G Model ANTAGE-4558; No. of Pages 6
ARTICLE IN PRESS International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts Joachim Müller a , Vreni Balmer a , Pablo Winzer a , Mahbubur Rahman a , Vera Manser a , Richard K. Haynes b , Andrew Hemphill a,∗ a
Institute of Parasitology, Vetsuisse Faculty, University of Berne, Länggass-Strasse 122, CH-3012 Berne, Switzerland Centre of Excellence for Pharmaceutical Sciences, Faculty of Health Sciences G2, North-West University Potchefstroom Campus, 11 Hoffman Street, Potchefstroom 2531, South Africa b
a r t i c l e
i n f o
Article history: Received 18 December 2014 Accepted 25 February 2015 Keywords: Neospora caninum Trioxolane Artemisinin Artemisone Chemotherapy Resistance
a b s t r a c t From a panel of 34 artemisinin derivatives tested in vitro, artemisone, GC007 and GC012 were most efficacious at inhibiting Neospora caninum replication (IC50 values of 3–54 nM), did not notably impair the invasiveness of tachyzoites and were non-toxic for human foreskin fibroblasts (HFFs). Transmission electron microscopy of drug-treated N. caninum-infected HFFs demonstrated severe alterations in the parasite cytoplasm, changes in the composition of the matrix of the parasitophorous vacuole (PV) and diminished integrity of the PV membrane. To exert parasiticidal activity, parasites had to be cultured continuously in the presence of 5 M artemisone or GC007 for 3 weeks. N. caninum tachyzoites readily adapted to a stepwise increase in concentrations (0.5–10 M) of GC012, but not to artemisone or GC007. Drugs induced the expression of elevated levels of NcBAG1 and NcSAG4 mRNA, but only NcBAG1 could be detected by immunofluorescence. Thus, artemisinin derivatives represent interesting leads that should be investigated further. © 2015 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction Neospora caninum is an apicomplexan parasite that invades and proliferates within a wide range of cell types. In an immunocompetent host, N. caninum differentiates into bradyzoites, which form tissue cysts and establish chronic infection [1]. Since its discovery in 1984, N. caninum has emerged as one of the most important infectious causes of bovine abortion, with an economic impact of over US$1.3 billion [2]. There is currently no vaccine available, thus drugs could be a promising option if effective compounds can be identified. Artemisinin is a naturally occurring 1,2,4-trioxane extracted from the sweet wormwood plant Artemisia annua. Semisynthetic artemisinin derivatives such as artemether, artesunate and dihydroartemisinin are widely used as antimalarials [3]. Newer 10-alkylamino derivatives such as artemisone (see Fig. 1) and artemiside exhibit superior antimalarial activities [4] and good efficacy in the treatment of experimental toxoplasmosis in mice [5]. Their mode of action has been the subject of considerable
∗ Corresponding author. Tel.: +41 31 631 2384; fax: +41 31 631 2477. E-mail address: [email protected] (A. Hemphill).
debate [4]. The peroxide group within the trioxane moiety is essential for biological activity. For blocking proliferation of Toxoplasma gondii, the nature of the substitution at C-10 is important [6]. Various compounds against N. caninum have been investigated, but only a few were evaluated in small animal models, including sulfadiazine, amprolium [7], toltrazuril [7], miltefosine [8] and two pentamidines [9,10]. Low micromolar concentrations of artemisone inhibited Neospora development in vitro, and in an in vivo gerbil model a non-toxic dose of artemisone prevented cerebral symptoms [11]. Here we investigated the in vitro activities of three artemisinin derivatives and 31 fully synthetic trioxolanes against N. caninum tachyzoites. In vitro data from the three most effective compounds, namely the artemisinin derivatives artemisone, GC007 and GC012, are presented.
