Tt

La

b

a

ARRAA

KNCTCL

1

ctthfsoac

tr

Tdf

h0

Molecular & Biochemical Parasitology 196 (2014) 75–81

Contents lists available at ScienceDirect

Molecular & Biochemical Parasitology

he chloramphenicol acetyltransferase vector as a tool for stableagging of Neospora caninum

uiz Miguel Pereiraa,b, Ana Patrícia Yatsudaa,b,∗

Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Av do Café, sn/n, 14040-903 Ribeirão Preto, SP, BrazilNúcleo de Apoio à Pesquisa em Produtos Naturais e Sintéticos, Universidade de São Paulo, Ribeirão Preto, SP, Brazil

r t i c l e i n f o

rticle history:eceived 8 April 2014eceived in revised form 3 July 2014ccepted 4 August 2014vailable online 12 August 2014

eywords:eospora caninumATransfectionhloramphenicol acetyltransferase

a b s t r a c t

Neospora caninum is an obligate intracellular Apicomplexa, a phylum where one of the current methodsfor functional studies relies on molecular genetic tools. For Toxoplasma gondii, the first method described,in 1993, was based on resistance against chloramphenicol. As in T. gondii, we developed a vector con-stituted of the chloramphenicol acetyltransferase gene (CAT) flanked by the N. caninum dihydrofolatereductase-thymidylate synthase (DHFR-TS) 5′ coding sequence flanking region. Five weeks after trans-fection and under the selection of chloramphenicol the expression of CAT increased compared to thewild type and the resistance was retained for more than one year. Between the stop codon of CAT andthe 3′ UTR of DHFR, a Lac-Z gene controlled by the N. caninum tubulin 5′ coding sequence flanking regionwas ligated, resulting in a vector with a reporter gene (Ncdhfr-CAT/NcTub-tetO/Lac-Z). The stability wasmaintained through an episomal pattern for 14 months when the tachyzoites succumbed, which was

ac-Z gene an unexpected phenomenon compared to T. gondii. Stable parasites expressing the Lac-Z gene allowedthe detection of tachyzoites after invasion by enzymatic reaction (CPRG) and were visualised macro- andmicroscopically by X-Gal precipitation and fluorescence. This work developed the first vector for stableexpression of proteins based on chloramphenicol resistance and controlled exclusively by N. caninumpromoters.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

Neospora caninum is an apicomplexan parasite that has beenlosely related to abortion and loss of fertility in cattle [1]. One ofhe key events to deeply investigate the invasion/replication sys-em of the parasite is based on molecular exploitation. Althoughighly developed for Toxoplasma gondii, molecular tools are rare

or N. caninum. The genetic manipulation in N. caninum with exclu-ive promoters was initiated by our group through the mutationf DHFR-TS and insertion of the Lac-Z gene [2]. Another possiblelternative is the development of parasites that are resistant tohloramphenicol.

Chloramphenicol is an antibiotic that inhibits translationhrough the binding to a peptidyltransferase enzyme on the 50Sibosome protein, both in Gram-negative and -positive bacteria

∗ Corresponding author at: Departamento de Análises Clínicas, Bromatológicas eoxicológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidadee São Paulo, 14040-903 Ribeirão Preto, Brazil. Tel.: +55 16 3602 4724;ax: +55 16 3602 4725.

E-mail addresses: ayatsuda@fcfrp.usp.br, ayatsuda@gmail.com (A.P. Yatsuda).

ttp://dx.doi.org/10.1016/j.molbiopara.2014.08.001166-6851/© 2014 Elsevier B.V. All rights reserved.

[3]. It probably acts in a manner similar to that in the apicom-plexan apicoplast, a non-photosynthetic plastid-like organelle [4].In bacteria, resistance against chloramphenicol is achieved bychloramphenicol acetyltransferase (CAT). CAT transfers an acetylgroup of Acetyl-S-CoA to chloramphenicol to yield the initialproduct acetoxychloramphenicol, which is devoid of significantantibiotic activity [5]. T. gondii has sensitivity to chloramphenicol;however, the tachyzoites acquire resistance after transfection withthe gene of resistance, CAT, flanked by a 5′ coding sequence flank-ing region and a 3′ downstream region of an active and non-lethaltachyzoite gene [6].

The development of a gene-controlling system in T. gondii, basedon tetracycline transactivation [7,8], allowed the evaluation ofgenes related to the gliding and invasion processes. For N. can-inum, there is a lack of options for the control of gene expression,particularly those responsive to a drug addiction. In this work,we performed the insertion of Lac-Z and tagged the tachyzoitesthrough chloramphenicol resistance. The tachyzoites were able toreceive and express the resistance gene, and the expression of thereporter gene was responsive to the concentration of the drug

applied. Our findings will contribute to the development of moreelaborate experiments for the genetic manipulation of N. caninum.

