Correction: Fhit modulation of the Akt-survivin pathway in lung cancer cells: Fhit-tyrosine 114 (Y114) is essential.

After publication of this Article the authors noticed errors in several figures. In Fig. 2b the Gapdh panels are incorrect. The lysates are identical to those used in Fig. 1b, therefore the Gapdh panels should be the same in both figures. In Fig. 3b the Gapdh panels for Ad-Fhit-wt and Ad-Fhit-Y114F are incorrect and have been replaced with scans from original films. In Fig. 4A the Gapdh panels are incorrect. The lysates are identical to those used in Fig. 3b, therefore the Gapdh panels should be the same in both figures. In Fig. 4Bb the Gapdh panels for Fhit siRNA were incorrect and have been replaced with scans from original films. All resupplied figures are provided below. In Fig. 5C several panels are incorrect. The Authors were unable to locate the original films for all of these panels so Fig. 5c has been deleted. The scientific conclusions of this paper have not been affected.

Fhit modulation of the Akt-survivin pathway in lung cancer cells: Fhit-tyrosine 114 (Y114) is essential.

The Fhit tumor suppressor binds and hydrolyses diadenosine polyphosphates and the Fhit-substrate complex has been proposed as a proapoptotic effector, as determined by infection of susceptible cancer cells with adenoviruses carrying wild-type fragile histidine triad (FHIT) or catalytic site mutants. The highly conserved Fhit tyrosine 114 (Y114), within the unstructured loop C-terminal of the catalytic site, can be phosphorylated by Src family tyrosine kinases, although endogenous phospho-Fhit is rarely detected. To explore the importance of Y114 and identify Fhit-mediated signaling events, wild-type and Y114 mutant FHIT-expressing adenoviruses were introduced into two human lung cancer cell lines. Caspase-dependent apoptosis was effectively induced only by wild-type but not Y114 mutant Fhit proteins. By expression profiling of FHIT versus mutant FHIT-infected cells, we found that survivin, an Inhibitor of Apoptosis Protein (IAP) family member, was significantly decreased by wild-type Fhit. In addition, Fhit inhibited activity of Akt, a key effector in the phosphatidylinositol 3-OH kinase (PI3K) pathway; loss of endogenous Fhit expression caused increased Akt activity in vitro and in vivo, and overexpression of constitutively active Akt inhibited Fhit-induced apoptosis. The results indicate that the Fhit Y114 residue plays a critical role in Fhit-induced apoptosis, occurring through inactivation of the PI3K-Akt-survivin signal pathway.

Stereochemical retention of the configuration in the action of Fhit on phosphorus-chiral substrates.

Fhit is the protein product of FHIT, a candidate human tumor suppressor gene. Fhit catalyzes the hydrolysis of diadenosine triphosphate (Ap3A) to AMP and ADP. Fhit is here shown to catalyze the hydrolysis in H218O with production of adenosine 5'-[18O]phosphate and ADP, proving that the substitution of water is at Palpha and not at Pbeta. The chain fold of Fhit is similar to that of galactose-1-phosphate uridylyltransferase, which functions by a double-displacement mechanism through the formation of a covalent nucleotidyl-enzyme intermediate and overall retention of configuration at Palpha. The active site of Fhit contains a histidine motif that is reminiscent of the HPH motif in galactose-1-phosphate uridylyltransferases, in which the first histidine residue serves as the nucleophilic catalyst to which the nucleotidyl group is bonded covalently in the covalent intermediate. In this work, the Fhit-catalyzed cleavage of (RP)- and (SP)-gamma-(m-nitrobenzyl) adenosine 5'-O-1-thiotriphosphate (mNBATPalphaS) in H218O to adenosine 5'-[18O]thiophosphate is shown to proceed with overall retention of configuration at phosphorus. gamma-(m-Nitrobenzyl) adenosine 5'-O-triphosphate (mNBATP) is approximately as good a substrate for Fhit as Ap3A, and both (RP)- and (SP)-mNBATPalphaS are substrates that react at about 0.5% of the rate of Ap3A. The stereochemical evidence indicates that hydrolysis by Fhit proceeds by a double-displacement mechanism, presumably through a covalent AMP-enzyme intermediate.