2. Materials and methods 2.1. Compounds, media and biochemicals Drugs (Fig. 1) were synthesised at the Hong Kong University of Science and Technology (Hong Kong, People’s Republic of China), and stock solutions of 50 mM in dimethyl sulphoxide (DMSO) were stored at −20 ◦ C. Culture media were purchased from Gibco-BRL
http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020 0924-8579/© 2015 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
Please cite this article in press as: Müller J, et al. In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts. Int J Antimicrob Agents (2015), http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020
G Model ANTAGE-4558; No. of Pages 6
ARTICLE IN PRESS J. Müller et al. / International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
2
Fig. 1. Effects of artemisone, GC007 and GC012 on intracellular proliferation and host cell invasion of Neospora caninum tachyzoites. (A) Human foreskin fibroblasts (HFFs) were grown to confluence in a 96-well plate, treated with artemisone, GC007 or GC012 at various concentrations or with dimethyl sulphoxide (DMSO) as a solvent control, and were infected with Nc--gal tachyzoites (103 per well). After 3 days, -galactosidase activity was determined and is shown here as a percentage of the solvent control. Mean ± standard error (SE) values correspond to four replicates. IC50 values (inhibitory concentration of 50%) were calculated as described and are shown beside the compound structures (with 95% confidence intervals). (B) Six-well tissue culture plates were seeded with 5 × 104 HFFs/well and were grown to confluence. Isolated Nc-Liv tachyzoites were pre-treated with 5 M of the drugs at 37 ◦ C for 1 h or 3 h, respectively, and were used for infection of HFF monolayers at 2 × 105 parasites/well. Control wells were infected with parasites incubated in the appropriate amount of DMSO. After 2 h, the medium was removed and the wells were washed with 4 mL of phosphate-buffered saline (PBS). Subsequently, cells were harvested and intracellular tachyzoites were quantified by real-time PCR. Mean ± SE values correspond to three replicates.
(Zürich, Switzerland) and biochemical reagents were from Sigma (St Louis, MO). 2.2. Host cells and parasites Human foreskin fibroblasts (HFFs) and Vero cells were cultured as previously described [12]. -Galactosidase-expressing N. caninum (Nc--gal) tachyzoites were maintained by serial passage in Vero cells in the same medium and were used for screening purposes. The N. caninum Nc-Liverpool (Nc-Liv) isolate was cultured identically and was used for all other assays. 2.3. Treatment of human foreskin fibroblasts and parasites Viability assessments of HFFs and drug efficacy tests using Nc--gal were performed as previously reported [9]. IC50 values (inhibitory concentration of 50%) were calculated as reported [12]. The effects of drugs on extracellular Nc-Liv tachyzoites were assessed by seeding six-well plates (Sarstedt, Sevelen, Switzerland) with 5 × 104 HFFs/well and infecting monolayers with tachyzoites at 2 × 105 parasites/well after pre-treatment with 5 M of each drug at 37 ◦ C for 1 h or 3 h. After 2 h, the medium was removed, the wells were washed with 4 mL of phosphate-buffered saline (PBS) and cells were removed with a cell scraper and processed for quantitative real-time PCR [9]. Treatment-and-re-growth experiments
were carried out in T25 tissue culture flasks (Sarstedt) containing HFFs infected with 8 × 105 Nc-Liv. After 2 h, either 500 nM or 5 M of artemisone, GC007 or GC012 were added. On Days 3, 6, 14 and 21 post-infection, drug-containing medium were removed and was substituted by normal medium, after which the flasks were microscopically checked daily for signs of re-emerging parasite replication. The experiment was terminated as soon as plaques appeared. To assess adaptation to increased drug concentrations, N. caninum-infected HFFs in T25 tissue culture flasks were initially cultured in the presence of 500 nM of each compound and drug levels were increased stepwise (or not), typically every 3–4 days, during an overall experimental period of 40 days. The decision to increase the drug concentration was taken after light microscopy inspection of cultures. Every 6–9 days, parasites were passaged onto fresh HFFs. The experiment was terminated at Day 40. 2.4. Transmission electron microscopy (TEM) HFFs were grown to confluence in T25 tissue culture flasks and were infected with Nc-Liv tachyzoites. After 2 days, cultures were treated with 5 M artemisone, GC007 or GC012 for 24 h and 72 h. Monolayers were washed with 100 mM sodium cacodylate buffer (pH 7.3) and were fixed in 2.5% glutaraldehyde in cacodylate buffer. Post-fixation in 2% OsO4 , dehydration, embedding in EPON 820 epoxy resin and cutting of ultrathin sections was done as described
