| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
Dipartimento di Patologia Sperimentale, Università degli Studi di Bologna, Italy;
* Dipartimento di Medicina Sperimentale, Sezione di Patologia Molecolare e Immunologia, Università degli Studi di Parma, Italy;
Cattedra di Farmacologia, Facoltà di Scienze Motorie, Università degli Studi di Urbino, Italy; and
Istituto di Farmacologia e Farmacognosia, Università degli Studi di Urbino, Italy
2Correspondence: Dipartimento di Patologia sperimentale, Via San Giacomo 14, 40126 Bologna, Italy. E-mail: brigotti{at}alma.unibo.it
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
50 Kb DNA fragments, and changes in morphology associated with apoptosis were negative. Furthermore, a block of translation comparable to that induced by RIPs, but obtained with cycloheximide, did not induce nuclear damage. Such damage is consistent with the enzymatic activity (removal of adenine) of RIPs acting in vitro on RNA-free chromatin and DNA. The results unequivocally indicate that RIPs can damage nuclear DNA in whole cells by means that are not secondary to ribosome inactivation or apoptosis.—Brigotti, M., Alfieri, R., Sestili, P., Bonelli, M., Petronini, P. G., Guidarelli, A., Barbieri, L., Stirpe, F., Sperti, S. Damage to nuclear DNA induced by Shiga toxin 1 and ricin in human endothelial cells.
Key Words: ribosome-inactivating proteins • alkaline-halo assay • AP sites • cytotoxins
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2
). Most RIPs are of plant origin and consist of a single chain (A chain) containing the catalytic site. Ricin and a few other plant RIPs contain in addition to the A chain a second chain (B chain) that binds to cell surface receptors mediating endocytosis of the RIP. Some of the two-chain RIPs, including ricin, are highly toxic.
Shiga toxin (Stx) from Shigella dysenteriae and the structurally and functionally related cytotoxins (Stx1, Stx2, and their variants) produced by Escherichia coli serotypes are unique among bacterial toxins in sharing RNA-N-glycosidase activity with plant RIPs (for review, see ref 3
). In Stxs, the enzymatic A chain is noncovalently associated to five B chains that mediate binding of the holotoxins to glycolipid receptors (usually Gb3, globotriaosylceramide, but also Gb4, globotetraosylceramide, in the case of some Stx2 variants) on target cells.
For almost two decades it has largely been assumed that RIPs act only on rRNA within ribosomes (4
, 5)
. Recently, however, all plant RIPs and Stxs tested have been shown to remove in vitro several adenine residues from naked RNAs and from DNA (6
7
8
9
10)
. In the latter instance, RIPs mimic repair DNA glycosylases (11)
whose activity is directed only against inappropriate or damaged bases and is often followed by cleavage of the DNA strand at the resulting apurinic/apyrimidinic (AP) sites (AP lyase activity). In contrast, in vitro depurination of DNA by plant RIPs, best observed at acidic pH (6
, 7)
but also occurring at neutral pH, removes a normal base (adenine) in the absence of DNA cleavage (12)
.
The damage to DNA induced in vitro by Stx1 both at acidic and physiological pH has been studied extensively using as substrate an [3H]DNA labeled in the adenine ring (9
, 13
, 14)
. Data indicate that as with plant RIPs, damage involves primarily the release of adenine from multiple sites and that the single-strand breaks observed are not due to an associated AP lyase activity but to a weakening of the DNA sugar-phosphate backbone after extensive adenine removal (14)
. Damage to DNA might be relevant to the mechanism of action of Stxs, whose role in the pathogenesis of bacillary dysentery, enterohemorrhagic colitis, and hemolytic uremic syndrome (HUS) is well documented (3)
. The present paper addresses the damage to DNA induced by Stx1 in human umbilical vein endothelial cell (HUVEC) containing the Gb3 receptor and thus susceptible to the toxin (15)
. The effect of the plant RIP ricin from Ricinus communis on the same cells was also investigated.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and saporin-L1 (a single-chain RIP from the leaves of Saponaria officinalis) as described in ref 17
. Stx1 obtained from a prototype Stx1 producer E. coli C600 (H19J), kindly supplied by Dr. Alison O’Brien (Department of Microbiology and Immunology, USU of the Health Sciences, Bethesda, MD), was purified by receptor analog affinity chromatography (18)
on globotriose-Fractogel (IsoSep AB). L-[4,5-3H]leucine (2.5 mCi/mmol, 2 µCi/ml) was obtained from Amersham Pharmacia Biotech (Bucks, U.K.) and [14C]thymidine (250 µCi/mmol, 0.05 µCi/ml) from NEN/Dupont (Boston, MA). Disposable plastics for laboratory use were obtained from Costar (Broadway, Cambridge, MA). Reagents of analytical grade were purchased from Sigma Chemical Co. (St. Louis, MO).