Genetic, biochemical, and crystallographic characterization of Fhit-substrate complexes as the active signaling form of Fhit.

Alterations in the FHIT gene at 3p14.2 occur as early and frequent events in the development of several common human cancers. The ability of human Fhit-negative cells to form tumors in nude mice is suppressed by stable reexpression of Fhit protein. Fhit protein is a diadenosine P1,P3-triphosphate (ApppA) hydrolase whose fungal and animal homologs form a branch of the histidine triad (HIT) superfamily of nucleotide-binding proteins. Because the His-96 --> Asn substitution of Fhit, which retards ApppA hydrolase activity by seven orders of magnitude, did not block tumor-suppressor activity in vivo, we determined whether this mutation affected ApppA binding or particular steps in the ApppA catalytic cycle. Evidence is presented that His-96 --> Asn protein binds ApppA well and forms an enzyme-AMP intermediate extremely poorly, suggesting that Fhit-substrate complexes are the likely signaling form of the enzyme. The cocrystal structure of Fhit bound to Ado-p-CH2-p-ps-Ado (IB2), a nonhydrolyzable ApppA analog, was refined to 3.1 A, and the structure of His-96 --> Asn Fhit with IB2 was refined to 2.6 A, revealing that two ApppA molecules bind per Fhit dimer; identifying two additional adenosine-binding sites on the dimer surface; and illustrating that His-98 is positioned to donate a hydrogen bond to the scissile bridging oxygen of ApppA substrates. The form of Fhit bound to two ApppA substrates would present to the cell a dramatically phosphorylated surface, prominently displaying six phosphate groups and two adenosine moieties in place of a deep cavity lined with histidines, arginines, and glutamines.

The role of the FHIT/FRA3B locus in cancer.

Common fragile sites form gaps at characteristic chromosome bands in metaphases from normal cells after aphidicolin induction. The distribution of common fragile sites parallels the positions of neoplasia-associated chromosomal rearrangements, prompting the proposal that fragility disposes to chromosomal rearrangements. Implicit in this hypothesis is that genes at fragile sites are altered by chromosome rearrangement and thus contribute to neoplastic growth. Chromosome band 3p14.2, encompassing the most inducible common fragile region, FRA3B, has been cloned and the FHIT gene, straddling FRA3B, characterized. The gene is inactivated by deletions in cancer-derived cell lines and primary tumors and Fhit protein is absent or reduced in lung, stomach, kidney, and cervical carcinomas, consistent with function as a tumor suppressor. FRA3B thus fulfills the prophecy that fragile site alterations contribute to the neoplastic process through inactivation of a tumor suppressor gene.

Purification and crystallization of complexes modeling the active state of the fragile histidine triad protein.

Fragile histidine triad protein (Fhit) is a diadenosine triphosphate (ApppA) hydrolase encoded at the human chromosome 3 fragile site which is frequently disrupted in tumors. Reintroduction of FHIT coding sequences to cancer cell lines with FHIT deletions suppressed the ability of these cell lines to form tumors in nude mice even when the reintroduced FHIT gene had been mutated to allow ApppA binding but not hydrolysis. Because this suggested that the tumor suppressor activity of Fhit protein depends on substrate-dependent signaling rather than ApppA catabolism, we prepared two crystalline forms of Fhit protein that are expected to model its biologically active, substrate-bound state. Wild-type and the His96Asn forms of Fhit were overexpressed in Escherichia coli, purified to homogeneity and crystallized in the presence and absence of ApppA and an ApppA analog. Single crystals obtained by vapor diffusion against ammonium sulfate diffracted X-rays to beyond 2.75 A resolution. High quality native synchrotron X-ray data were collected for an orthorhombic and a hexagonal crystal form.