76 L.M. Pereira, A.P. Yatsuda / Molecular & Biochemical Parasitology 196 (2014) 75–81

F uct. (A5 3′ dowt o Ncdh

2

2

m(21

2N

dasb

sw(ctwCCwAp

gflbT(ee

2

caKEaeG3z

ig. 1. The Neospora caninum Ncdhfr-CAT and Ncdhfr-CAT/NcTub-tetO/Lac-Z constr′ coding sequence flanking region, chloramphenicol coding sequence (CAT) and

ransient expression of Lac-Z in Neospora caninum, NcTub-tetO/Lac-Z, was ligated t

. Materials and methods

.1. N. caninum culture

Vero cell cultures were maintained in RPMI-1640edium (Sigma) supplemented with 5% foetal calf serum

Gibco/Invitrogen), 50 �g/ml of kanamycin at 37 ◦C and 5% CO2 in T-5 cm2 or 75 cm2 tissue culture flasks. N. caninum tachyzoites of Nc-

isolate were maintained in Vero cell monolayers and purified [9].

.2. Construction of Ncdhrf-CAT andcdhfr-CAT/NcTub-tetO/Lac-Z

The T. gondii construct, TgtubYFP-TetR/sag-CAT [8], wasesigned from the pCAT-GFP, which was originally derived from

pKS backbone plasmid [10]. TgtubYFP-TetR/sag-CAT was succes-ively treated with restriction enzymes and ligated as describedelow.

The vector Ncdhfr-CAT was constituted of the CAT codingequence (663 bp), amplified from the TgtubYFP-TetR/sag-CAT [8]ith forward (CCCAGATCTATGCATGAGAAAAAAATC) and reverse

CCCCATATGTTATGCCCCGCCCTGCCA) primers, controlled by the 5′

oding sequence flanking region of the dihydrofolate reductase-hymidylate synthase (DHFR-TS) gene of N. caninum (Fig. 1A); thisas amplified (1428 bp) using the forward (TTTAAGCTTTGGGCAT-ACTGAGGGACTT) and reverse (TTTCCTAGGCATGTTTCGCTGCA-AACTC) primers and the 3′ UTR sequence of the same gene (886 bp)as amplified using the forward (CCCCTGCAGTGGAAAAATCTGA-ATATATA) and reverse (TCCGCGGCCGCCTTTCTCGCAAGTCTCCTG)rimers (Fig. 1A).

The Lac-Z expression was obtained with ligation of the Lac-Zene downstream of the N. caninum tubulin 5′ coding sequenceanking region. Treatment of Ncdhrf-CAT with Pstl allowed ligationetween the CAT coding sequence and the 3′ UTR region of NcDHFR-S gene, generating the Ncdhfr-CAT/NcTub-tetO/Lac-Z constructFig. 1B). The tetO region is a site with affinity to the TetR proteinxtracted from a T. gondii construct [8], but only the �-galactosidasexpression was used in this manuscript.

.3. Stably transfected tachyzoites and CAT-ELISA

The plasmids (25 �g) were inserted in freshly purified N.aninum tachyzoites (1 × 108) by the enzyme restriction medi-ted integration (REMI) with HindIII, in cytomix buffer (120 mMCl, 0.15 mM CaCl2, 10 mM K2HPO4/KH2PO4, 25 mM Hepes, 2 mMDTA, 5 mM MgCl2, pH 7.6) supplemented with 5 mM glutathione,s described by [11]. Cells were transferred to a 0.4 cm gap cuvette,

lectroporated with 1.8 kV at 25 mFd and 100 � in a BioRadenePulser Xcell and incubated for 18 h at 37 ◦C, 5% CO2, when0 �M of chloramphenicol was added. After the lytic cycle, tachy-oites were purified and cultured with 20 �M of chloramphenicol

) The vector for CAT expression was constructed by successive ligations of NcDHFRnstream region of NcDHFR in the TgtubYFP-TetR/sag-CAT [8]. (B) The vector forfr-CAT after treatment with Pstl.

until the development of resistant tachyzoites in the third week.Two weeks later, stable transfected tachyzoites (1 × 107) werecompared with the wild type strain for the expression of chloram-phenicol acetyltransferase using the CAT-ELISA kit (Roche AppliedScience, USA), following the manufacturer’s instructions. The reac-tion was read at an absorbance of 405 nm (Synergy H1, BioTek,software Gen5 2.01) and absorbance of the Vero cells was used asblank; values were analysed by one-way ANOVA with the softwarePrism 5.1, and post-analysed by Tukey test.

2.4. Selection of Lac-Z tachyzoites under chloramphenicol

Two equal and independent aliquots of purified tachyzoites(3 × 107) were transfected, in separated cuvettes, with 5 �g ofplasmid Ncdhfr-CAT/NcTub-tetO/Lac-Z, previously treated withHindIII, and incubated in 5% CO2 at 37 ◦C for 18 h, at which pointfour concentrations of chloramphenicol (5, 10, 20 and 40 �M)were added. After each lytic cycle, the tachyzoites were purifiedin Sephadex G-25 (PD-10 columns, GE) and 1 × 104 tachyzoiteswere inoculated in a new Vero cell culture until the next lyticcycle. The purified tachyzoites were counted in a haemocytome-ter and 1 × 103, 1 × 104, 1 × 105 and 1 × 106 (only after the thirdcycle) were incubated with 5 mM of CPRG (chlorophenolred-�-d-galactopyranoside, Roche) for 18 h at 37 ◦C. The samples weretransferred to a 96-well plate in duplicate and the absorbance wasmeasured at 570 nm (Synergy H1, BioTek). The absorbance equiv-alent to 1 × 105 tachyzoites with the standard error was calculatedand plotted against each lytic cycle.