Cell cultures
HUVEC cultures (19)
were kindly provided by Dr. Janette A. M. Maier (University of Milan, Italy). Cells were maintained in complete growth medium M199 containing antibiotics (100 U/ml penicillin, 100 µg/ml streptomycin) and supplemented with 1.4 mM glutamine, 10% fetal bovine serum, 50 µg/ml endothelial cell growth factor. All cultures were kept in an incubator at 37°C in a water-saturated 5% CO2 atmosphere in air. Subpassages were made by rinsing confluent monolayers with phosphate-buffered saline (PBS) and detaching cells by treatment with a trypsin-EDTA solution. Cell suspensions were inoculated into new flasks coated with 2% sterile gelatin at a density of 5 x 104/cm2 and cells were used when confluent, usually within 3 days after plating. Cells used in this study were all of early (< 10th) passages and were routinely checked for the expression of von Willebrand factor by immunocytochemistry using a rabbit anti-human von Willebrand factor (Dako, Milan, Italy) as primary antibody.
Determination of protein synthesis
Protein synthesis was measured as the rate of incorporation of labeled leucine during a 30 min incubation of the cell monolayers in complete minimal essential medium containing 0.4 mM leucine and trace amounts of [3H]leucine. This procedure has been described in detail elsewhere (20)
.
Damage to DNA measured by the alkaline-halo assay in intact cells
The alkaline-halo assay described in ref 21
with minor modifications was used. After treatment with RIPs, the cells were resuspended at 2.0 x 104 cells/100 µl in 1.5% low-melting agarose in PBS containing 5 mM EDTA and immediately sandwiched between an agarose-coated slide and a coverslip. After complete gelling, the coverslips were removed and the slides were immersed in an alkaline hypotonic buffer (0.1 M NaOH/1 mM EDTA, 
pH 13, for 20 min), washed, and stained for 5 min with 10 µg/ml ethidium bromide. The ethidium bromide-labeled DNA was visualized using a Bio-Rad DVC 250 confocal laser microscope (Bio-Rad, Richmond, CA); the resulting images were taken and processed with a Hamamatsu chilled CCD 5985 camera (Hamamatsu Italy S.p.a., Milan, Italy) coupled with an Apple Macintosh computer using the public domain NIH Image program (developed at the U.S. National Institutes of Health and available on the Internet at http://rsb.info.nih.gov/nih-image/). The level of DNA single-strand breakage was quantitated by calculating the nuclear spreading factor value, which represents the ratio between the area of the halo (obtained by subtracting the area of the nucleus from the total area, e.g., nucleus+halo) and that of the nucleus, from 50 to 75 randomly selected cells/experiment/treatment conditions. Data are expressed as relative nuclear spreading factor values (rNSF), calculated by subtracting the nuclear spreading factor value of control cells from that of treated cells. It should be stressed that the method does not distinguish between primitive DNA breaks and the formation of AP sites. In fact, cleavage of DNA at the AP sites occurs during incubation of the samples with the highly alkaline denaturing buffer.
