Hydroxyurea-The Good, the Bad and the Ugly

Abstract

Hydroxyurea (HU) is mostly referred to as an inhibitor of ribonucleotide reductase (RNR) and as the agent that is commonly used to arrest cells in the S-phase of the cycle by inducing replication stress. It is a well-known and widely used drug, one which has proved to be effective in treating chronic myeloproliferative disorders and which is considered a staple agent in sickle anemia therapy and-recently-a promising factor in preventing cognitive decline in Alzheimer's disease. The reversibility of HU-induced replication inhibition also makes it a common laboratory ingredient used to synchronize cell cycles. On the other hand, prolonged treatment or higher dosage of hydroxyurea causes cell death due to accumulation of DNA damage and oxidative stress. Hydroxyurea treatments are also still far from perfect and it has been suggested that it facilitates skin cancer progression. Also, recent studies have shown that hydroxyurea may affect a larger number of enzymes due to its less specific interaction mechanism, which may contribute to further as-yet unspecified factors affecting cell response. In this review, we examine the actual state of knowledge about hydroxyurea and the mechanisms behind its cytotoxic effects. The practical applications of the recent findings may prove to enhance the already existing use of the drug in new and promising ways.

Keywords: DNA replication checkpoint; cell cycle arrest; hydroxyurea; replication stress; ribonucleotide reductase.

Conflict of interest statement

The authors declare no conflict of interest. The founder had no role in the design of the study; in the data collection, analysis and interpretation; in the writing of the manuscript, or in the decision to publish.

Figures

Figure 1

The structures of two homodimeric subunits of ribonucleotide reductase and the impact of hydroxyurea on the PCET transfer: (A) the dATP in the allosteric center of subunit α2 of human RNR (PDB: 2WGH) [36]; (B) the Fe(3+) ion center of the human β2 subunit (PDB: 2UW2); (C) a 3D model of the α2 subunit with the allosteric sites and active (substrate) sites (PDB: 2WGH); (D) a 3D model of the β2 subunit with the di-iron center placement marked (PDB: 2UW2). (E) Bidirectional PCET can occur only when both subunits form an α2β2 heterodimeric structure. This binding allows for substrate–effector interaction. (F) Hydroxyurea inhibits PCET within RNR by proton-coupled electron transfer, most probably mediated by a hydrogen-bonded proton wire [35]. The images in (A,B) were generated using Mol*Viewer [46].

Figure 2

The cell response to HU-induced depletion of dNTPs during replication. Origins licensed on the DNA strand are grouped into replication “factories” the activation of which is strictly defined in time, from early- to late-replicating. In normal conditions, only a small percentage of origins will become activated and start to replicate DNA; the other origins remain dormant and will be replicated passively unless they are needed (A). If the currently replicating factory already has the desired amount of active forks, an ATR/Chk1 signaling mechanism limits dormant origin firing and also limits the activation of other factories that are supposed to replicate later (C). In the absence of free dNTPs (B), active forks stall and expose fragile ssDNA (B,C). ATR kinases stabilize stalled forks and also trigger the signaling pathway of the S-phase checkpoint, which arrests the cell cycle, inhibits origin firing in inactive factories, and restores RNR activity (C). The cell is unable to enter mitosis unless DNA replication is finished and any damage repaired (C).

Figure 3

HU changes the activity of replication as well as several key proteins in root meristem cells of Vicia faba. (A) Nuclear DNA labeled with 5-bromo-2′-deoxyuridine (BrdU; a1–a3); (B) PCNA (b1–b3); (C) Chk1 kinase (c1–c3). The DNA within the nucleus is detected with propidium iodide (orange; c1). The control levels of Chk1 are consistent with reports that cells maintain a basic level of kinases in order to prevent dormant origin firing (c1,c3). (D) Histone H2AX (d1–d3), which is a hallmark of double-strand DNA damage (H2AXS139ph), shows increased levels after HU exposure (d2,d3). The presented factors were detected using immunocytochemical staining after 32 h incubation in water (first column, normal conditions; a1,b1,c1,d1) or in 2.5 mM HU (second column; a2,b2,c2,d2). The third column contains 3D renderings of BrdU/protein localization within the nucleus, based on statistical analysis of experimental data and microscope observations (a3,b3,c3,d3). The left side of the column shows the agent’s activity under normal conditions and the right panel shows the activity after HU treatment. The experimental procedures for immunocytochemical detection of: (i) PCNA, (ii) Chk1 (phosphorylated on serine 317), and (iii) H2AX (phosphorylated on serine 139) were identical (BrdU detection required hydrolysis with HCl). The experimental procedure for the DNA replication assay with BrdU (on entire cells) was as follows: control and HU-treated (2.5 mM, 24 h) seedlings were pulsed for 30 min with 30 µM BrdU solution at 20 °C in the dark. Excised 3 mm long meristems were then washed with ice-cold Tris buffer (10 mM Tris, 10 mM EDTA-2Na, 100 mM NaCl, pH 7.2) for 5 min and fixed for 45 min at 4 °C in freshly prepared 4% paraformaldehyde. After fixation, root tips were washed, squashed onto slides, and treated with 1.5 M HCl (for 1.5 h at 20 °C, for partial denaturation of nuclear DNA). In contrast, detection of PCNA/Chk1S317ph/H2AXS139ph required an enzymatic maceration step, i.e., incubation in a citric acid-buffered digestion solution for 45 min at 37 °C. Subsequently, root tips were washed again, squashed onto slides, and treated with the following primary antibodies: (i) mouse monoclonal anti-BrdU (Sigma-Aldrich, Saint Quentin, France) diluted in Tris-buffer (1:50) or (ii) the rabbit monoclonal antibodies anti-PCNA (Abcam, Cambridge, United Kingdom), anti-Chk1(S317ph) (Cell Signaling Technology, Beverly, MA, USA), and anti-H2AX(S139ph) (Cell Signaling Technology; Beverly, MA, USA; 1:250), diluted in PBS. Following overnight incubation (at least 16 h) at 4 °C, slides were washed in Tris/PBS buffer, respectively, and incubated for 1 h with AlexaFluor 488-conjugated goat anti-mouse/mouse anti-rabbit secondary antibodies, respectively (Cell Signaling Technology, Beverly, MA, USA; 1:500), washed, and embedded in PBS/glycerol mixture (9:1) with 3% DABCO.

