Identification of human autoantibodies to the transcriptional repressor ZF5.

ZF5 was originally cloned as a transcriptional repressor on the mouse c-myc promoter. It contains the Kruppel-type zinc fingers and a conserved POZ domain, which is found in a growing number of zinc finger proteins and mediate protein-protein interactions. Autoantibodies against transcription factors are sometimes found in sera from patients with high levels of anti-nuclear antibody (ANA). Using Western blotting with ZF5 and sera of autoimmune disease, we detected one serum, named M6 serum, which contains the antibody against a transcriptional repressor ZF5. The confirmed epitope was specific to ZF5 and was not reactive to the other analogous factors: BCL-6, ZID, and Sp1. This epitope also has a molecular mimicry of the viral proteins. From these results, we predict that viral protein which mimics host ZF5 antigen triggers self-reactive T cell clones and induces the autoantibody in M6 serum after the destruction of host tissues.

Analysis of the consensus binding sequence and the DNA-binding domain of ZF5.

Murine ZF5 is a transcription factor with five zinc finger motifs that represses the c-myc gene by binding to two GC-rich elements at the promoter region. Because of its ubiquitous expression in a variety of tissues, elucidation of biological functions and cellular target genes of ZF5 is of great interest. As the first step of identifying cellular target genes, we have attempted to determine the consensus binding motif for ZF5. We succeeded in isolating 19 oligonucleotide duplex DNAs to which ZF5 binds and determined the binding sequences with DNase I footprinting analysis. From these sequences, we deduced the consensus binding motif for ZF5 to be GSGCGCGR. In addition, we have analyzed the DNA-binding domain of ZF5 by testing a series of deletion mutants. It turned out that the zinc fingers 3 and 4 of the five finger motifs play a critical role in DNA binding.

ZF5, which is a Kruppel-type transcriptional repressor, requires the zinc finger domain for self-association.

ZF5, which we have cloned as a transcriptional repressor on the mouse c-myc promoter, has the POZ domain at the amino-terminus and the Kruppel-type zinc finger domain at the carboxy-terminus. In this report, we showed that ZF5 has two contradictory functions in transcription: activation of human immunodeficiency virus (HIV) promoter and repression of the HSV thymidine kinase (TK) promoter. The POZ domain contributed to the repressor activity, whereas the active function resulted from the DNA-binding ability of the zinc finger domain. We demonstrated that the POZ domain has a function mediating homomeric protein-protein interaction and this interaction requires the zinc finger domain. Furthermore, the POZ domain decreased the DNA-binding activity of the zinc finger domain. These results can provide evidence indicating the important interaction between the POZ and zinc finger domains.

Genomic cloning and characterization of the mouse POZ/zinc-finger protein ZF5.

We isolated genomic DNA containing the entire sequence of ZF5, which was originally identified by its ability to repress the mouse c-myc promoter and which was characterized as one of the POZ (Poxvirus and zinc finger) proteins. The POZ motif is a protein-protein interaction interface found at the N-terminal region of zinc finger proteins. Sequence analysis demonstrated that the ATG translation initiation codon was separately located from the remainder of the coding sequence. Using both RNase protection and primer extension assay, a single major transcription start site was determined. Promoter analysis by transient transfection assay suggested positive autoregulation by ZF5 itself. The ZF5 N-terminal region, including the POZ domain, was required for this regulation. Sp1 also activated the ZF5 promoter and this activity was repressed by addition of ZF5. ZF5 expression was stronger in mouse ovary, lung and brain than in other organs.

Detection of mouse skeletal muscle-specific product, which includes ZF5 zinc fingers and a VP16 acidic domain, by reverse transcriptase PCR.

ZF5, which we have cloned as a repressor on the mouse c-myc promoter, is a zinc finger protein containing Kruppel-type zinc finger and ZiN/POZ domains. In a reverse transcriptase PCR assay using mouse skeletal muscle RNA, we identified a 827 bp PCR product including the zinc finger domain of ZF5 and the acidic domain of VP16. The presence of the VP16 acidic domain induced the reduction of DNA-binding activity of the zinc finger domain. In addition, the inhibitory effect of the VP16 acidic domain was demonstrated on the human immunodeficiency virus (HIV) promoter, but there was no effect on the thymidine kinase (TK) promoter.