Please cite this article in press as: Müller J, et al. In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts. Int J Antimicrob Agents (2015), http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020
G Model ANTAGE-4558; No. of Pages 6
ARTICLE IN PRESS J. Müller et al. / International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
3
Fig. 2. Ultrastructural effects of drug treatment on Neospora caninum tachyzoites. Nc-Liv-infected human foreskin fibroblast (HFF) monolayers were cultured for 2 days and were then treated with 5 M artemisone, GC007 or GC012 for 24 h and 72 h and then fixed and processed for transmission electron microscopy. (A) Controls cultured in the presence of solvent only, showing tachyzoites residing within a parasitophorous vacuole delineated by a parasitophorous vacuole membrane. The parasitophorous vacuole matrix is composed of a tubular network (PVtn marked with black arrows). nu, tachyzoite nucleus; rho, rhoptries; dg, dense granules. Bar = 0.75 m. (B–D) Neospora caninum-infected HFFs treated with 5 M GC007 for 24 h. (C and D) Higher magnification views of (B). Note the electron-dense and rather granular appearance of the matrix of the parasitophorous vacuole (PVma) and the tachyzoites as well as other undefined inclusions embedded in the matrix. White arrows point towards the still intact tachyzoite plasma membrane and open triangles indicate areas where the membrane of the parasitophorous vacuole is missing. con, conoid. Bars in (B) = 1.6 m; and (C,D) = 0.4 m. (E) Largely still intact tachyzoites are seen after 24 h of artemisone treatment, with well-defined rhoptries (rho), dense granules (dg) and micronemes (mic) and conoid (con) at the apical tip, although in some tachyzoites the structure of the nuclear interior appears dissolved (nuc). In addition, parts of the membrane of the parasitophorous vacuole are still well defined (black triangles), whilst in other locations this membrane appears dissolved (open triangles). Note the irregular structures (membranes, undefined electron-dense material and vacuoles) within the matrix of the parasitophorous vacuole (PVma). Bar = 0.4 m. (F) After 72 h of artemisone exposure, most vacuoles contain electron-dense matrix components with embedded, and largely non-viable, tachyzoites with a distorted cytoplasm and nucleus with condensed chromatin (nuc). Bar = 0.3 m.
previously [9]. Specimens were viewed on a Philips 400 transmission electron microscope (Philips Electron Optics, Eindhoven, The Netherlands) operating at 80 kV. 2.5. Measurement of NcBAG1 and NcSAG4 RNA expression Confluent HFF monolayers in six-well plates were infected with 2.5 × 104 N. caninum tachyzoites and, after 2 days of treatment with 5 M artemisone, GC007, GC012 or DMSO alone,
were initiated (=Day 0). After 0, 1, 2 and 4 days, monolayers were washed once in PBS and the adherent cells were scraped off, RNA was extracted and quantitative reverse transcriptase PCR (RT-PCR) was performed [12] using the following primers: for NcBAG1 (AB475042.1), NcBAG1 F (CTCGACTTCATGGATGAGG) and Nc BAG1 R (CTTCTATGGTAACGTCATCC); and for NcSAG4 (AY763105.1), NcSAG4 F (CGCGTGCTCGGCGTGG) and NcSAG4 R (GTGCTACCAGTTTGCATCC). From the quantitative RT-PCR, mean values (±standard error) from four replicates were assessed and
Please cite this article in press as: Müller J, et al. In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts. Int J Antimicrob Agents (2015), http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020
G Model ANTAGE-4558; No. of Pages 6
ARTICLE IN PRESS J. Müller et al. / International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
4
expression levels were given as values in arbitrary units relative to the amount of 28S rRNA (L49389.1) quantified using the primers Nc28S F (TCTCTCTCACCAGGTTTAGG) and Nc28S R (CCGTGTTTCAAGACGGGTC). 2.6. Immunofluorescence HFF monolayers grown in 24-well plates (Sarstedt) on poly-l-lysine-coated coverslips were infected with 2 × 104 N. caninum tachyzoites. After 24 h, specimens were exposed to 5 M artemisone, GC007 or GC012 for a period of 4 days. Specimens were then washed once in PBS and fixed and processed for immunofluorescence as described previously [13]. 3. Results 3.1. Artemisinin derivatives restrict the proliferation of Neospora caninum tachyzoites in vitro Thirty-one trioxolanes and three artemisinin derivatives were evaluated for their capacity to interfere with N. caninum proliferation. Artemisone, GC007 and GC012 were the only compounds with an IC50 value of 50 M, and for non-confluent
monolayers IC50 values of GC007 and GC012 were ca. 14 M. None of the compounds exhibited a dramatic effect on the infectivity of extracellular tachyzoites (Fig. 1B). As shown previously [14], N. caninum tachyzoites exhibited a considerably reduced invasion capability already after 3 h of extracellular maintenance even in the absence of drugs.