Damage to DNA measured by the alkaline filter elution assay
Cells were labeled for 24 h with [14C]thymidine and, after 6 h incubation in a label-free medium, treated with RIPs. Samples were withdrawn immediately after damage induction and analyzed by the alkaline filter elution assay. This procedure was carried out as described by Kohn et al. (22)
. Briefly, 5 x 105 cells were gently loaded onto 25 mm, 2 µm pore polycarbonate filters and rinsed twice with 10 ml of ice-cold PBS containing 5 mM Na2EDTA. Cells were then lysed with 5 ml of 2% sodium dodecyl sulfate (SDS), 0.025 M Na4EDTA, pH 10.1, and the lysates were rinsed with 7 ml of 0.02 M Na4EDTA. Finally, DNA elution was carried out in the dark with tetraethyl ammonium hydroxide/0.02 M EDTA (free acid)/0.1% SDS, pH 12.1, at a flow rate of 0.04 ml/min. The same eluting solution buffered at pH 12.6 by addition of an appropriate amount of tetraethyl ammonium hydroxide was used to assess the presence of alkali-labile sites (23)
. Fractions of 3.5 ml were collected and processed for liquid scintillation counting. The amount of DNA remaining on the filters and of the DNA recovered from the interior of the membrane holders was also determined by scintillation counting. Under these conditions, intact DNA remains on the filters in contrast to single-stranded DNA fragments, which pass through the membrane during the elution process.
Detection of apoptosis
Early phases of apoptosis were evaluated as described previously (24
, 25)
by 1) viable cell staining for DNA with the permeant fluorescent stain Hoechst 33342 performed simultaneously with a dye exclusion test using propidium iodide; 2) FITC-conjugated annexin V assay (Bender MedSystems) based on annexin V binding to phosphatidylserine exposed on the outer leaflet of the plasma membrane lipid bilayer of cells entering the apoptotic pathway; 3) caspase 3 activity (Caspase Colorimentric Assay kit, MBL Intern. Corp., Watertown, MA). The latter assay is based on spectrophotometric detection of the chromophore p-nitroanilide (pNA) after cleavage from the labeled substrate DEVD-pNA. The ratio between the absorbance of pNA from the apoptotic sample and that from an untreated control gives the fold increase in caspase 3 activity (relative activity). For the assay, adherent HUVEC were scraped from dishes and added to spontaneously detached cells when present. Cells were washed by centrifugation and resuspended in lysis buffer. After incubation for 10 min on ice, cell debris were centrifuged at 12 000 g for 2 min at 4°C and the supernatants were assayed for caspase activity. Protein content of the lysate supernatants was measured using the Bio-Rad protein assay (26)
.
Electrophoresis of genomic DNA
Electrophoretic analysis of double-stranded DNA fragments was performed using the programmable autonomously controlled electrode (PACE) assay. Preparation of agarose plugs and the running conditions were as described earlier (27)
. PACE electrophoresis was carried out using a Bio-Rad DRIII variable angle system (Bio-Rad). The gels were cast using 1% w/v chromosomal grade agarose in 0.5x TBE buffer [composition of the 0.5x concentrated buffer: 44.5 mM Tris/HCl, 44.5 mM boric acid, 1 mM Na2EDTA (pH 8.3)] and run with a three block assay. During the first block, the switch time was 75 min for 5 h at 14°C, with a field strength of 1.2 V/cm and a reorientation angle of 106°. In the second block the temperature was decreased to 9°C, the switch time was 30 min for 30 h, and the reorientation angle was 106° using a field strength of 2.2 V/cm. In the third block the switch time was linearly ramped from 10 to 90 s for 7 h and field strength increased to 6 V/cm. The reorientation angle was changed to 120°. Gels were stained with ethidium bromide, viewed with an UV transilluminator, and photographed.