Induction of a germline mutation at a hypervariable mouse minisatellite locus by 252Cf radiation.

Male C3H/HeN mice were exposed to 252Cf radiation and mated with unirradiated C57BL/6N females. F1 mice were analyzed for germline mutation at the paternally derived C3H/HeN allele of a hypervariable minisatellite locus, Ms6hm. This locus exhibited a high frequency of length change mutation spontaneously, and the mutation frequency of the paternally derived C3H/He allele in F1 mice born to unirradiated males was 8.4%. Exposure of male mice to 252Cf radiation resulted in even higher frequency of germline mutation. The spermatid stage germ cells were most sensitive to neutrons, and the mutation frequencies of the paternal allele were elevated to 18%, 26% and 24% for 0.35, 0.7 and 1.02 Gy of 252Cf radiation, respectively. Spermatozoa and spermatogonia stages were less sensitive and the mutation frequencies for 1.02 Gy of 252Cf radiation were 16% and 19%, respectively. The 252Cf radiation consisted of 35% gamma-rays and 65% neutrons. Assuming that these two radiations act additively, RBE of 252Cf neutrons for the induction of minisatellite mutation was calculated to be 5.9 for spermatozoa stage irradiation, 2.6 for spermatid stage irradiation and 6.5 for spermatogonia irradiation.

Genetic status of p53 and induction of apoptosis by radiation or isoflavones in human gastric carcinoma cell lines.

We have shown previously that various human cancer cell lines undergo morphological changes and internucleosomal DNA fragmentation characteristic of apoptosis after exposure to ionizing radiation or isoflavones. Here, we assessed the role of p53 gene in cell cycle and apoptosis following treatment of 11 gastric carcinoma cell lines with gamma-rays, genistein, biochanin A, or daidzein. Cell survival was measured by trypan blue staining, and apoptosis was assessed by fluorochrome staining. The rate of cell survival and apoptosis of the cells by gamma-irradiation or isoflavones did not correlate with p53 gene abnormalities. Flow cytometric measurement of DNA content demonstrated that while gamma-irradiation and genistein induced G(2) arrest, biochanin A and daidzein blocked the cell cycle of all carcinoma cells at G(1) phase. At multiple time points following irradiation, G(2) arrest was observed at 12-16 h in the wild-type and mutant p53 cell lines. Induction of p53 and p21 proteins was not observed in wild-type p53 lines after exposure to gamma-irradiation or isoflavones by Western blotting. Moreover, transfection of the wild-type p53 gene into MKN-1 cells failed to induce G(1) arrest by gamma-irradiation and genistein. Based on these results, we hypothesize that gastric cancer cells may possess a signal pathway which is different from the usual mechanisms of the p53-mediated DNA damage response in normal or hematopoietic tumor cells.

Association of minisatellite instability with c-myc amplification and K-ras mutation in methylcholanthrene-induced mouse sarcomas.

Instability of microsatellite sequences are frequently found in human tumors. In addition, minisatellite sequences, another group of highly unstable sequences, serve as sensitive markers of genetic instability. We have studied minisatellite instability in methylcholanthrene-induced mouse sarcomas. These sarcomas frequently carry the amplified c-myc gene. Seven sarcomas without the amplification and seven others with the amplification were selected randomly. Regardless of the state of the c-myc gene amplification, these sarcomas exhibited a varying degree of transplantability in syngeneic mice. The hypervariable mouse minisatellite locus Ms6hm was found to be highly unstable, specifically among sarcomas with the amplified c-myc gene. However, chromosome instability, as analyzed by micronucleus assay, was observed similarly for two groups of sarcomas. In addition, transversion of G to C and A to T was detected at the K-ras gene in four of the seven sarcomas with the amplified c-myc gene, and these mutations are thought to be induced directly by methylcholanthrene. Thus, concomitant occurrence was observed for three seemingly unrelated mutations, amplification of the c-myc locus, point mutation of the K-ras gene, and instability at the hypervariable mouse minisatellite locus. The present study indicates a possible involvement of K-ras mutation and c-myc amplification in induction of genetic instability in methylcholanthrene-induced mouse sarcomas.

Radiation-induced apoptotic cell death in human gastric epithelial tumour cells; correlation between mitotic death and apoptosis.