2.5. Rescue of plasmid from transfected tachyzoites

Genomic DNA from stable tachyzoites (5 × 107), selected using20 �M of chloramphenicol for 2 months, was extracted (Wizard®

Genomic DNA Purification kit, Promega), treated with Pstl (0.5 U)for 18 h, purified (Illustra GFX PCR DNA and Gel Band Purifica-tion Kit, GE) and eluted in 30 �l of deionised water. A volume of17 �l was submitted to a ligation reaction with 0.4 U Weiss of T4ligase (Fermentas) for 18 h at 16 ◦C. The ligated N. caninum DNAwas electroporated in E. coli Top10 (2.5 kV; 25 �F; 200 �), and theampicillin-resistant colonies were either PstI- or EcoRl-treated forvisualisation of the whole plasmid or for verification of the Ncdhfr-CAT insertion, respectively.

2.6. Semi-quantification of rescued plasmids

Genomic DNA from stable tachyzoites (1 × 107) selected with 5,10, 20 and 40 �M of chloramphenicol from the seventh lytic cycle

were extracted (Wizard® Genomic DNA Purification kit, Promega)and 50 ng (GeneQuant Pro, The GE Healthcare Lifesciences,formerly Amersham Biosciences) was amplified using the PCRprimers Forward Ncdhfr (TTTAAGCTTTGGGCATCACTGAGGGACTT)

Bioch

aSw37rurC

2

a(Ccb3i

2

tseptbagct

2

Noac0s1�PAwtCTTaH

2

p1amLtw

L.M. Pereira, A.P. Yatsuda / Molecular &

nd reverse CAT (CCCCATATGTTATGCCCCGCCCTGCCA) with Hot-tarTaq DNA Polymerase (Qiagen). The touchdown PCR conditionsere: one cycle 94 ◦C, 1 min; 25 cycles 94 ◦C, 30 s; five cycles 60 ◦C,

0 s; five cycles 55 ◦C, 30 s; 15 cycles 50 ◦C, 30 s and 25 cycles2 ◦C, 2 min. The 5′ coding sequence flanking region from theibosomal gene NcRPS13 (NCBI: XM 003883829.1) was amplifiednder the same conditions as a control, using the forward andeverse NcRPS13 primers (CCCAAGCTTCCCAGGTTTGCGTTGT andCGTTCCGAAGCTGTCGAATTCTCTC, respectively).

.7. ˇ-Galactosidase detection by CPRG

The cultures from invasion and proliferation assays (items 2.10nd 2.11) were lysed with �-galactosidase lysis buffer (100 mM 4-2-hydroxyethyl)-1-piperazineethanesulfonic acid, pH 8.0; 1 mMaCl2; 1% Triton X-100, 0.5% SDS; 5 mM DTT) for 1 h at 50 ◦C andentrifuged for 10 min at 10,000 × g. Each lysate (20 �l) was incu-ated with 5 mM CPRG in �-galactosidase lysis buffer for 18 h at7 ◦C in an ELISA plate (Nunc) and the reading was taken at 570 nm

n an ELISA reader (Synergy H1, BioTek).

.8. ˇ-Galactosidase detection by X-Gal precipitation

The N. caninum tachyzoites expressing Lac-Z were detected byhe X-Gal precipitation with the kit �-Gal Staining Set (Roche). Theamples from the culture supernatant, and of invasion and prolif-ration assays (items 2.10 and 2.11) were submitted to a similarrocedure. The samples were fixed with 2% formaldehyde, washedwice with PBS and incubated with the working solution (Ironuffer and X-Gal solution from the kit �-Gal Staining Set) for 2 ht 37 ◦C. The samples were washed twice with PBS, mounted inlycerol 60% and the images captured with the AxioCam MRc 5amera in the microscope Zeiss Axioskop 40 and processed usinghe AxioVison 4.6 software.

.9. ˇ-Galactosidase detection by fluorescence

The supernatant of a 5-day culture of tachyzoites (2 × 107) from. caninum selected by chloramphenicol (5, 10, 20 and 40 �M,f the seventh cycle) was fixed with 2% formaldehyde and lig-ted to a poly-l-lysine coated coverslip (Sigma) for 30 min. Theoverslips were washed twice with PBS and permeabilised with.2% PBS-Triton 100 for 30 min and washed twice with PBS. Theamples were blocked with PBS-BSA (3% BSA, glycine 50 mM) for8 h at 4 ◦C, washed twice with PBS, incubated with mouse anti--galactosidase (Sigma) for 1 h at 37 ◦C and washed twice withBS; this was followed by incubation with conjugated anti-mouselexa-488 (Invitrogen) and washing twice with PBS. The coverslipsere mounted with DAPI (Santa Cruz) and the images were cap-

ured using a Confocal Microscope (Leica TCS SP5 Laser Scanningonfocal Microscope, Leica Microsystems, Heidelberg, Germany).he images were captured by a 60× objective in oil immersion.hree images of 0.6 �m from each channel were captured, groupednd processed by the program Image J 1.41 (National Institutes ofealth, USA).