Release of adenine from RNA-free chromatin
For this assay, ricin was activated by 1 h incubation at 37°C with 5% 2-mercaptoethanol (28)
. Activation of Stx1 was performed by treatment with trypsin, urea, and dithiothreitol (29)
. Nuclei were isolated as described elsewhere (30)
with slight modifications. After removal of the culture medium, cell monolayers were washed with ice-cold PBS, then cells were scraped into the same buffer and pelleted at 500 g for 5 min. The pellet was resuspended in 2 vol of lysis buffer (10 mM Tris/HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl2, 0.5% (v/v) Nonidet P-40); after 5 min of incubation on ice, the mixture was centrifuged at 500 g for 10 min. The pelleted nuclei were resuspended in 20 mM HEPES/KOH, pH 6.8, 15 mM KCl, 1.5 mM MgCl2, 1.5 mM DTT and sedimented through 1.7 M sucrose in the same buffer by centrifuging 100 min at 33 000 g. Chromatin was prepared by homogenizing nuclei in the presence of 0.075 M NaCl containing 0.024 M EDTA, pH 7.9, followed by treatment with decreasing NaCl concentrations as described in ref 31
. Incubation for 5 min at 25°C with RNAase A (10 µg/ml) finally yielded RNA-free chromatin. Adenine released by RIPs from RNA-free chromatin was measured as described in ref 32
except for quantification, which was by LC/MS on a Waters Alliance/zq apparatus. Chromatography to separate adenine was on a Waters XTerra MS C18 column (2.1x50 mm, 2.5 µm beads) equilibrated and eluted with 10 mM ammonium acetate (solvent A)/methanol (solvent B) at 0.3 ml/min at 15°C. Equilibration was in 98/2 (A/B). After sample injection (120 µl), the column was washed (2 min) with equilibration solvent and eluted with 90/10 (A/B) for 5 min. Tightly bound material was finally eluted with 20/80 (acetonitrile/A) (0.6 min) and riequilibration was obtained with 1.2 min of 90/10 (A/B), followed by 9 min of 98/2 (A/B). MS analysis was in positive electrospray with single ion recording (135+1 m/z) on a splitted flow of 50 µl/min. Parameters were optimized manually for maximum sensitivity and duplicate chromatograms were combined with Micromass MaxLynx software to reduce noise.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. It is now generally assumed that the major portion of the histopathological lesions observed in these conditions is the consequence of the interaction of these toxins with the endothelial lining of intestine and kidney, in addition to the interaction with the absorptive villus epithelial cells in the gut.
Previous studies have shown that HUVEC are sensitive to the toxic action of Stxs (15
, 33
34
35)
. Since besides depurinating RNA 28S within ribosomes, ricin and Stx1 also depurinate DNA in vitro, the rate of protein synthesis of HUVEC treated with the two toxins was measured and the relationship with a possible nuclear DNA damage was investigated.
In the alkaline-halo assay (21)
used to assess DNA damage in whole cells, control and toxin-treated cells are embedded in melted agarose, spread onto microscope slides, and lysed with a hypotonic solution at denaturing pH (pH 13), which induces the radial diffusion from the nucleus of single-stranded DNA fragments and generates, upon incubation with ethidium bromide, a fluorescent image similar to a halo concentric to the nuclear remnants. The greater the level of DNA fragmentation the bigger the area of the halo is (Fig. 1
), thus allowing a quantitative determination of the nuclear injury. Quantification of the nuclear damage can thus be pinpointed to the level of individual cells. Figure 1
clearly shows that both Stx1 and ricin, at very low concentrations (see above), possess DNA damaging activity on HUVEC. In our experiments, heterogeneity in the response of HUVEC to Stx1 and ricin was never observed.
|
In Fig. 2
the time course of inhibition of protein synthesis is compared with the DNA damage measured as rNSF in HUVEC cultured in the presence of Stx1 (0.01 nM) or ricin (1 nM). Calculation of the rNSF values was from 50–75 selected images/treatment condition. Although starting somewhat earlier in the case of Stx1, inhibition of protein synthesis with both RIPs reaches a plateau value approaching 100% within 3 h. Nuclear spreading follows the same time course with ricin (Fig. 2B
) but is delayed 3 h with Stx1 (Fig. 2A
).
|
Ricin induces apoptosis in organs of poisoned animals (36)
, and several reports have claimed that apoptosis is triggered by Stxs and ricin in different cell types (37
38
39
40
41
42
43
44)
. Therefore, a crucial point to address was the demonstration that the nuclear damage detected in HUVEC by the alkaline-halo assay was directly induced by RIPs and not secondary to the onset of apoptosis.