3.2. Long-term treatment of Neospora caninum-infected human foreskin fibroblasts The putative parasiticidal activity of the three compounds was evaluated. Continuous treatment with 500 nM of all compounds failed to eliminate tachyzoites, and even after 21 days parasites resumed proliferation after 4–6 days of maintenance in the absence of drugs. When treatment with 5 M was stopped between 3 days and 14 days of drug exposure, tachyzoites reappeared between 6 days and 8 days following treatment irrespective of the treatment duration. Only after 21 days of continuous treatment with 5 M of each of artemisone and GC007 were all tachyzoites eliminated. Neospora could readily adapt to the presence of 0.5 M GC012 as the drug concentration could be increased stepwise up to 10 M within 20 days. In the presence of artemisone and GC007, however, parasites could not adapt to the drug pressure and proliferation remained continuously suppressed.
Fig. 3. Bradyzoite gene expression increases during treatment with artemisone, GC007 and GC012. (A) Immunofluorescent double-labelling with anti-NcSAG1 (tachyzoites, red) and anti-NcBAG1 (bradyzoites, green). Staining was performed with control cultures or with cultures treated with 5 M of each of artemisone, GC007 or GC012 for 4 days. (B) NcBAG1 and NcSAG4 mRNA levels. Confluent human foreskin fibroblast (HFF) monolayers grown in six-well plates were infected with 2.5 × 104 Neospora caninum tachyzoites. Treatment with each of 5 M of artemisone, GC007 or GC012 or with dimethyl sulphoxide (DMSO) (control) were initiated 2 days post-infection (=Day 0). After different time points, adherent cells were harvested for RNA extraction. mRNA levels were quantified by real-time reverse transcriptase PCR (RT-PCR). Mean ± standard error values from four replicates were assessed and expression levels were given as values in arbitrary units relative to the amount of 28S rRNA.
Please cite this article in press as: Müller J, et al. In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts. Int J Antimicrob Agents (2015), http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020
G Model ANTAGE-4558; No. of Pages 6
ARTICLE IN PRESS J. Müller et al. / International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
3.3. Treatment with artemisone, GC007 or GC012 induces distinct ultrastructural alterations TEM revealed distinct structural and morphological alterations induced by these compounds. Non-treated controls (Fig. 2A) showed intracellular parasites residing in the cytoplasm of their host cells within a parasitophorous vacuole (PV) surrounded by a PV membrane. A tubular network forming a vacuole matrix was evident. The most rapid alterations were observed upon treatment with GC007, which had a noteworthy impact already at 24 h, but more extensive damage was evident at 72 h of treatment (Fig. 2B–D). Many tachyzoites were still located intracellularly, although the PV membrane was no longer readily discernible and the PV matrix was composed of highly electron-dense material of unknown nature intermingled with small vacuoles. Most tachyzoites exhibited a distorted cytoplasm with vacuoles, lipid droplets and condensed nuclei, but the plasma membrane appeared mostly unaffected. Exposure of N. caninum tachyzoites to artemisone led to similar changes, but the impact was less rapid as after 24 h only modest ultrastructural alterations were visible (Fig. 2E). However, 72 h of treatment with artemisone induced the formation of PVs that were filled with electron-dense matrix and tachyzoites exhibiting similar characteristics as described for GC007 (Fig. 2F). Changes in parasites induced by treatment with GC012 were less dramatic. After 24 h and 72 h of treatment, tachyzoites appeared rather undisturbed, except that the PV membrane and the vacuolar matrix were absent or degraded in some instances (data not shown). 3.4. Drug treatment induces differential expression of BAG1 and SAG4 genes The fact that a small fraction of N. caninum tachyzoites appeared ‘resistant’ to these compounds for extended periods of time suggested that these parasites adapted to the adverse conditions by differentiation to bradyzoite-like forms. Upon treatment of infected HFFs with the compounds, specific expression of the NcBAG1 protein could be detected (Fig. 3). Moreover, mRNA levels of NcBAG1 were increased by a factor of >40 in cells treated with artemisone or GC007, and NcSAG4 levels were increased by a factor of >50 in cells treated with GC007 and GC012 (Fig. 3). 4. Discussion Artemisone is a 10-alkylaminoartemisinin that was developed as a potent antimalarial drug, whilst artemiside is a more lipophilic compound. Both artemisone and especially artemiside showed promising efficacy in mice experimentally infected with T. gondii [15]. In this study, we have assessed the ability of a small panel of artemisinins to act against N. caninum in vitro, and only artemisone, GC007 and GC012 exhibited IC50 values of