.10. Invasion assay

The invasion of tachyzoites was measured by CPRG and X-Galrecipitation. Purified tachyzoites were diluted (1 × 106; 5 × 105;

× 105, 5 × 104 and 1 × 104 for CPRG and 1 × 106; 5 × 105; 1 × 105

nd 5 × 104 for X-Gal precipitation) and incubated on a Vero cell

onolayer in a 24-well (TPP) plate or 8-well chamber slide (Nunc

ab-Tek®) respectively for CPRG and X-Gal precipitation. The cul-ures were incubated for 2 h at 37 ◦C, in 5% CO2, washed three timesith PBS and lysed with �-galactosidase lysis buffer (CPRG) or fixed

emical Parasitology 196 (2014) 75–81 77

and incubated with the working solution (X-Gal precipitation). Theoutcomes from CPRG were plotted for the calculation of the R2

linear regression by the program GraphPad Prism 5.1. The imagesfrom X-Gal precipitation were captured with the AxioCam MRc 5camera in the microscope Zeiss Axioskop 40 and processed by theAxioVison 4.6 software.

2.11. Proliferation assay

Purified tachyzoites (1 × 103/well) were incubated in Vero cellmonolayers in 24-well plates (TPP) for CPRG or 8-well chambers(TPP) for X-Gal precipitation for 18 h, at 37 ◦C, 5% CO2; to themedium, different dilutions of pyrimethamine (Sigma) were added:16, 8, 4, 2, 1, 0.5 and 0.25 �M for CPRG and 4, 2, 1, 0.5 and 0.25 �Mfor X-Gal. These samples were incubated for a further 54 h (total of72 h) at 37 ◦C, 5% CO2. The samples for CPRG reaction or X-Gal pre-cipitation were processed as previously described (2.7 and 2.8). Thepercentage of inhibition, for CPRG reaction, was calculated usingthe formula (100 – ((T/C) × 100)), where T is the absorbance of thetreated well and C the absorbance of the wells without any drug.A similar calculation was performed for X-Gal precipitation, whereT and C represent the mean of the 20 field counts of treated andcontrol wells, respectively, in an optic microscope.

3. Results

3.1. Detection of CAT enzyme expression in N. caninumtachyzoites

The chloramphenicol acetyltransferase activity in transfectedtachyzoites exhibited a significant absorbance of 1.296 for Ncdhfr-CAT (data not show). The tachyzoites were evaluated 2 weeksafter the rise of stable tachyzoites (which corresponds to a totalof 5 weeks of chloramphenicol selection). The resistance of stabletransfected tachyzoites was retained for 14 months under chlor-amphenicol selection.

3.2. Expression of Lac-Z after selection with chloramphenicol inN. caninum tachyzoites

The selection under chloramphenicol generated tachyzoiteswith different patterns of Lac-Z expression (Fig. 2). Parasitesselected with higher concentrations (40 and 20 �M) reached aplateau of Lac-Z expression after the fourth lytic cycle, when thenumber of tachyzoites increased, similarly to the wild culture afterthe fifth cycle. Chloramphenicol at 10 �M has a slight effect ontachyzoite proliferation and the Lac-Z expression peak was reachedafter the seventh cycle. The proliferation of tachyzoites under 5 �Mwas not affected and the expression of Lac-Z was similar along theprocess of selection. The control transfected and non-selected indi-cated a �-galactosidase activity similar to the selected with 5 �M ofdrug in the first lytic cycle. The Lac-Z expression decreased until thewild type levels after the third cycle. The expression of Lac-Z of theparasites under 40, 20, 10 and 5 �M exhibited a relative absorbanceper 105 tachyzoites of 33.7; 23.3; 0.32 and 0.0045, respectively.

3.3. Episomal stability of parasites

Tachyzoites Ncdhfr-CAT or Ncdhfr-CAT/Lac-Z selected with20 �M of chloramphenicol were stable in culture for 14 months,at which point the growth started to decrease, until being com-pletely extinguished after 18 months of culture (inclusive in the

absence of chloramphenicol). After treatment of the genomic DNAwith PstI (Fig. S1A) and recircularisation, only plasmids withoutgenomic DNA of N. caninum were found, with a size of 5.9 kb (Fig.S1B). The treatment of the rescued plasmids with EcoRI (Fig. S1C)

78 L.M. Pereira, A.P. Yatsuda / Molecular & Biochemical Parasitology 196 (2014) 75–81

Fig. 2. Selection of Neospora caninum tachyzoites Ncdhfr-CAT/NcTub-tetO/Lac-Zwith chloramphenicol. Four groups of purified tachyzoites (3 × 107), in two parallelassays, were transfected with 5 �g of Ncdhfr-CAT/NcTub-tetO/Lac-Z and incubatedfor 18 h. Each culture was incubated with 5, 10, 20 and 40 �M of chloramphenicol,purified after the lytic cycle and inoculated (1 × 104) in a new Vero culture. The puri-fied tachyzoites from each lytic cycle were counted and the �-galactosidase activitymeasured by CPRG reaction and the absorbance 570 nm per 105 tachyzoites calcu-llr