An early event in apoptosis is the translocation of phosphatidylserine from the inner to the outer layer of the plasma membrane, and an early test for apoptosis is the binding of annexin V, a membrane-impermeable protein, to the externalized phosphatidylserine. The annexin V binding test was negative for both Stx1 and ricin at the time of maximal DNA spreading observed (6 and 3 h, respectively) whereas doubling the time of treatment (12 and 6 h, respectively) induced a gradual increase in the number of cells exhibiting annexin V reactivity (data not shown).
Independent from the initial signal and the subsequent intracellular molecular events, diverse apoptotic pathways converge in the activation of executioner caspases, such as caspase 3, and finally in the large-scale digestion of nuclear DNA with the formation of 50 Kb-paired DNA fragments that are considered an unequivocal landmark of apoptosis (45
, 46)
. Figure 2
shows that the nuclear spreading induced by Stx1 and ricin appears before the onset of the execution phase of apoptosis as measured by the cleavage of a colorimetric substrate specific for caspase 3. With Stx1 (Fig. 2A
), the caspase activity was found increased (threefold) after 16 h of treatment of the cells, i.e., more than 10 h after the maximum nuclear spreading observed, at which time no caspase 3 activity was detectable. With ricin (Fig. 2B
), the same threefold increase was observed 3 h after the maximal value of nuclear spreading detected and a fivefold increase appeared after 16 h of treatment. Finally, PACE analysis (which allows the simultaneous determination of DNA double-strand breaks and DNA fragment size, see ref 47
) of DNA from toxin-treated cells (6 and 3 h of exposure to Stx1 and ricin, respectively; see Fig. 2
) did not reveal the typical and discrete formation of 50 Kb paired DNA fragments (Fig. 3
) (27
, 45
, 46)
. Indeed, DNA fragments 50 Kb can be seen in HUVEC subjected to an apoptotic regimen (10 µM etoposide for 30 min, followed by another 6 h in drug-free complete culture medium) included as a positive control (48)
. PACE analysis also reveals that treatment of HUVEC with the two toxins does not result in random generation of double-stranded DNA fragments, indicating that neither Stx1 nor ricin is able to produce DNA double-strand breaks.
|
Apoptosis is associated with typical changes in cellular morphology. After 24 h of treatment with Stx1 or ricin, 30% of the cells exhibited typical changes in morphology, including chromatin condensation and fragmentation into apoptotic bodies, visualized with a fluorescent cell-permeant DNA-stain (Fig. 4
). No such figures were observed earlier, when nuclear spreading was observed, or in untreated cells.
|
Taken together, the negative response of the annexin V binding assay, the delayed activation of caspase 3, and the absence of large-scale (50 Kb) DNA cleavage and morphological modifications strongly suggest that the early damage to DNA observed with the toxic RIPs precedes apoptosis and may well be the consequence of the enzymatic activity of Stx1 and ricin on DNA.