The mode of cell death in cells undergoing mitotic death after gamma-irradiation was studied in seven human gastric epithelial tumour cell lines and two strains of normal gastric fibroblasts. Apoptotic cells were frequently observed in all tumour lines after irradiation, whereas the two fibroblast strains were quite low in apoptosis frequency. The advent of apoptosis depended on the radiation doses and incubation time. Detailed analysis of one of the carcinoma lines, SH101-P4, revealed that G2-phase arrest was maximum at 12 h postirradiation. The cells began to escape G2 arrest by 24 h. Apoptotic cells began to increase at 12 h postirradiation and became maximal from 72 to 96 h. Apoptosis developed in the G1 phase of the cell cycle subsequent to the irradiation. These results suggest that apoptosis is one of the modes of mitotic death after irradiation.

A radiation-induced murine ovarian granulosa cell tumor line: introduction of v-ras gene potentiates a high metastatic ability.

A non-metastatic epithelial tumor cell line, OV3121, was established from ovarian granulosa cell tumor in B6C3F1 mouse irradiated with 60Co-gamma rays. OV3121 cells showed an epithelial morphology and grew in monolayer with a population doubling time of 28-30 h. The production of estradiol and the expression of cytokeratin confirmed the epithelial origin of the line. No pulmonary metastasis was observed from solid tumors after subcutaneous (s.c.) injection or after intravenous (i.v.) injection of a clonal subline, OV3121-1 cells. We examined the experimental metastasis of individual clones of OV3121-1 cells, containing various introduced viral oncogenes: v-Ha-ras, v-Ki-ras, v-fms, v-mos, v-raf, v-src, v-sis, v-fos and v-myc. Among them, only OV3121-1 cells with v-Ha-MuSV or v-Ki-MuSV produced lung colonies at high frequencies. In a more detailed analysis, the v-Ha-ras transfectants OV-ras4 and OV-ras7 were found to form colonies in various organs by metastasis from tumors after s.c. injection, as well as lung colonies after i.v. injection. Moderately metastatic OV-ras7 cells showed high gelatinolytic activity at 72 kDa (MMP-2) and 92 kDa (MMP-9) as compared with the parental OV3121-1 and OV-Neo control cells by zymographic analysis. However, more metastatic OV-ras4 cells produced progressively weaker bands of 72 kDa gelatinolytic activity. No gross alterations in the expression of MMP-1, MMP-3, TIMP-1 and TIMP-2 transcripts were detected in these cell lines. These results suggest that this ovarian granulosa cell tumor line may provide a useful system for understanding the mechanisms by which oncogenes influence the occurrence of metastasis.

CSF shunt physics: factors influencing inshunt CSF flow.

Cerebrospinal fluid (CSF) in a shunt does not have a constant flow rate. The flow fluctuates from 0.01 ml/min to 1.93 ml/min according to each patient's own daily supine rhythmic pattern. We determined and evaluated the factors influencing CSF flow in a shunt in 19 cases of hydrocephalus. Postural changes, such as head elevation, led to increases by over 0.04 ml/min in inshunt CSF flow, while inshunt CSF flow in the supine position was less than 0.04 ml/min. Respiratory changes, such as coughing and apnea-hyperventilation, also influenced inshunt CSF flow. Changes in intracranial pressure (ICP) corresponded to changes in inshunt CSF flow. Inshunt CSF flows were higher than average during the night, the flows being stimulated by increases in ICP especially during REM sleep.

Over-expressed ZF5 gene product, a c-myc-binding protein related to GL1-Kruppel protein, has a growth-suppressive activity in mouse cell lines.

ZF5 encodes a zinc finger protein, which contains five C2H2-type zinc fingers showing homology with the zinc finger of the Kruppel family, and binds to two sites in the mouse c-myc promoter. We report the effect of over-expression of ZF5 on cell growth. ZnCl2 treatment suppressed the growth of a mouse fibroblast cell line (L cells) transfected with the wild-type ZF5 gene driven by the metallothionein promoter. Cells transfected with the wild-type ZF5 gene formed colonies two- to fivefold less efficiently than those transfected with the mutant ZF5 gene in P19, NIH3T3, 3T3-L1 and L cells. Over-expression of ZF5 did not cause c-myc down-regulation or arrest of the cell cycle, but increased the DNA content.

