Sodium butyrate enhances the cytotoxic effect of cisplatin by abrogating the cisplatin imposed cell cycle arrest.

BACKGROUND: Histone deacetylase inhibitors have been proposed as potential enhancers of the cytotoxic effect of cisplatin and other anticancer drugs. Their application would permit the use of lower therapeutic doses and reduction of the adverse side effects of the drugs. However, the molecular mechanisms by which they sensitize the cells towards anticancer drugs are not known in details, which is an obstacle in developing effective therapeutic protocols. RESULTS: In the present work, we studied the molecular mechanisms by which sodium butyrate sensitizes cancer cells towards cisplatin. HeLa cells were treated with 5 mM butyrate, with 8 microM cis-diaminedichloroplatinum II (cisplatin), or with both. Cells treated with both agents showed approximately two-fold increase of the mortality rate in comparison with cells treated with cisplatin only. Accordingly, the life span of albino mice transfected with Ehrlich ascites tumor was prolonged almost two-fold by treatment with cisplatin and butyrate in comparison with cisplatin alone. This showed that the observed synergism of cisplatin and butyrate was not limited to specific cell lines or in vitro protocols, but was also expressed in vivo during the process of tumor development. DNA labeling and fluorescence activated cell sorting experiments showed that cisplatin treatment inhibited DNA synthesis and arrested HeLa cells at the G1/S transition and early S phase of the cell cycle. Western blotting and chromatin immunoprecipitation revealed that this effect was accompanied with a decrease of histone H4 acetylation levels. Butyrate treatment initially reversed the effect of cisplatin by increasing the levels of histone H4 acetylation in euchromatin regions responsible for the G1/S phase transition and initiation of DNA synthesis. This abrogated the cisplatin imposed cell cycle arrest and the cells traversed S phase with damaged DNA. However, this effect was transient and continued only a few hours. The long-term effect of butyrate was a massive histone acetylation in both eu- and heterochromatin, inhibition of DNA replication and apoptosis. CONCLUSION: The study presents evidence that cell sensitization towards cisplatin by sodium butyrate is due to hyperacetylation of histone H4 in specific chromatin regions, which temporarily abrogates the cisplatin imposed cell cycle arrest.

Fluorescence study of steroid hormone binding activity of Helix pomatia agglutinin.

Helix pomatia agglutinin (HPA) is a N-acetylgalactosamine (GalNAc) binding lectin, found in the reproductive gland of a Roman snail. The present study has shown that HPA, in addition to its carbohydrate binding capacity possesses a hydrophobic binding activity. This protein binds with high affinity (k(D)=1.9-2.4 microM) steroid hormones: testosterone and progesterone, identified as putative ligands for the animal lectin HPA. Additionally, we have found that this lectin also interacts with adenine (k(D)=5.4+/-0.5 microM) and arylaminonaphthalene sulfonate TNS (k(D)=12+/-0.3 microM). Binding of HPA to hormones and adenine was accompanied by a significant increase of the intrinsic Trp fluorescence (up to 50%), characterizing the conformational changes in the lectin molecule. The hyperbolic shape of the binding curves indicated one high affinity site for the two steroid hormones and adenine, and more than one hydrophobic site for TNS, showed by the sigmoidal curve fit and Hill coefficient of (n(H)=1.5+/-0.2). Hormones and adenine compete for an identical binding site, suggested to occupy the central hydrophobic cavity of the HPA hexamer. Fluorescence resonance energy transfer (FRET) was applied to calculate the intramolecular distance between TNS and Trp chromophores.

DNA interstrand crosslinks repair in mammalian cells.

We studied the formation of double strand breaks (DSBs) as intermediates in the repair of DNA interstrand crosslinks (ICLs) by homologous recombination (HR). The plasmid EGFP-N1 was crosslinked with trioxsalen to give one ICL per plasmid on average. HeLa cells were transfected with the crosslinked plasmids and the ICL repair was monitored by following the restoration of the GFP expression. It was accompanied by gamma-H2AX foci formation suggesting that DSBs were formed during the process. However, the same amount of gamma-H2AX foci was observed when cells were transfected with native plasmid, which indicated that gamma-H2AX foci appearance could not be used to determine the amount of DSBs connected with the ICL repair in this system. For this reason we further monitored the DSB formation by determining the amount of linearized plasmids, since having one crosslink per plasmid on average, any ICL-driven DSB formation would lead to plasmid linearization. Native and crosslinked plasmids were incubated in repair-competent cell-free extracts from G1 and S phase HeLa cells. Although a considerable part of the ICLs was repaired, no linearization of the plasmids was observed in the extracts, which was interpreted that DSBs were not formed as intermediates in the process of ICL repair. In another set of experiments HR-proficient HeLa and HR-deficient irs3 cells were transfected with native and crosslinked plasmids, and 6 h and 12 h later the plasmid DNA was isolated and analyzed by electrophoresis. The same amount of linear plasmid molecules was observed in both cell lines, regardless of whether they were transfected with native or crosslinked pEGFP-N1, which further confirmed that DSB formation was not an obligatory step in the process of ICL repair by HR.