The single-stranded DNA fragments visualized by alkaline-halo assay may result from a frank, direct breakage of the DNA strands or, alternatively, from the presence of alkali-labile sites such as apurinic sites, which are converted to DNA single-strand breaks under alkaline hydrolysis at high pH values of 13. The experiments shown in Fig. 5
were specifically designed to investigate the nature of the DNA lesions in HUVEC treated with Stx1 (0.01 nM, 6 h) and ricin (1 nM, 3 h). The alkaline filter elution, a well-established assay to detect the presence of DNA lesions in genomic DNA (23)
, was adapted to detect the presence of alkali-labile sites. The modification consists simply of the elution of the same samples at two different pH values, i.e., 12.1 and 12.6 (23)
. The higher the pH, the more rapid the hydrolysis of alkali-labile sites. As a result, at the higher pH value and in the presence of alkali-labile sites, the elution rate of single-stranded DNA fragments is faster and the amount of DNA retained in the filter is lower vs. DNA eluted at pH 12.1. The elution profiles at pH 12.1 and pH 12.6 of toxin-treated cells are compared in Fig. 5
. DNA from Stx1- and ricin-treated cells both exhibit a higher rate of elution at pH 12.6. CaCrO4, an agent that induces alkali-labile sites, has also been included as a positive control (49)
. The importance of these findings is twofold: 1) the DNA damaging activity of Stx1 and ricin has been confirmed with an alternative, independent technique, and 2) the differences of the elution profiles at pH 12.1 and 12.6 suggest that the DNA damaging activity of the toxins may result from the presence of alkali-labile sites produced by the depurinating activity of the two toxins on DNA. This hypothesis is supported by experiments reported in Table 1
on the release of adenine from isolated RNA-free chromatin after incubation with the toxins at physiological pH. The release occurred with both Stx1 and ricin. In these experiments, relatively high concentrations of toxins had to be used to produce a release of adenine directly detectable with the present technique. These amounts of Stx1 and ricin would lead to the generation of a very high number of AP sites (106) for chromatin derived from a single nucleus. Thus, even a much lower depurination rate obtained in HUVEC treated with lower concentrations (Figs. 2
and 5)
of toxins may well explain the nuclear injury observed in these cells using more sensitive techniques, such as the alkaline-halo assay (Fig. 2)
and the alkaline filter elution assay (Fig. 5)
.
|
|
We can conclude that the nuclear damage to DNA induced by Stx1 and ricin is consistent with the enzymatic activity (removal of adenine) of RIPs and involves the formation of AP sites and single-stranded DNA breaks, the latter probably arising from the action of DNA repair enzymes on AP sites.
AP sites are unceasingly formed in cells and continuously repaired by specialized enzymes. By reducing the turnover of DNA repair enzymes, the arrest of protein synthesis induced by RIPs might conceivably allow the accumulation of spontaneous AP sites. In this case, a nuclear damage appearing simultaneously (ricin) or delayed (Stx1) with respect to the inhibition of protein synthesis might be merely a secondary effect of a pure translation inhibitor. Figure 6
shows, however, that prolonged inhibition of protein synthesis (< 6 h) induced by an inhibitor of translation having a different mechanism of action such as cycloheximide is neither accompanied nor followed by DNA damage as revealed by nuclear spreading. Moreover, cycloheximide did not induce apoptosis in HUVEC at the dosage (1 µg/ml) and time (24 h) tested as evaluated by the morphological appearance of cells and an absence of caspase 3 activity (Fig. 6)
. Although a delayed apoptosis triggered by high doses of cycloheximide has been reported in other cell lines (50)
, it is well known that cycloheximide at a low dosage, rather than inducing apoptosis, has a protective effect in HUVEC challenged with various proapoptotic stimuli such as serum depletion (51)
or tissue factor pathway inhibitor (52)
. The data with cycloheximide (Fig. 6)
vs. those obtained with ricin and Stx1 (Fig. 2)
once again stress the point (53)
that a strong inhibition of protein synthesis is not sufficient per se to trigger apoptosis. Programmed cell death induced by toxic RIPs occurs only when both protein synthesis inhibition and nuclear damage have been elicited.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
8
9)
in vitro. Moreover, the data clearly indicate that the early DNA damage, observed in parallel or after the arrest of protein synthesis in HUVEC treated with ricin and Stx1 is not a consequence of apoptosis. Only after considerable lag time do the intoxicated cells in fact choose to participate in their own demise through the apoptotic pathway.