ARTICLE IN PRESS International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts Joachim Müller a , Vreni Balmer a , Pablo Winzer a , Mahbubur Rahman a , Vera Manser a , Richard K. Haynes b , Andrew Hemphill a,∗ a
Institute of Parasitology, Vetsuisse Faculty, University of Berne, Länggass-Strasse 122, CH-3012 Berne, Switzerland Centre of Excellence for Pharmaceutical Sciences, Faculty of Health Sciences G2, North-West University Potchefstroom Campus, 11 Hoffman Street, Potchefstroom 2531, South Africa b
a r t i c l e
i n f o
Article history: Received 18 December 2014 Accepted 25 February 2015 Keywords: Neospora caninum Trioxolane Artemisinin Artemisone Chemotherapy Resistance
a b s t r a c t From a panel of 34 artemisinin derivatives tested in vitro, artemisone, GC007 and GC012 were most efficacious at inhibiting Neospora caninum replication (IC50 values of 3–54 nM), did not notably impair the invasiveness of tachyzoites and were non-toxic for human foreskin fibroblasts (HFFs). Transmission electron microscopy of drug-treated N. caninum-infected HFFs demonstrated severe alterations in the parasite cytoplasm, changes in the composition of the matrix of the parasitophorous vacuole (PV) and diminished integrity of the PV membrane. To exert parasiticidal activity, parasites had to be cultured continuously in the presence of 5 M artemisone or GC007 for 3 weeks. N. caninum tachyzoites readily adapted to a stepwise increase in concentrations (0.5–10 M) of GC012, but not to artemisone or GC007. Drugs induced the expression of elevated levels of NcBAG1 and NcSAG4 mRNA, but only NcBAG1 could be detected by immunofluorescence. Thus, artemisinin derivatives represent interesting leads that should be investigated further. © 2015 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction Neospora caninum is an apicomplexan parasite that invades and proliferates within a wide range of cell types. In an immunocompetent host, N. caninum differentiates into bradyzoites, which form tissue cysts and establish chronic infection [1]. Since its discovery in 1984, N. caninum has emerged as one of the most important infectious causes of bovine abortion, with an economic impact of over US$1.3 billion [2]. There is currently no vaccine available, thus drugs could be a promising option if effective compounds can be identified. Artemisinin is a naturally occurring 1,2,4-trioxane extracted from the sweet wormwood plant Artemisia annua. Semisynthetic artemisinin derivatives such as artemether, artesunate and dihydroartemisinin are widely used as antimalarials [3]. Newer 10-alkylamino derivatives such as artemisone (see Fig. 1) and artemiside exhibit superior antimalarial activities [4] and good efficacy in the treatment of experimental toxoplasmosis in mice [5]. Their mode of action has been the subject of considerable
∗ Corresponding author. Tel.: +41 31 631 2384; fax: +41 31 631 2477. E-mail address: [email protected] (A. Hemphill).
debate [4]. The peroxide group within the trioxane moiety is essential for biological activity. For blocking proliferation of Toxoplasma gondii, the nature of the substitution at C-10 is important [6]. Various compounds against N. caninum have been investigated, but only a few were evaluated in small animal models, including sulfadiazine, amprolium [7], toltrazuril [7], miltefosine [8] and two pentamidines [9,10]. Low micromolar concentrations of artemisone inhibited Neospora development in vitro, and in an in vivo gerbil model a non-toxic dose of artemisone prevented cerebral symptoms [11]. Here we investigated the in vitro activities of three artemisinin derivatives and 31 fully synthetic trioxolanes against N. caninum tachyzoites. In vitro data from the three most effective compounds, namely the artemisinin derivatives artemisone, GC007 and GC012, are presented.