Cine magnetic resonance imaging of aqueductal stenosis.

Cerebral aqueductal stenosis is one of the most common causes of congenital and acquired hydrocephalus, but the etiology, pathophysiology and cerebrospinal fluid (CSF) dynamics of aqueductal stenosis have yet to be clarified. Utilizing cardiac gated cine magnetic resonance (MR) imaging, we evaluated aqueductal configuration and pulsatile motion of brain and CSF flow stimulated by cardiac pulsation in five patients with non-tumoral aqueductal stenosis. Cine MR of four cases revealed obliteration of the aqueduct by thickening mesencephalic tectum, turbulent CSF flow in the III ventricle, and absence of flow-related signal void, which in all normal cases indicates CSF movement within the aqueduct. In the remaining fifth case, with proximal dilation of the aqueduct resulting from thinning of the tectum, distortion of caudal (distal) tectum related to pulsatile motion of the brain caused funnel-like narrowing of the aqueduct, leading to incomplete obstruction and the absence of upward CSF flow during diastole.

Antiproliferative effects of isoflavones on human cancer cell lines established from the gastrointestinal tract.

Seven isoflavones, biochanin A, daidzein, genistein, genistin, prunectin, puerarin, and pseudobaptigenin were tested for cytostatic and cytotoxic effects on 10 newly established cancer cell lines of the human gastrointestinal origin. Proliferation of HSC-41E6, HSC-45M2, and SH101-P4 stomach cancer cell lines was strongly inhibited by biochanin A and genistein, whereas other stomach, esophageal, and colon cancer lines were moderately suppressed by both compounds. Biochanin A and genistein were cytostatic at low concentrations (< 20 micrograms/ml for biochanin A, < 10 micrograms/ml for genistein) and were cytotoxic at higher concentrations (> 40 micrograms/ml for biochanin A, > 20 micrograms/ml for genistein). DNA fragmentation was observed at cytotoxic doses of both compounds, indicating the apoptotic mode of cell death by the compounds. Chromatin condensation and nuclear fragmentation of each cell line were also observed. The advent of apoptosis was dose dependent for both isoflavones. Biochanin A suppressed tumor growth of HSC-45M2 and HSC-41E6 lines in athymic nude mice. Our results suggest that two of isoflavone derivatives, biochanin A and genistein, inhibit the cell growth of stomach cancer cell lines in vitro through activation of a signal transduction pathway for apoptosis. Moreover, in vivo experiments demonstrate that biochanin A can be used as an anticancer agent.

Transcriptional repressor ZF5 identifies a new conserved domain in zinc finger proteins.

We have cloned a cDNA encoding a new murine C2H2 zinc finger protein, ZF5. The 51.3 kD protein contains five GL1-Kruppel type zinc fingers at the C-terminus. At its N-terminus, ZF5 has a 41 amino acid region which was found to be homologous to the N-termini of several other zinc finger proteins. This region defines a new motif within zinc finger proteins which we have named the Zinc finger N-terminal (ZiN) domain. ZF5 binds to two sites in the c-myc promoter and to the -50 bp site of the herpes simplex thymidine kinase promoter. ZF5 is a transcriptional repressor and its repression domain is located N-terminal to the zinc finger domains. A single 4 kb ZF5 mRNA is expressed widely.

The same external signal differentially induced the c-myc expression in Burkitt lymphoma and B-lymphoblastoid cell lines.

An extracellular signal, such as phorbol-12-myristate-13-acetate (PMA), was found to reduce the c-myc expression in Burkitt lymphoma (BL) cells but augment the expression of the same gene in a B-lymphoblastoid cell line (B-LCL). Studies with Epstein-Barr virus (EBV)-converted BL cells demonstrated that the differential effect of PMA on c-myc expression was not due to alterations in the structure of the translocated c-myc gene, but to the difference in the intracellular milieu of the B cells. Experiments on the degradation rate in c-myc RNA suggested that this phenomenon in c-myc expression was exerted, at least in part, at the transcriptional level.

Necessity for ICP monitoring to supplement GCS in head trauma cases.