Internalized Stx1 is known to reach the nuclear envelope (54)
, and cells treated with Stx1 show the toxin predominantly in the nuclear fraction (40)
. However, to our knowledge this is the first report that Stx1 directly challenges the integrity of nuclear DNA. In view of the present results, the possibility should be considered that a failure of the suicide program in cells intoxicated with this medically important toxin might lead to mutagenic effects.
The mechanism by which ricin induces apoptosis has been studied. The involvement of various caspases, caspase-like proteases, serine proteases (43
, 44)
, and poly(ADP-ribose) polymerase cleavage activity (44)
has been reported. Hu et al. (44)
also suggested that protein synthesis inhibition was not the sole cause of ricin-induced apoptosis. In contrast, the mechanism of apoptosis induced by Stxs is not fully understood. Many reports (reviewed in ref 55
) suggest that multiple pathways, some of which are peculiar to a particular member of the Stx family, are involved. Among the most accredited pathways are 1) inhibition of protein synthesis (suggested for all Stxs); 2) interaction of the B chains with the Gb3 receptor on target cells (Stx1; ref 37
) and/or direct activation of the caspase cascade by the internalized B chains (Stx1; ref 39
); and 3) interaction of the holotoxin with mitochondrial Bcl-2 resulting in caspase 3 activation (Stx2; ref 40
). At odds with Stx1, Stx2 is detected not only in the nucleus, but also in the mitochondrial fraction, where its A chain interacts directly with Bcl-2, forming a complex necessary for cell death induction in target cells. Our data obtained with HUVEC cultured in the presence of a very low concentration of Stx1 (0.01 nM) point to an additional mechanism of apoptosis centered on the enzymatically induced damage of DNA. A novel mechanism in triggering apoptosis can be argued by the inability of Stx1 to interact with Bcl-2 (40)
and by the very high concentration (100 nM) required for apoptosis mediated by the Stx1 B chain alone (37)
. Moreover, data obtained with cycloheximide on HUVEC support the view that inhibition of translation by Stx1 is not sufficient to induce programmed cell death. It seems likely that in the fully evolved apoptotic program, several mechanisms (inhibition of translation, transduction of signal through the Gb3 receptor, and DNA damage) might be concurrent and interactive.
In addition to its role in the induction of cell death, depurination of DNA by Stx1 may be involved in more subtle effects at the cellular level. It has been suggested that the microangiopathic lesions typical of hemolytic uremic syndrome might be caused by Stx through the block of the production of procoagulating or anticoagulating molecules and through an imbalance between vasodilators and vasoconstrictors produced by endothelial cells (3)
. Moreover, several recent data indicate that a cooperation between Stxs and inflammatory cytokines is fundamental in the pathogenesis of hemolytic uremic syndrome (55)
. Dealing with a human colon epithelial cell line, Yamasaki et al. (56)
have demonstrated that extremely low concentrations of Stx1 and Stx2 ( 0.1 pM) are sufficient to induce interleukin 8, tumor necrosis factor 
, and MCP-1 production and that the enzymatic activity of Stx1 is essential for cytokine induction. In the presence of such low concentrations of toxins, a selective removal of very few adenines, undetectable by the alkaline-halo assay but occurring in critical DNA sequences of endothelial cells, might well induce alterations in gene expression and regulation. A validation of this hypothesis and the location of the genes targeted by the toxin await the identification of DNA recognition sequences for Stxs.
This work was supported by grants from MURST, Ministero della Salute, Pallotti’s Legacy for Cancer Research, University of Bologna (Funds for selected research topics).
|
FOOTNOTES |
|---|
Received for publication July 13, 2001.
Revision received November 15, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-D-galactose-(1
4)-ß-D-galactose-(1 4)-ß-D-glucose-(1) receptor ligand-based chromatography. FEBS Lett 258,320-322[CrossRef][Medline]
antisense oligomer. Science 249,1570-1574
B. Infect. Immun. 65,3337-3344[Abstract]
-sarcin Induction of cell death via apoptosis. Eur. J. Biochem. 268,2113-2123[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