2. Materials and methods 2.1. Compounds, media and biochemicals Drugs (Fig. 1) were synthesised at the Hong Kong University of Science and Technology (Hong Kong, People’s Republic of China), and stock solutions of 50 mM in dimethyl sulphoxide (DMSO) were stored at −20 ◦ C. Culture media were purchased from Gibco-BRL
http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020 0924-8579/© 2015 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
Please cite this article in press as: Müller J, et al. In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts. Int J Antimicrob Agents (2015), http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020
G Model ANTAGE-4558; No. of Pages 6
ARTICLE IN PRESS J. Müller et al. / International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
2
Fig. 1. Effects of artemisone, GC007 and GC012 on intracellular proliferation and host cell invasion of Neospora caninum tachyzoites. (A) Human foreskin fibroblasts (HFFs) were grown to confluence in a 96-well plate, treated with artemisone, GC007 or GC012 at various concentrations or with dimethyl sulphoxide (DMSO) as a solvent control, and were infected with Nc--gal tachyzoites (103 per well). After 3 days, -galactosidase activity was determined and is shown here as a percentage of the solvent control. Mean ± standard error (SE) values correspond to four replicates. IC50 values (inhibitory concentration of 50%) were calculated as described and are shown beside the compound structures (with 95% confidence intervals). (B) Six-well tissue culture plates were seeded with 5 × 104 HFFs/well and were grown to confluence. Isolated Nc-Liv tachyzoites were pre-treated with 5 M of the drugs at 37 ◦ C for 1 h or 3 h, respectively, and were used for infection of HFF monolayers at 2 × 105 parasites/well. Control wells were infected with parasites incubated in the appropriate amount of DMSO. After 2 h, the medium was removed and the wells were washed with 4 mL of phosphate-buffered saline (PBS). Subsequently, cells were harvested and intracellular tachyzoites were quantified by real-time PCR. Mean ± SE values correspond to three replicates.
(Zürich, Switzerland) and biochemical reagents were from Sigma (St Louis, MO). 2.2. Host cells and parasites Human foreskin fibroblasts (HFFs) and Vero cells were cultured as previously described [12]. -Galactosidase-expressing N. caninum (Nc--gal) tachyzoites were maintained by serial passage in Vero cells in the same medium and were used for screening purposes. The N. caninum Nc-Liverpool (Nc-Liv) isolate was cultured identically and was used for all other assays. 2.3. Treatment of human foreskin fibroblasts and parasites Viability assessments of HFFs and drug efficacy tests using Nc--gal were performed as previously reported [9]. IC50 values (inhibitory concentration of 50%) were calculated as reported [12]. The effects of drugs on extracellular Nc-Liv tachyzoites were assessed by seeding six-well plates (Sarstedt, Sevelen, Switzerland) with 5 × 104 HFFs/well and infecting monolayers with tachyzoites at 2 × 105 parasites/well after pre-treatment with 5 M of each drug at 37 ◦ C for 1 h or 3 h. After 2 h, the medium was removed, the wells were washed with 4 mL of phosphate-buffered saline (PBS) and cells were removed with a cell scraper and processed for quantitative real-time PCR [9]. Treatment-and-re-growth experiments
were carried out in T25 tissue culture flasks (Sarstedt) containing HFFs infected with 8 × 105 Nc-Liv. After 2 h, either 500 nM or 5 M of artemisone, GC007 or GC012 were added. On Days 3, 6, 14 and 21 post-infection, drug-containing medium were removed and was substituted by normal medium, after which the flasks were microscopically checked daily for signs of re-emerging parasite replication. The experiment was terminated as soon as plaques appeared. To assess adaptation to increased drug concentrations, N. caninum-infected HFFs in T25 tissue culture flasks were initially cultured in the presence of 500 nM of each compound and drug levels were increased stepwise (or not), typically every 3–4 days, during an overall experimental period of 40 days. The decision to increase the drug concentration was taken after light microscopy inspection of cultures. Every 6–9 days, parasites were passaged onto fresh HFFs. The experiment was terminated at Day 40. 2.4. Transmission electron microscopy (TEM) HFFs were grown to confluence in T25 tissue culture flasks and were infected with Nc-Liv tachyzoites. After 2 days, cultures were treated with 5 M artemisone, GC007 or GC012 for 24 h and 72 h. Monolayers were washed with 100 mM sodium cacodylate buffer (pH 7.3) and were fixed in 2.5% glutaraldehyde in cacodylate buffer. Post-fixation in 2% OsO4 , dehydration, embedding in EPON 820 epoxy resin and cutting of ultrathin sections was done as described