lt1p

i2

z1uatopw(2

3

LtaTp(

i2

3

o

Fig. 3. Detection of Ncdhfr-CAT/NcTub-tetO/Lac-Z by PCR and Lac-Z expression byCPRG. Detection of the Ncdhfr-CAT and the region of the 5′ coding sequence flankingregion NcRPS13 from genomic DNA of Neospora caninum Lac-Z tachyzoites selectedwith 5, 10, 20 and 40 �M. (A) Agarose electrophoresis (0.8%) of the Ncdhfr-CAT andNcRPS13 amplification; lane 1 = wild type N. caninum, lanes 2–5 = N. caninum Lac-Z,

ated. (A) Logarithm of Absorbance/105 tachyzoites plotted against the respectiveytic cycle. (B) Tachyzoites counted in a haemocytometer and plotted against theespective lytic cycle.

iberated the predicted fragment of 1170 bp (Fig. S1D), with theypical pattern of circular Ncdhfr-CAT. Therefore, the resistance for4 months against the drug occurred due to episomal forms of thelasmid which were detected in all of the rescued clones.

Supplementary Fig. S1 related to this article can be found,n the online version, at http://dx.doi.org/10.1016/j.molbiopara.014.08.001.

The detection of plasmids by PCR was readily observed for tachy-oites selected with 20 and 40 �M and a faint band was observed for0 �M, whereas for 5 �M and wild type samples, the plasmids werendetectable (Fig. 3A, lanes 5, 4, 3, 2 and 1, respectively). In parallel,mplification of part of the 5′ coding sequence flanking region fromhe ribosomal gene NcRPS13 (651 bp) indicated a similar amountf DNA extracted from each sample (Fig. 3A). The same batch ofarasites was tested for Lac-Z expression. The tachyzoites selectedith 40 �M (ABS 20.87) improved the �-galactosidase activity 1.5

ABS 13.4), 52 (ABS 0.4) and 4496 (ABS 0.0046) times compared to0 �M, 10 �M and 5 �M, respectively (Fig. 3B).

.4. Detection of Lac-Z tachyzoites after invasion

Two hours after invasion, the N. caninum tachyzoites expressedac-Z detectable up to 1 × 104/well. The absorbance of each dilu-ion was 1.5 (1 × 106); 1.0 (5 × 105); 0.163 (1 × 105); 0.117 (5 × 104)nd 0.025 (1 × 104), excluding the absorbance of wild tachyzoites.he number of tachyzoites and their absorbance were sufficientlyroportional to generate a linear regression curve with R2 of 0.955Fig. S2).

Supplementary Fig. S2 related to this article can be found,n the online version, at http://dx.doi.org/10.1016/j.molbiopara.014.08.001.

.5. Detection of Lac-Z expression by X-Gal precipitation

The precipitation of X-Gal was proportional to the concentrationf chloramphenicol, being very intense in tachyzoites selected with

selected with, respectively, 5, 10, 20 and 40 �M of chloramphenicol. (B) Relativeabsorbance per 105 tachyzoites selected with chloramphenicol from the seventhcycle, evaluated by CPRG reaction.

20 and 40 �M (Fig. S3B and S3A), indicating the consistency of thetubulin promoter from N. caninum and the adaptation of the vectorwithin the parasite expression machinery. The tachyzoites selectedwith lower concentrations of the drug (10 and 5 �M) did not exhibita visible precipitation, as observed in the wild type (Fig. S3C, D andE).

Supplementary Fig. S3 related to this article can be found,in the online version, at http://dx.doi.org/10.1016/j.molbiopara.2014.08.001.

The Lac-Z tachyzoites were visualised macroscopically, forminga blue pellet after centrifugation (Fig. 4A), in contrast to the wildones (Fig. 4B). The staining allowed the detection of isolated tachy-zoites (Fig. 4C) including the visualisation of replicating (Fig. 4D)and early egressed forms (Fig. 4E). Tachyzoites were detected afterinvasion in Vero cells and different amounts of incubated parasiteswere distinguishable at concentrations of 1 × 106, 5 × 105, 1 × 105

and 5 × 104 (Fig. 4F, G, H and I, respectively).

3.6. Detection of ˇ-galactosidase by fluorescence

The �-galactosidase form expressed by N. caninum was detectedwith an antiserum against the enzyme form from E. coli. The fluores-cence allowed the discrimination of isolated parasites and indicateda uniform distribution of the enzyme along the parasite (Fig. 5A–D).The concentration of chloramphenicol during the selection of para-sites had an influence on the �-galactosidase expression and,consequentially, on the fluorescence detection. Parasites selectedwith 40 and 20 �M of drug demonstrated an intense fluorescence,whereas expression under 10 and 5 �M (Fig. 5G and H, respec-tively) was not sufficient to allow detection by this method andthe tachyzoites were undifferentiated when compared to the wildtype (Fig. 5I).