Fhit, a putative tumor suppressor in humans, is a dinucleoside 5',5"'-P1,P3-triphosphate hydrolase.

Human Fhit (fragile histidine triad) protein, encoded by the FHIT putative tumor suppressor gene, is a typical dinucleoside 5',5"'-P1,P3-triphosphate (Ap3A) hydrolase (EC 3.6.1.29) on the basis of its enzymatic properties we report here. Ap3A is the preferred substrate among ApnA (n = 3-6), and AMP is always one of the reaction products. Mn2+ and Mg2+ are equally stimulatory, while Zn2+ is inhibitory with Ap3A as the substrate. Values of the K(m) for Ap3A and Ap4A are 1.3 and 4.6 microM, respectively. Values of the specificity constant, kcat/K(m), for Ap3A and Ap4A are 2.0 x 10(6) and 6.7 x 10(3) s-1 M-1, respectively, for a glutathione S-transferase (GST)-Fhit fusion protein. Site-directed mutagenesis of FHIT demonstrated that all four conserved histidines are required for full activity, and the central histidine of the triad is absolutely essential for Ap3A hydrolase activity. This putative tumor suppressor is the first evidence for a connection between dinucleotide oligophosphate metabolism and tumorigenesis. Also, Fhit is the first HIT protein in which the histidine residues have been demonstrated by mutagenesis to be critical for function.

Cloning of the Schizosaccharomyces pombe gene encoding diadenosine 5',5"'-P1,P4-tetraphosphate (Ap4A) asymmetrical hydrolase: sequence similarity with the histidine triad (HIT) protein family.

Diadenosine 5',5"'-P1,P4-tetraphosphate (Ap4A) asymmetric hydrolase (EC 3.6.1.17) is a specific catabolic enzyme of Ap4A found in Schizosaccharomyces pombe. We have previously described the partial purification of Ap4A hydrolase from S. pombe [Robinson, de la Peña and Barnes (1993) Biochim. Biophys. Acta 1161, 139-148]. We determined the sequence of the N-terminal 20 amino acids of Ap4A hydrolase and designed two degenerate PCR primers based on the sequence. The 60 bp DNA fragment obtained by PCR, which is specific to Ap4A hydrolase, was used to isolate the Ap4A hydrolase gene, aph1, from S. pombe by screening a genomic DNA library in a multicopy plasmid. Ap4A hydrolase activity from the crude supernatant of a positive S. pombe transformant was about 25-fold higher than the control. There was no detectable stimulation of enzymic activity by phosphate. The aph1 gene from S. pombe contains three introns. The intron boundaries were confirmed by sequencing the cDNA of the aph1 gene from a S. pombe cDNA library. The deduced open reading frame of the aph1 gene codes for 182 amino acids. Two regions of significant local similarity were identified between the Ap4A hydrolase and the histidine triad (HIT) protein family [Séraphin (1992) DNA Sequence 3, 177-179]. HIT proteins are present in prokaryotes, yeast, plants and mammals. Their functions are unknown, except that the bovine protein inhibits protein kinase C in vitro. All four histidine residues which are conserved among the HIT proteins, including the HxHxH putative Zn(2+)-binding motif, are conserved in the Ap4A hydrolase. In addition, there are two regions of similarity between the Ap4A phosphorylases I and II from Saccharomyces cerevisiae and Ap4A hydrolase from S. pombe. These regions overlap with the HIT protein similarity regions. The aph1 gene from S. pombe is the first asymmetrical Ap4A hydrolase gene to be cloned and sequenced.

Synthesis and use of a chromogenic substrate analog for Ap4A catabolic enzymes.

