation, probably through transcriptional activation. This work provides a novel link between Bam and epigenetic transcriptional control.

Characterization of the isoforms of MOVO zinc finger protein, a mouse homologue of Drosophila Ovo, as transcription factors.

We previously described two isoforms (MOVO-A and -B) of a novel zinc finger protein MOVO, a mouse homologue of Drosophila Ovo protein. Here, we isolated cDNA encoding the third isoform MOVO-C, which had a transactivation domain and zinc finger domain, but lacked an N-terminal potential repression domain that was present in MOVO-A. Three isoform mRNAs were expressed highly in mouse testis and also in the ovary at lower levels. The structural analyses of the isolated Movo gene and mRNAs demonstrated that three different Movo transcripts were differentially processed to generate three isoforms. Major mRNA species encoded MOVO-B with a zinc finger domain alone, and minor mRNA species encoded MOVO-A (potential repressor) and MOVO-C (potential activator). To assign MOVO to a transcriptional factor, we characterized DNA-binding and transactivation properties. Random oligonucleotide selection, electrophoretic mobility shift assay and footprinting indicated that MOVO bound to the sequence, 5'-G(G/C/T)GGGGG-3'. These motifs were found in the 5'-flanking regions of Movo and other testis-specific genes. Nuclear proteins binding to this motif were detected in mouse testis, and the expression of MOVO mRNA was restricted in spermatocytes. The luciferase assay demonstrated that MOVO-C activated Movo promoter and MOVO-A repressed it, but MOVO-B had no effects. Mutated MOVO-binding motifs in the Movo promoter reduced the luciferase activity. All the isoforms had no effects on SV40 promoter without MOVO-binding motifs. MOVO-A partially rescued oogenesis of a Drosophila ovo mutant. These results suggest that MOVO isoforms are transcription factors to regulate genes carrying the MOVO-binding motifs in the testis.