3.7. Evaluation of N. caninum (Ncdhfr-CAT/NcTub-tetO/Lac-Z)susceptibility to pyrimethamine

The expression of �-galactosidase allowed the measurement oftachyzoite proliferation by two methods: CPRG (Fig. 6A) and count-ing after X-Gal precipitation (Fig. 6B–H). Concentrations higherthan 1 �M of pyrimethamine inhibited the replication by ∼80% in

L.M. Pereira, A.P. Yatsuda / Molecular & Biochemical Parasitology 196 (2014) 75–81 79

Fig. 4. Detection of N. caninum Lac-Z tachyzoites (Ncdhfr-CAT/NcTub-tetO/Lac-Z) byX-Gal precipitation. The tachyzoites were detected by three approaches with the kit�-Gal Staining Set (Roche): in isolated forms, after cell invasion and the supernatantof a 5-day infected culture. (A and B) Macroscopic visualisation of X-Gal precipitationin Lac-Z and wild tachyzoites, respectively. (C) Isolated Lac-Z tachyzoites. (D) Lac-Ztat

tidasc

4

obo[

gaa

Fig. 5. Detection of N. caninum Lac-Z (Ncdhfr-CAT/NcTub-tetO/Lac-Z) by fluo-rescence. Tachyzoites selected with chloramphenicol were fixed, permeabilised,blocked and incubated with mouse anti-�-galactosidase and anti-mouse Alexa 488labelled conjugate and the nucleus stained with DAPI. (A) The nuclei were visualisedwith DAPI. (B) Detection of �-galactosidase. (C) Visualisation of tachyzoites by optic

achyzoites replicating on a Vero cell. (E) Early egressed Lac-Z tachyzoites. (F, G, Hnd I) Vero cells monolayers incubated with 1 × 106; 5 × 105; 1 × 105 and 5 × 104

achyzoites, respectively, for 2 h.

he CPRG assay (Fig. 6A) and ∼75% by X-Gal precipitation and count-ng (Fig. 6B). The visualisation of the cultures incubated with therug indicated a similar pattern, when the replication was inhibitedt concentrations higher than 1 �M (Fig. 6F, G and H) and demon-trated a partial proliferation with 0.5 and 0.25 �M (Fig. 6D and E)ompared to the control (Fig. 6C).

. Discussion

Despite being poorly developed for N. caninum, the insertionf genes in T. gondii started more than 20 years ago with cassettesased on CAT [6], followed by mutated DHFR [12], complete knock-uts [13], the insertion of reporter genes such as Lac-Z [14] or GFP10] and the site-specific insertion of genes [15].

We demonstrate that N. caninum, which is closely related to T.ondii, is a suitable model for the expression of chloramphenicolcetyltransferase (CAT). The design of vectors based on this enzymellowed the insertion of Lac-Z gene and the tagging of tachyzoites.

microscopy. (D) Images A, B and C merged. (E–I) Visualisation of nucleus and �-galactosidase from tachyzoites selected with 5, 10, 20 and 40 �M of chloramphenicoland the wild type, respectively.

The expression of CAT in N. caninum was robust, similar to thefindings reported for T. gondii [6]. Our work also evaluated CATexpression by the CAT ELISA, which directly indicates the increaseof enzyme expression. The tagging of N. caninum with CAT is a com-plementary method used to detect tachyzoites via in vitro assays,as performed in eukaryotic cells [16,17].

The episomal pattern allowed a preliminary form of geneexpression control in N. caninum. The dosage of chlorampheni-col determined the proportional expression of Lac-Z, according tothe selection with 5 or 10 �M and 20 or 40 �M of the drug byCPRG assay. X-Gal precipitation and fluorescence confirmed theapplication of chloramphenicol as a drug for expression control,complementing the insertion of genes. In addition to the low �-galactosidase detection, the tachyzoites selected with 5 and 10 �Mof chloramphenicol exhibited the same fluorescence and X-Galprecipitation as the wild type, which were clearly detectable onparasites selected with 20 or 40 �M. The possibility to improve theexpression of inserted genes by the addition of chloramphenicolis the first molecular tool for the evaluation of genes in N. can-inum. The overexpression of genes will contribute to the evaluationand design of tachyzoites applicable for protective tests in vivo orfor evaluating the effects of genes related to invasion, gliding ormetabolism. The overexpression of reporter genes such as GFP, YFPand luciferase might allow the design of more sensitive tests for

the detection of N. caninum without modifying the genome. Addi-tionally, the improvement of expression by chloramphenicol willprobably be useful for the detection of poorly expressed proteinsin N. caninum.

80 L.M. Pereira, A.P. Yatsuda / Molecular & Biochemical Parasitology 196 (2014) 75–81

Fig. 6. Incubation of N. caninum Lac-Z (Ncdhfr-CAT/NcTub-tetO/Lac-Z) with pyrimethamine. The proliferation of N. caninum was evaluated by CPRG reaction and X-Galprecipitation after incubation with pyrimethamine. Tachyzoites (1 × 103/well) were incubated in 24-well plates (for CPRG) or 8-well chambers (for X-Gal precipitation) andserial dilutions of pyrimethamine were added after 18 h, followed by incubation for a further 54 h (total of 72 h). (A) Percentage of inhibition of N. caninum proliferationi ipitatp

ticgstorsoztwibcatfito5tto

ncubated with pyrimethamine and evaluated after CPRG reaction (A) or X-Gal precyrimethamine (0, 0.25, 0.5, 1, 2 and 4 �M).