Enhanced repair of DNA interstrand crosslinks in S phase.

Hela cells synchronized in G1 and S phases of the cell cycle were transfected with pEGFP crosslinked with trioxsalen. Twelve hours later the number of fluorescent cells was determined by fluorescent microscopy. Cells in S phase have repaired 0.2-0.3 ICL/kb over the 12h period, while cells in G1 phase repaired interstrand crosslinks much more poorly. The crosslinked plasmids were efficiently recruited to the nuclear matrix both in G1 phase and S-phase, which showed that the poor repair of G1 cells was a result of a lack of DNA replication rather than of a lack of matrix attachment.

Comparison of the global genomic and transcription-coupled repair rates of different lesions in human cells.

There are two subclasses of nucleotide excision repair (NER). One is the global genomic repair (GGR) which removes lesions throughout the genome regardless of whether any specific sequence is transcribed or not. The other is the transcription-coupled repair (TCR), which removes lesions only from the transcribed DNA sequences. There are data that GGR rates depend on the chemical nature of the lesions in a manner that the lesions inflicting larger distortion on the DNA double helix are repaired at higher rate. It is not known whether the TCR repair rates depend on the type of lesions and in what way. To address this question human cells were transfected with pEGFP and pEYFP plasmids treated with UV light, cis-diamminedichloroplatinum(II) (cisplatin) and angelicin and 24 h later the restored fluorescence was measured and used to calculate the respective NER rates. In a parallel series of experiments the same plasmids were incubated in repair-competent protein extracts to determine GGR rates in the absence of transcription. From the two sets of data, the TCR rates were calculated. We found out that cisplatin, UV light and angelicin lesions were repaired by GGR with different efficiency, which corresponded to the degree of DNA helix distortion induced by these agents. On the other hand the three lesions were repaired by TCR at very similar rates which showed that TCR efficiency was not directly connected with the chemical nature of the lesions.

Histone deacetylase inhibitor sodium butyrate enhances cellular radiosensitivity by inhibiting both DNA nonhomologous end joining and homologous recombination.

HDAC inhibitors have been proposed as radiosensitizers in cancer therapy. Their application would permit the use of lower radiation doses and would reduce the adverse effects of the treatment. However, the molecular mechanisms of their action remain unclear. In the present article, we have studied the radiosensitizing effect of sodium butyrate on HeLa cells. FACS analysis showed that it did not abrogate the γ-radiation imposed G2 cell cycle arrest. The dynamics of γ-H2AX foci disappearance in the presence and in the absence of butyrate, however, demonstrated that butyrate inhibited DSB repair. In an attempt to clarify which one of the two major DSBs repair pathways was affected, we synchronized HeLa cells in G1 phase and after γ-irradiation followed the repair of the DSBs by agarose gel electrophoresis. Since HR is not operational during G1 phase, by this approach we determined the rates of NHEJ only. The results showed that NHEJ decreased in the presence of butyrate. In another set of experiments, we followed the dynamics of disappearance of RAD51 foci in the presence and in the absence of butyrate after γ-radiation of HeLa cells. Since RAD51 takes part in HR only, this experiment allows the effect of butyrate on DSB repair by homologous recombination to be assessed. It showed that HR was also obstructed by butyrate. These results were confirmed by host cell reactivation assays in which the repair of plasmids containing a single DSB by NHEJ or HR was monitored. We suggest that after a DSB is formed, HDACs deacetylated core histones in the vicinity of the breaks in order to compact the chromatin structure and prevent the broken DNA ends from moving apart from each other, thus ensuring effective repair.

Tumor-specific protein human galectin-1 interacts with anticancer agents.