ATP was coupled with 5-bromo-4-chloro-3-indolyl phosphate using a water-soluble carbodiimide to yield 5-bromo-4-chloro-3-indolyl tetraphospho-5'-adenosine (BCIp4A) which is an analog of diadenosine 5',5"'-P1,P4-tetraphosphate (Ap4A). BCIp4A is a chromogenic substrate for three different types of Ap4A catabolic enzyme in alkaline phosphatase-coupled reactions. Ap4A phosphorylase I from Saccharomyces cerevisiae was used as a model enzyme to demonstrate that BCIp4A stains for enzymic activity in polyacrylamide gels under nondenaturing conditions. A yeast colony assay was developed to detect Ap4A phosphorylase I activity in situ using BCIp4A as a chromogenic substrate. Ap4A phosphorylase I was assayed in situ in yeast transformed with a multicopy plasmid containing APA1, the gene encoding Ap4A phosphorylase I. BCIp4A should facilitate screening of genomic or cDNA libraries for genes encoding Ap4A catabolic enzymes.

Physical mapping of the Saccharomyces cerevisiae Ap4A phosphorylase I-encoding gene by the Achilles' cleavage method.

LacI-mediated Achilles' cleavage (AC) is a method for selective fragmentation of chromosomes at special lac operator sites introduced by gene targeting methods [Koob and Szybalski, Science 250 (1990) 271-273]. The Saccharomyces cerevisiae APA1 gene, coding for diadenosine 5', 5"'-P1, P4-tetraphosphate phosphorylase I, has previously been shown to be located on chromosome III [Kaushal et al., Gene 95 (1990) 79-84]. We have now used the AC method to map APA1 gene to a site 44 kb from the left terminus of the chromosome, between the HIS4 and HML genes. This location was confirmed by the comparison of restriction maps of the APA1 gene region to published restriction maps of chromosome III.

Changes in diadenosine tetraphosphate levels in Physarum polycephalum with different oxygen concentrations.

Cellular levels of diadenosine tetraphosphate (Ap4A) were measured, by a specific high-pressure liquid chromatography method, in microplasmodia of Physarum polycephalum subjected to different degrees of hypoxia, hyperoxia, and treatment with H2O2. Ap4A levels increased three- to sevenfold under anaerobic conditions, and the microplasmodia remained viable after such treatment. Elevated levels of Ap4A returned to the basal level within 5 to 10 min upon reoxygenation of the microplasmodia. The increases in Ap4A levels were larger in stationary-phase or starved microplasmodia than in fed, log-phase microplasmodia. The maximal increase measured in log-phase microplasmodia was twofold. No significant changes in Ap4A levels occurred in microplasmodia subjected to mild hypoxia, hyperoxia, or treatment with 1 mM H2O2. These results indicate that in P. polycephalum, Ap4A may function in the metabolic response to anaerobic conditions rather than in the response to oxidative stress.

Inhibition of prolactin secretion by antiserum to the alpha- and beta-subunits of gonadotropin.

This study was undertaken to determine whether antiserum to the alpha- and beta-subunits of human gonadotropin could affect prolactin secretion from human pituitary cell monolayers. The addition of antiserum to the alpha-subunit of follicle-stimulating hormone, but not to the alpha-subunit of luteinizing hormone, at a 1:10,000 dilution inhibited basal prolactin secretion. The addition of these antisera failed to inhibit prolactin secretion in the face of 10(-5) mol/L gonadotropin-releasing hormone. Coincubation of antiserum to the beta-subunit of luteinizing hormone or the beta-subunit of follicle-stimulating hormone at a 1:10,000 dilution failed to inhibit basal prolactin secretion; however, coincubation of gonadotropin-releasing hormone with antiserum to the beta-subunit of either luteinizing hormone or follicle-stimulating hormone significantly inhibited the release of prolactin. These observations led us to conclude that basal and gonadotropin-releasing hormone-mediated prolactin secretion can be affected by gonadotropin subunits.

In vivo levels of diadenosine tetraphosphate and adenosine tetraphospho-guanosine in Physarum polycephalum during the cell cycle and oxidative stress.