Please cite this article in press as: Müller J, et al. In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts. Int J Antimicrob Agents (2015), http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020
G Model ANTAGE-4558; No. of Pages 6
ARTICLE IN PRESS J. Müller et al. / International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
3
Fig. 2. Ultrastructural effects of drug treatment on Neospora caninum tachyzoites. Nc-Liv-infected human foreskin fibroblast (HFF) monolayers were cultured for 2 days and were then treated with 5 M artemisone, GC007 or GC012 for 24 h and 72 h and then fixed and processed for transmission electron microscopy. (A) Controls cultured in the presence of solvent only, showing tachyzoites residing within a parasitophorous vacuole delineated by a parasitophorous vacuole membrane. The parasitophorous vacuole matrix is composed of a tubular network (PVtn marked with black arrows). nu, tachyzoite nucleus; rho, rhoptries; dg, dense granules. Bar = 0.75 m. (B–D) Neospora caninum-infected HFFs treated with 5 M GC007 for 24 h. (C and D) Higher magnification views of (B). Note the electron-dense and rather granular appearance of the matrix of the parasitophorous vacuole (PVma) and the tachyzoites as well as other undefined inclusions embedded in the matrix. White arrows point towards the still intact tachyzoite plasma membrane and open triangles indicate areas where the membrane of the parasitophorous vacuole is missing. con, conoid. Bars in (B) = 1.6 m; and (C,D) = 0.4 m. (E) Largely still intact tachyzoites are seen after 24 h of artemisone treatment, with well-defined rhoptries (rho), dense granules (dg) and micronemes (mic) and conoid (con) at the apical tip, although in some tachyzoites the structure of the nuclear interior appears dissolved (nuc). In addition, parts of the membrane of the parasitophorous vacuole are still well defined (black triangles), whilst in other locations this membrane appears dissolved (open triangles). Note the irregular structures (membranes, undefined electron-dense material and vacuoles) within the matrix of the parasitophorous vacuole (PVma). Bar = 0.4 m. (F) After 72 h of artemisone exposure, most vacuoles contain electron-dense matrix components with embedded, and largely non-viable, tachyzoites with a distorted cytoplasm and nucleus with condensed chromatin (nuc). Bar = 0.3 m.
previously [9]. Specimens were viewed on a Philips 400 transmission electron microscope (Philips Electron Optics, Eindhoven, The Netherlands) operating at 80 kV. 2.5. Measurement of NcBAG1 and NcSAG4 RNA expression Confluent HFF monolayers in six-well plates were infected with 2.5 × 104 N. caninum tachyzoites and, after 2 days of treatment with 5 M artemisone, GC007, GC012 or DMSO alone,
were initiated (=Day 0). After 0, 1, 2 and 4 days, monolayers were washed once in PBS and the adherent cells were scraped off, RNA was extracted and quantitative reverse transcriptase PCR (RT-PCR) was performed [12] using the following primers: for NcBAG1 (AB475042.1), NcBAG1 F (CTCGACTTCATGGATGAGG) and Nc BAG1 R (CTTCTATGGTAACGTCATCC); and for NcSAG4 (AY763105.1), NcSAG4 F (CGCGTGCTCGGCGTGG) and NcSAG4 R (GTGCTACCAGTTTGCATCC). From the quantitative RT-PCR, mean values (±standard error) from four replicates were assessed and
Please cite this article in press as: Müller J, et al. In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts. Int J Antimicrob Agents (2015), http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020
G Model ANTAGE-4558; No. of Pages 6
ARTICLE IN PRESS J. Müller et al. / International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
4
expression levels were given as values in arbitrary units relative to the amount of 28S rRNA (L49389.1) quantified using the primers Nc28S F (TCTCTCTCACCAGGTTTAGG) and Nc28S R (CCGTGTTTCAAGACGGGTC). 2.6. Immunofluorescence HFF monolayers grown in 24-well plates (Sarstedt) on poly-l-lysine-coated coverslips were infected with 2 × 104 N. caninum tachyzoites. After 24 h, specimens were exposed to 5 M artemisone, GC007 or GC012 for a period of 4 days. Specimens were then washed once in PBS and fixed and processed for immunofluorescence as described previously [13]. 3. Results 3.1. Artemisinin derivatives restrict the proliferation of Neospora caninum tachyzoites in vitro Thirty-one trioxolanes and three artemisinin derivatives were evaluated for their capacity to interfere with N. caninum proliferation. Artemisone, GC007 and GC012 were the only compounds with an IC50 value of 50 M, and for non-confluent
monolayers IC50 values of GC007 and GC012 were ca. 14 M. None of the compounds exhibited a dramatic effect on the infectivity of extracellular tachyzoites (Fig. 1B). As shown previously [14], N. caninum tachyzoites exhibited a considerably reduced invasion capability already after 3 h of extracellular maintenance even in the absence of drugs.