Lac-Z in T. gondii appeared as direct alternatives to evaluatehe effects of drugs on parasites [18,19], lysis [20] and detectionn tissues [14]. The Lac-Z N. caninum adhered/invaded the Veroells with a correlation between the number of tachyzoites and �-alactosidase activity (R2) of 0.955. The standardisation of a fast andimple invasion assay amplified the range of inhibitors/enhancerso be tested on invasion, thus optimising the screening processf drug candidates. Furthermore, the episomal maintenance ofesistance generates tagged tachyzoites with unaltered genomictructure and the presence of Lac-Z expression during 14 monthsf selection represents a long time window for the use of the tachy-oites. Despite the application of REMI transfection, which elevatedhe integration rate in T. gondii [11] no evidence of integrationas found and the episomal form was predominant for N. can-

num. The parasites could be detected not only macroscopically (thelue pellet was visible after centrifugation), but also microscopi-ally (either isolated or after invasion in Vero cells). As this wasn enzymatic reaction, the colour appeared as a complement forhe detection and quantification by microscopic counting, which,or N. caninum, was mainly based on immunofluorescence [21,22],mmunohistochemistry [23] and direct staining [9,24]. The detec-ion of tachyzoites under different stages of replication in Vero cellspens up the possibility of refining the assay via the substitution of

′ coding sequence flanking regions, which can improve or decreasehe X-Gal precipitation or have a time-specific expression for detec-ion in different periods of the life cycle, as seen for the promoterf BAG1 activated in the bradyzoite stage [25].

ion and counting (B). (C–H) Images of N. caninum incubated with serial dilutions of

Lac-Z tachyzoites allowed the design of proliferation assaysand the evaluation of drugs in this process. The parasites wereinhibited by pyrimethamine, an important drug for toxoplasmo-sis [26,27] and an auxiliary drug for malaria prophylaxis [28].Despite the great potential for neosporosis combat, the appli-cation of pyrimethamine on N. caninum proliferation assays hasbeen described once, based on microscopic counting after Giemsastaining of the cultures [29], where the concentration of total inhi-bition was 0.05 �g/ml [29]. In contrast, in our conditions, 0.25 �M(0.062 �g/ml) inhibited the proliferation in ∼30% of cells. An effec-tive dose to block N. caninum proliferation at levels of ∼80% wasfound to be 1 �M (0.25 �g/ml). These results will provide guidanceto a more specific form to determine the ideal dosage for in vivoapproaches, which will upgrade the application of folate inhibitorson neosporosis combat. Giemsa and other forms of subjective eval-uation of proliferation demand specialised technique support incontrast to the CPRG assay. Also, X-Gal precipitation exhibits pro-liferation without staining the host cells, as observed for Giemsa.

The application of the insertion of CAT vectors into genesamplifies the tools for N. caninum detection and the evaluationof molecular manipulation effects. The episomal maintenance anddose-response pattern of tachyzoites selected with chloramphen-icol indicates a new form of expression control of inserted genes

in N. caninum, which will lead to the development of more refinedmethods for gene evaluation. The tagging of tachyzoites withoutgenomic modification is the functional contribution of this work,offering parasites that are more closely related to wild type ones.

Bioch

A

(f(s

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

L.M. Pereira, A.P. Yatsuda / Molecular &

cknowledgements

We would like to thank CNPq (478767/2007-2) and FAPESP2005/53785-9) for the research Grants; FAPESP for the PhDellowship to L.M.P. (2009/07713-7) and Intervet Internationalcurrent Merck Animal Health) for the plasmid with TetO and Lac-Zequences (pRPS13LacZ, [8]).

eferences

[1] Dubey JP, Schares G. Neosporosis in animals – the last five years. Vet Parasitol2011;180(1–2):90–108.

[2] Pereira LM, Baroni L, Yatsuda AP. A transgenic Neospora caninum strain basedon mutations of the dihydrofolate reductase-thymidylate synthase gene. ExpParasitol 2014;138:40–7.

[3] Neu HC, Gootz TD. Antimicrobial chemotherapy. In: Baron S, editor. Medicalmicrobiology. Galveston, TX: University of Texas Medical Branch at Galveston;1996.

[4] Ramya TN, Mishra S, Karmodiya K, Surolia N, Surolia A. Inhibitors of non-housekeeping functions of the apicoplast defy delayed death in Plasmodiumfalciparum. Antimicrob Agents Chemother 2007;51(1):307–16.

[5] Shaw WV, Packman LC, Burleigh BD, Dell A, Morris HR, Hartley BS. Primarystructure of a chloramphenicol acetyltransferase specified by R plasmids.Nature 1979;282(5741):870–2.

[6] Kim K, Soldati D, Boothroyd JC. Gene replacement in Toxoplasma gondiiwith chloramphenicol acetyltransferase as selectable marker. Science1993;262(5135):911–4.

[7] Meissner M, Schluter D, Soldati D. Role of Toxoplasma gondii myosin A inpowering parasite gliding and host cell invasion. Science 2002;298(5594):837–40.