Cellular levels of diadenosine tetraphosphate (Ap4A) and adenosine tetraphospho-guanosine (Ap4G) were specifically measured during the cell cycle of Physarum polycephalum by a high-pressure liquid chromatographic method. Ap4A was also measured indirectly by a coupled phosphodiesterase-luciferase assay. No cell cycle-specific changes in either Ap4A or Ap4G were detected in experiments involving different methods of assay, different strains of P. polycephalum, or different methods of fixation of macroplasmodia. Our results on Ap4A are in contrast with those reported previously (C. Weinmann-Dorsch, G. Pierron, R. Wick, H. Sauer, and F. Grummt, Exp. Cell Res. 155:171-177, 1984). Weinmann-Dorsch et al. reported an 8- to 30-fold increase in Ap4A in early S phase in P. polycephalum, as measured by the phosphodiesterase-luciferase assay. We also measured levels of Ap4A, Ap4G, and ATP in macroplasmodia treated with 0.1 mM dinitrophenol. Ap4A and Ap4G transiently increased three- to sevenfold after 1 h and then decreased concomitantly with an 80% decrease in the level of ATP after 2 h in the presence of dinitrophenol. These results do not support the hypothesis that Ap4A is a positive pleiotypic activator that modulates DNA replication, but they are consistent with the hypothesis proposed for procaryotes that Ap4A and Ap4G are signal nucleotides or alarmones of oxidative stress (B.R. Bochner, P.C. Lee, S.W. Wilson, C.W. Cutler, and B.N. Ames, Cell 37:225-232, 1984).

Assay of diadenosine tetraphosphate hydrolytic enzymes by boronate chromatography.

A new procedure was described for assay of diadenosine tetraphosphate (Ap4A) hydrolases based on boronate chromatography. Potential reaction products, AMP, ADP, and ATP, of the hydrolysis of Ap4A were separated from residual substrate by chromatography on a boronate-derivatized cation-exchange resin, Bio-Rex 70. Separation was achieved by changing the concentrations of ethanol and ammonium acetate in the elution buffers. Picomole masses of products were detectable, blank dpm values were less than 0.5% of the total dpm, and auxiliary enzymes were not required. The procedure was specifically described for Ap4A pyrophosphohydrolase from Physarum polycephalum. The assay is generally applicable for dinucleoside polyphosphate hydrolases which hydrolyze other substrates such as Ap3A, Ap5A, Ap6A, and Gp4G. Dinucleotide polyphosphates are readily purified by chromatography on this boronate resin in a volatile buffer. Tes, Tricine, and Tris buffers significantly interfered with the chromatography of ATP.

Assay of adenosine 5'-P1-tetraphospho-P4-5"'-adenosine and adenosine 5'-P1-tetraphospho-P4-5"'-guanosine in Physarum polycephalum and other eukaryotes. An isocratic high-pressure liquid-chromatography method.

A5'pppp5'A has been proposed to serve as a molecular signal that triggers DNA replication. When published methods proved to be inadequate for the assay of A5'pppp5'A in Physarum polycephalum by h.p.l.c. (high-pressure liquid chromatography), a set of purification procedures was developed that allowed assay of as little as 2pmol of A5'pppp5'A. A5'pppp5'A was purified from cellular extract by covalent boronate chromatography, treated with alkaline phosphatase to hydrolyse residual mononucleotides and analysed by isocratic ion-exchange h.p.l.c. The analysis was facilitated by a pre-column switching procedure that allowed early-eluted species to be diverted from the analytical column. By using this procedure A5'pppp5'A has been detected in Physarum polycephalum (1.4 pmol/mg of protein), Saccharomyces cerevisiae (3.6 pmol/mg of protein) and rat liver (3.3 pmol/mg of protein). In each case a minor peak was also seen, which was identified as A5'pppp5'G. The identity of both peaks was confirmed by co-elution with standards on isocratic and gradient h.p.l.c. and treatment with enzymes, including a dinucleoside polyphosphate pyrophosphohydrolase from Physarum polycephalum.

Diadenosine 5',5"'-P1,P4-tetraphosphate pyrophosphohydrolase from Physarum polycephalum. Substrate specificity.