Figure 4

The consequences of ATR inhibition and malfunction of the S-phase checkpoint. A disrupted ATR/Chk1 pathway does not arrest the cell in the S phase and allows cycle progression with under-replicated DNA. This causes the accumulation of SSBs and DSBs, fork collapse, and deregulation of the inhibition of late origin firing. Moreover, a lack of radical maintenance and inhibition of new RNR synthesis leads to the accumulation of HU-induced ROS. The circled images show anaphase aberrations detected using Feulgen staining (top) and the accumulation of ROS (H2O2) in the cytoplasm in DAB-stained root meristem cells [99] (bottom) of Vicia faba subsp. minor. The experimental procedure for 3,3-diaminobenzidine (DAB) staining (to detect H2O2 by means of DAB polymerization) was as follows: roots of V. faba were submerged in Tris-buffered (10 mM Tris, 10 mM EDTA-2Na, 100 mM NaCl) DAB-HCl (1 mg mL−1; pH 7.5) dissolved in distilled water or in 2.5 mM HU (HU-treated plants). Following 24 h incubation, excised 3 mm root tips were fixed for 45 min in PBS-buffered 4% paraformaldehyde (20 °C), washed (3 times) with PBS, and placed in citric acid-buffered 2.5% pectinase (pH 5.0; 37 °C for 45 min). Digested root meristem cells were washed with PBS and squashed onto microscope slides in a mixture of glycerol and PBS (9:1; v/v). Production of H2O2 was observed under the microscope as reddish-brown areas in the cells (indicated by a black arrow in the circled image).

Figure 5

Visual examples of cell death, DNA damage, and chromosome breakage induced by overexposure to HU or bypassing of the S-phase cycle induced by 2.5 mM HU and 5 mM caffeine (CF) in root meristem cells of Vicia faba subsp. minor. (A) Programmed cell death (PCD) visible after double staining with acridine orange (AO) and ethidium bromide (EB) (a1,a2) [100]. AO is able to enter living cells and makes them appear green (a1). EB is taken up only when cell membrane integrity is compromised (usually in dead or dying cells), making them look red (a2). The range of colors (from green through yellow to red) indicates the existence of living, dying, and dead cells. The scale bar equals 50 µm. (B) DNA damage observed under normal conditions (b1) and after HU treatment (b2). The longer and more visible the comet’s tail is, the more DNA damage is present. DNA was stained with YOYO-1 (green; b1–b3). The scale bar equals 10 µm. (C) Feulgen-stained DNA and chromosome aberrations during mitosis, visible under a red spectrum in a fluorescent microscope (c1–c4) [101]. The experimental procedure for Feulgen staining (d1–d4,e1,e2) was as follows: root tips were fixed in cold absolute ethanol and glacial acetic acid (3:1, v/v) for 1 h, washed several times with ethanol, rehydrated, hydrolyzed in 4 M HCl (1.5 h), and stained with Schiff’s reagent (pararosaniline; Sigma-Aldrich, Saint Quentin, France) according to standard methods. After rising in SO2–water (three times) and distilled water, 1.5 mm long apical segments of roots were cut off, placed in a drop of 45% acetic acid, and squashed onto microscope slides. Following freezing using dry ice, coverslips were removed, and the dehydrated slides were mounted in Canada balsam. The top right inserts show mitotic morphology under normal conditions; DNA was also Feulgen-stained and the fluorescent image colors were edited to green for better visibility. The scale bars equal 10 µm. (D) Additional examples of anaphase aberrations (Feulgen staining under visible light). The single black asterisks (*) indicate mitotic chromosome fragments that were lost due to a lack of a connection to kinetochores (d1,d3); the double black asterisks (**) indicate O-shaped chromosomes (d2); the three asterisks (***) indicate a chromosome bridge (d4). The scale bars equal 10 µm. (E) Post-mitotic aberrations of decondensing DNA (Feulgen staining under visible light). Chromosome fragments (lost between opposite poles of the cell), i.e., later micronuclei, are indicated with a red asterisk (e1,e2). The scale bars equal 10 µm. (F) Short schematic of an S-phase checkpoint malfunction caused by caffeine (CF), as used in the presented experiments.

Figure 6

A summary of (A) the effects of HU exposure; (B) the consequences of high dosage, long treatment, and checkpoint deregulation; and (C) the medical and scientific applications of hydroxyurea.