The present work shows a novel binding activity of the tumor specific lectin--recombinant human galectin-1 (hGal-1)--to three porphyrin compounds: (1) Zn-porphyrin (ZnTPPS); (2) Mn-porphyrin and (3) Au-porphyrin. These compounds are widely applied in the photodynamic therapy of cancer (PDT). Our data indicate that hGal-1, similar to some plant lectins, a bacterial lectin from Pseudomonas aeruginosa and an animal lectin from Helix pomatia, possesses dual functions binding to both carbohydrate and non-carbohydrate ligands. The interaction of ZnTPPS with hGal-1 was studied by the specific fluorescence emission of the porphyrin. The protein binding properties to Mn/Au-porphyrins and adenine were measured by intrinsic protein fluorescence quenching. The values determined for the apparent dissociation constants (K(D)) of 0.6-1.5 microM are similar to the K(D) for complexes of concanavalin A and porphyrin, and are indicative of the high affinity of hGal-1 for these porphyrins. In addition, the analysis of the hyperbolic binding curves obtained suggests the presence of one hGal-1 binding site for porphyrins or adenine. Additionally, we found that hGal-1 interacts with the fluorescent probe 2-(p-toluidinyl)naphthalene sulfonic acid (TNS), that was used to identify the hydrophobic regions within hGal-1. Homodimeric hGal-1 has more than one class of binding site for TNS as revealed by the sigmoidal shape of the fluorescence titration curve. hGal-1 can be characterized as a porphyrin-binding protein based on its interactions with the Zn/Mn- and Au-porphyrins, and this indicates that hGal-1 may have potential as a delivery molecule to target systems (e.g., tumor cells) with possible application in photodynamic therapy.

Repair of DNA interstrand crosslinks may take place at the nuclear matrix.

Host cell reactivation assay using Trioxsalen-crosslinked plasmid pEGFP-N1 showed that human cells were able to repair Trioxsalen interstrand crosslinks (ICL). To study the mechanism of this repair pathway, cells were transfected with the plasmids pEGFP-1, which did not contain the promoter of the egfp gene, and with pEGFP-G-, which did not contain the egfp gene. Neither of these plasmids alone was able to express the green fluorescent protein. After cotransfection with the two plasmids, 1%-2% of the cells developed fluorescent signal, which showed that recombination events had taken place in these cells to create DNA constructs containing the promoter and the gene properly aligned. When one or both of the plasmids were crosslinked with Trioxsalen, the recombination rate increased several fold. To identify the nuclear compartment where recombination takes place, cells were transfected with crosslinked pEGFP-N1 and the amount of plasmid DNA in the different nuclear fractions was determined. The results showed that Trioxsalen crosslinking increased the percentage of matrix attached plasmid DNA in a dose-dependent way. Immunoblotting experiments showed that after transfection with Trioxsalen crosslinked plasmids the homologous recombination protein Rad51 also associated with the nuclear matrix fraction. These studies provide a model system for investigating the precise molecular mechanisms that appear to couple repair of DNA ICL with nuclear matrix attachment.

DNase I sensitive site in the core region of the human beta-globin origin of replication.

HeLa cells were synchronized at late G1, early S, and late S phase of the cell cycle by nocodazole treatment. The cells were permeabilized with Triton X-100, digested with DNAse I, and extracted with 0.2 M ammonium sulfate to remove the digested chromatin. DNA was isolated from the residual chromatin attached to the nuclear matrix, digested with Hind III, and subjected to hybridization with [(32)P] labeled probe located upstream of the core region of the human beta-globin replication origin. The hybridization pattern revealed the existence of a DNase I sensitive site in the core region of the beta-globin replicator. The results suggest that association with the nuclear matrix induce alteration in the chromatin structure of the origin of replication that represents a more open chromatin configuration.

Nucleotide excision repair rates in rat tissues.

We have determined and compared nucleotide excision repair capability of several rat tissues by a method based on restoration of the transformation activity of UV-irradiated pBlueScript by incubation in repair-competent protein extracts. After 3 h of incubation, plasmid DNA was isolated and used to transform competent Escherichia coli cells. Damaged plasmids showed low transformation efficiency prior to incubation in repair-competent extracts. After incubation the transformation efficiency was restored to different extents permitting calculation of the repair capacity of the extracts. Our results showed that rapidly proliferating tissues such as liver, kidney and testis showed higher nucleotide excision repair capacity than slowly proliferating tissues such as heart, muscle, lung and spleen. When liver and splenocytes were stimulated to proliferation by partial hepatectomy and mitogen stimulation, their repair capability increased in parallel with the respective proliferative rates.

Relationship between DNA repair capacity and resistance to genotoxins in four human cell lines.

We have developed fast, reliable and simple fluorescent method to assess and compare repair capacity of cells. To this end plasmid pEGFP containing the gene for the enhanced green fluorescent protein was damaged in vitro by genotoxic agents and introduced into cells by transfection. The repair capacity of the cells was determined from the number of fluorescent cells counted with a fluorescent microscope 24 h after transfection. The ability of four human tumor cell lines--HEK293, HeLa, Namalwa and K562 to repair DNA lesions inflicted by cis-diamminedichloroplatinum(II), UV light, 8-methoxypsoralen and 4',5'-8-trimethylpsoralen were determined and compared to the survival rates of the cells after treatment with the same genotoxic agents. In most but not all cases, there was a good correlation between repair capacity and cell survival. This finding indicates that the DNA repair capacity could be used as a biomarker in risk assessment and/or drug resistance assays.