The substrate specificity of diadenosine 5',5"'-P1,P4-tetraphosphate pyrophosphohydrolase from Physarum polycephalum for dinucleoside polyphosphates has been determined by high-performance liquid chromatography (HP-LC). Elution of a strong anion-exchange resin with a pH and ionic strength gradient of ammonium phosphate separates a series of monoadenosine and diadenosine polyphosphates. Most of the corresponding guanine nucleotides are also resolved on this HPLC system. One mole each of Ap4A and Gp4G is symmetrically hydrolyzed to 2 mol of ADP and GDP, respectively. Ap3A, Ap5A, Ap6A, and Ap4 are hydrolyzed, and in each case ADP is one of the products. Gp3G, Gp5G, Gp6G, and Gp4 are also substrates, and in each case GDP is one of the products. AMP, ADP, ATP, Ap2A, ADPR, GMP, GDP, GTP, NAD+, and NADP+ are not substrates. No hydrolysis of the cap dinucleotides m7Gp3Am and m7Gp3Cm was detected by HPLC. Diadenosine tetraphosphate pyrophosphohydrolase preparations were also assayed for adenylate kinase, nucleotide diphosphate kinase, NAD(P)+ pyrophosphohydrolase, phosphodiesterase, cyclic nucleotide phosphodiesterase, phosphatase, and ribonuclease activities. These enzymic activities were not detectable in diadenosine tetraphosphate pyrophosphohydrolase. The symmetrical hydrolysis of Ap4A and Gp4G is an unique catalytic property that distinguishes diadenosine tetraphosphate pyrophosphohydrolase from P. polycephalum from diadenosine tetraphosphate phosphohydrolases from other organisms.

Cyclic AMP in the cell cycle of Physarum polycephalum.

Cyclic AMP, [3H]thymidine incorporation, and DNA content were measured in the cell cycle of Physarum polycephalum. A sensitive radioimmunoassay was employed to assay cyclic AMP so that plasmodia could be assayed individually. In contrast to previously published results (Lovely, J.R. and Threlfall, R.J. (1976) Biochem. Biophys. Res. Commun. 71, 789-795), no pre-mitotic peak of cyclic AMP was detected. In seven experiments levels of cyclic AMP showed only small changes in individual experiments and ranged from 1-6 pmol/mg protein in different experiments. When plasmodia in the immediate premitotic period were collected on the basis of nuclear mitotic morphology, no evidence of a peak of cyclic AMP was found. Light was found to increase plasmodial cyclic AMP in a rapid, transient fashion. However, the brief exposure of cell cycle samples to light during collection did not induce any apparent cell cycle specific peaks of cyclic AMP. Although the occurrence of extremely rapid transient peaks of cyclic AMP in the cell cycle cannot be ruled out, it appears likely that the P. polycephalum cell cycle can proceed normally without major changes in cyclic AMP.

Use of ethidium bromide fluorescence enhancement to detect duplex DNA and DNA bacteriophages during zone sedimentation in sucrose gradients: molecular weight of DNA as a function of sedimentation rate.

Duplex DNA molecules and DNA bacteriophages have been sedimented through 5--25% sucrose gradients containing ethidium bromide. The location of DNA within the gradients has been determined by illuminating gradients with ultraviolet light and observing the ethidium bromide fluorescence enhancement induced by the DNA. The relative sedimentation rates of linear, duplex DNAs from bacteriophages T4, T5, T7 and an 8.3% T7 deletion mutant have been determined. The distances sedimented by DNA have been corrected, when necessary, for a progressive decrease in sedimentation rate that occurs after the DNA has traversed 40% of the sucrose gradient. The corrected distances sedimented by two DNA molecules, r1' and r2', are related to the DNA molecular weights, m1 and m2, by the equation: r1'/r2' = (m1/m2)0.38 when 0.025--0.70 microgram of each type of DNA is sedimented. Intact bacteriophages were also sedimented in ethidium bromide--sucrose gradients and detected by fluorescence enhancement.