3.2. Long-term treatment of Neospora caninum-infected human foreskin fibroblasts The putative parasiticidal activity of the three compounds was evaluated. Continuous treatment with 500 nM of all compounds failed to eliminate tachyzoites, and even after 21 days parasites resumed proliferation after 4–6 days of maintenance in the absence of drugs. When treatment with 5 M was stopped between 3 days and 14 days of drug exposure, tachyzoites reappeared between 6 days and 8 days following treatment irrespective of the treatment duration. Only after 21 days of continuous treatment with 5 M of each of artemisone and GC007 were all tachyzoites eliminated. Neospora could readily adapt to the presence of 0.5 M GC012 as the drug concentration could be increased stepwise up to 10 M within 20 days. In the presence of artemisone and GC007, however, parasites could not adapt to the drug pressure and proliferation remained continuously suppressed.
Fig. 3. Bradyzoite gene expression increases during treatment with artemisone, GC007 and GC012. (A) Immunofluorescent double-labelling with anti-NcSAG1 (tachyzoites, red) and anti-NcBAG1 (bradyzoites, green). Staining was performed with control cultures or with cultures treated with 5 M of each of artemisone, GC007 or GC012 for 4 days. (B) NcBAG1 and NcSAG4 mRNA levels. Confluent human foreskin fibroblast (HFF) monolayers grown in six-well plates were infected with 2.5 × 104 Neospora caninum tachyzoites. Treatment with each of 5 M of artemisone, GC007 or GC012 or with dimethyl sulphoxide (DMSO) (control) were initiated 2 days post-infection (=Day 0). After different time points, adherent cells were harvested for RNA extraction. mRNA levels were quantified by real-time reverse transcriptase PCR (RT-PCR). Mean ± standard error values from four replicates were assessed and expression levels were given as values in arbitrary units relative to the amount of 28S rRNA.
Please cite this article in press as: Müller J, et al. In vitro effects of new artemisinin derivatives in Neospora caninum-infected human fibroblasts. Int J Antimicrob Agents (2015), http://dx.doi.org/10.1016/j.ijantimicag.2015.02.020
G Model ANTAGE-4558; No. of Pages 6
ARTICLE IN PRESS J. Müller et al. / International Journal of Antimicrobial Agents xxx (2015) xxx–xxx
3.3. Treatment with artemisone, GC007 or GC012 induces distinct ultrastructural alterations TEM revealed distinct structural and morphological alterations induced by these compounds. Non-treated controls (Fig. 2A) showed intracellular parasites residing in the cytoplasm of their host cells within a parasitophorous vacuole (PV) surrounded by a PV membrane. A tubular network forming a vacuole matrix was evident. The most rapid alterations were observed upon treatment with GC007, which had a noteworthy impact already at 24 h, but more extensive damage was evident at 72 h of treatment (Fig. 2B–D). Many tachyzoites were still located intracellularly, although the PV membrane was no longer readily discernible and the PV matrix was composed of highly electron-dense material of unknown nature intermingled with small vacuoles. Most tachyzoites exhibited a distorted cytoplasm with vacuoles, lipid droplets and condensed nuclei, but the plasma membrane appeared mostly unaffected. Exposure of N. caninum tachyzoites to artemisone led to similar changes, but the impact was less rapid as after 24 h only modest ultrastructural alterations were visible (Fig. 2E). However, 72 h of treatment with artemisone induced the formation of PVs that were filled with electron-dense matrix and tachyzoites exhibiting similar characteristics as described for GC007 (Fig. 2F). Changes in parasites induced by treatment with GC012 were less dramatic. After 24 h and 72 h of treatment, tachyzoites appeared rather undisturbed, except that the PV membrane and the vacuolar matrix were absent or degraded in some instances (data not shown). 3.4. Drug treatment induces differential expression of BAG1 and SAG4 genes The fact that a small fraction of N. caninum tachyzoites appeared ‘resistant’ to these compounds for extended periods of time suggested that these parasites adapted to the adverse conditions by differentiation to bradyzoite-like forms. Upon treatment of infected HFFs with the compounds, specific expression of the NcBAG1 protein could be detected (Fig. 3). Moreover, mRNA levels of NcBAG1 were increased by a factor of >40 in cells treated with artemisone or GC007, and NcSAG4 levels were increased by a factor of >50 in cells treated with GC007 and GC012 (Fig. 3). 4. Discussion Artemisone is a 10-alkylaminoartemisinin that was developed as a potent antimalarial drug, whilst artemiside is a more lipophilic compound. Both artemisone and especially artemiside showed promising efficacy in mice experimentally infected with T. gondii [15]. In this study, we have assessed the ability of a small panel of artemisinins to act against N. caninum in vitro, and only artemisone, GC007 and GC012 exhibited IC50 values of