[8] van Poppel NF, Welagen J, Duisters RF, Vermeulen AN, Schaap D. Tight controlof transcription in Toxoplasma gondii using an alternative tet repressor. Int JParasitol 2006;36(4):443–52.

[9] Pereira LM, Candido-Silva JA, De Vries E, Yatsuda AP. A new thrombospondin-related anonymous protein homologue in Neospora caninum (NcMIC2-like1).Parasitology 2011;138(3):287–97.

10] Striepen B, He CY, Matrajt M, Soldati D, Roos DS. Expression, selection, andorganellar targeting of the green fluorescent protein in Toxoplasma gondii. MolBiochem Parasitol 1998;92:325–38.

11] Black M, Seeber F, Soldati D, Kim K, Boothroyd JC. Restriction enzyme-mediatedintegration elevates transformation frequency and enables co-transfection of

Toxoplasma gondii. Mol Biochem Parasitol 1995;74:55–63.

12] Donald RG, Roos DS. Stable molecular transformation of Toxoplasma gondii:a selectable dihydrofolate reductase-thymidylate synthase marker basedon drug-resistance mutations in malaria. Proc Natl Acad Sci U S A1993;90(24):11703–7.

[

emical Parasitology 196 (2014) 75–81 81

13] Donald RG, Carter D, Ullman B, Roos DS. Insertional tagging, cloning, andexpression of the Toxoplasma gondii hypoxanthine-xanthine-guanine phospho-ribosyltransferase gene. Use as a selectable marker for stable transformation. JBiol Chem 1996;271(24):14010–9.

14] Seeber F, Boothroyd JC. Escherichia coli beta-galactosidase as an in vitro andin vivo reporter enzyme and stable transfection marker in the intracellularprotozoan parasite Toxoplasma gondii. Gene 1996;169(1):39–45.

15] Huynh MH, Carruthers VB. Tagging of endogenous genes in a Toxoplasma gondiistrain lacking Ku80. Eukaryot Cell 2009;8(4):530–9.

16] Gorman CM, Moffat LF, Howard BH. Recombinant genomes which expresschloramphenicol acetyltransferase in mammalian cells. Mol Cell Biol1982;2(9):1044–51.

17] New DC, Miller-Martini DM, Wong YH. Reporter gene assays and their appli-cations to bioassays of natural products. Phytother Res 2003;17(5):439–48.

18] Jin C, Jung SY, Kim SY, Song HO, Park H. Simple and efficient model sys-tems of screening anti-Toxoplasma drugs in vitro. Expert Opin Drug Discov2012;7(3):195–205.

19] McFadden DC, Seeber F, Boothroyd JC. Use of Toxoplasma gondii expressingbeta-galactosidase for colorimetric assessment of drug activity in vitro. Antimi-crob Agents Chemother 1997;41(9):1849–53.

20] Seeber F. An enzyme-release assay for the assessment of the lytic activitiesof complement or antimicrobial peptides on extracellular Toxoplasma gondii. JMicrobiol Methods 2000;39(3):189–96.

21] He P, Li J, Gong P, Liu C, Zhang G, Yang J, et al. Neospora caninum surface anti-gen (p40) is a potential diagnostic marker for cattle neosporosis. Parasitol Res2013;112(5):2117–20.

22] Borsuk S, Andreotti R, Leite FP, Pinto L da S, Simionatto S, Hartleben CP, et al.Development of an indirect ELISA-NcSRS2 for detection of Neospora caninumantibodies in cattle. Vet Parasitol 2011;177(1–2):33–8.

23] Medina-Esparza L, Macias L, Ramos-Parra M, Morales-Salinas E, QuezadaT, Cruz-Vazquez C. Frequency of infection by Neospora caninum in wildrodents associated with dairy farms in Aguascalientes, Mexico. Vet Parasitol2013;191(1–2):11–4.

24] Machado RZ, Mineo TW, Landim Jr LP, Carvalho AF, Gennari SM, Miglino MA.Possible role of bovine trophoblast giant cells in transplacental transmission ofNeospora caninum in cattle. Rev Bras Parasitol Vet 2007;16(1):21–5.

25] Kobayashi T, Narabu S, Yanai Y, Hatano Y, Ito A, Imai S, et al. Gene cloning andcharacterization of the protein encoded by the Neospora caninum bradyzoite-specific antigen gene BAG1. J Parasitol 2013;99(3):453–8.

26] Kaye A. Toxoplasmosis: diagnosis, treatment, and prevention in congenitallyexposed infants. J Pediatr Health Care 2011;25(6):355–64.

27] Rorman E, Zamir CS, Rilkis I, Ben-David H. Congenital toxoplasmosis-prenatalaspects of Toxoplasma gondii infection. Reprod Toxicol 2006;21(4):458–72.

28] Nwakanma DC, Duffy CW, Amambua-Ngwa A, Oriero EC, Bojang KA, Pinder M,

et al. Changes in malaria parasite drug resistance in an endemic populationover a 25-year period with resulting genomic evidence of selection. J Infect Dis2014;209:1126–35.

29] Lindsay DS, Dubey JP. Evaluation of anti-coccidial drugs’ inhibition of Neosporacaninum development in cell cultures. J Parasitol 1989;75(6):990–2.