The necessity for ICP monitoring together with GCS findings to detect deterioration in head trauma cases and determine the treatment required was studied. There were 18 subjects (14 males, 4 females) aged from 11 to 61. Cases of primary brain stem damage were excluded. Eight cases had GCS of 6-10, and 10 cases scores of 11-15. Initial CTs of these cases indicated the following conditions: thin acute extradural haematoma (A-EDH), thin acute subdural haematoma (A-SDH), brain contusion, and single or multiple intracerebral haematoma (ICH). However, in no case was any mass effect clearly shown. Medical decompression (osmotherapy, barbiturate, steroid and mechanical hyperventilation) was carried out with simultaneous ICP monitoring. Based upon our cases which showed a GCS score of 10 or less, ICP monitoring should accompany medical decompression. Where the ICP cannot be maintained below 20 mm Hg, there is a high risk (about 60%) of haematoma enlargement, delayed haematoma, or increasing brain oedema. ICP monitoring in these cases should be maintained for at least one week. Timely surgical decompression is necessary when the ICP stays above 20 mm Hg, the GCS score drops below 10, and repeat CT scan indicates progress of the mass effect.

Factors affecting cerebrospinal fluid flow in a shunt.

Nineteen hydrocephalic patients were studied to determine factors affecting cerebrospinal fluid (CSF) flow through shunts. This study was based on our previously reported method by which fluctuations in CSF flow through a shunt of from 0.01 ml min-1 to 1.93 ml min-1 were identified, each having its own rhythmic pattern. While CSF flow in a supine position was less than 0.01 ml min-1, head elevation to 60 degrees led to increases in CSF flow from 0.12 ml min-1 to 0.17 ml min-1. Sudden respiratory changes such as coughing also affected CSF flow. CSF flows were higher than average between 10 pm and 7 am, and changes in CSF flow were related to slight increases in ICP during REM sleep. There is no relationship between CSF flow in a shunt and daily fluid intake which varied from 27 ml kg-1 to 103 ml kg-1, and no significant changes in CSF flow resulting from rapid intravenous injection of Glycerol and Ringer's solution.

[Cerebrospinal fluid absorption mechanism--based on measurement of CSF flow rate in shunt tube].

The cerebrospinal fluid (CSF) absorption mechanism in cases of hydrocephalus was investigated on the basis of measurements of CSF flow in a shunt tube after ventriculo-peritoneal shunt surgery, monitoring of intracranial pressure, CT findings, radioisotope cisternography, cerebral blood flow, EEG, PSP tests and changes in neurological findings. The subjects were 6 males and 7 females aged from 18 to 70. CSF flow rates in the shunt tubes were between 0.01 and 1.93 ml/min. Calculating the daily volume of CSF flow, the subjects were divided into two groups: Group A (8 patients) with a volume of less than 150 ml/day (0.01-0.25 ml/min), and Group B (5 patients) with between 150 and 500 ml/day (0.01-1.93 ml/min). Monitoring of intracranial pressure prior to the shunt operation was performed in 10 cases. These pressure values ranged between 4 and 25 mmHg (mean: 7-8 mmHg), and there was no difference between the two groups. The pre-and post-operative radioisotope cisternography findings indicated improvement of ventricular dilatation, periventricular lucency and ventricular reflux. After the shunt operations, there was neurological improvement in 6 of the 8 Group A cases but only in 2 of the 5 Group B cases. Considering the CSF flow volumes of the two groups, it appears that in Group A the shunt tube is not the main CSF circulation pathway. This could mean that resistance to CSF absorption in the cerebrospinal space has decreased after the shunt operation and there has been recovery of the physiological CSF absorption pathways. In other words, neurological improvement can be expected in this group A.

A non-invasive CSF flowmeter.

A new non-invasive method for quantitative measurement of cerebrospinal fluid (CSF) flow in the ventriculo-peritoneal shunt tubing, used in hydrocephalus, has been developed. It is an implantable device which produces a bubble in the shunt tubing by electrolysis. This bubble is then detected in the tubing by an electrode arrangement using electric impedance or ultrasonically using a Doppler probe. The energy for electrolysis is supplied by extracorporeal high-frequency transmission. The CSF flow rate is calculated by the velocity of bubble flow in the tubing. CSF flow rates, ranging from 0.01 to 1.00 ml/min, have been measured in animal experiments with statistically good accuracy. In 11 clinical cases a flow range of between 0.01 and 1.93 ml/min have been observed.