|Home | About | Journals | Submit | Contact Us | Français|
Entamoeba histolytica is an important human pathogen and a major health problem worldwide. Many aspects of parasite biology can be studied with the exception of stage conversion, which cannot be reproduced adequately in E. histolytica. The reptile parasite Entamoeba invadens is a vital model system for studying stage conversion since it can be induced to undergo both encystation and excystation with high efficiency in vitro. However, functional studies using E. invadens have been limited by the lack of genetic tools in this species. Here, we report a new method for both transient and stable transfection of E. invadens. These new tools will greatly enhance research into Entamoeba development.
Amebiasis, caused by the protozoan parasite Entamoeba histolytica, is a major health problem in humans, affecting 50 million people a year and causing an estimated 100,000 deaths . The life cycle involves two stages: the invasive trophozoite form and the environmentally resistant cyst form. Spread of the parasite to new hosts requires conversion from the disease-causing trophozoite form to the infectious cyst form; thus, stage conversion is essential for disease transmission . Some information regarding the developmental cascade has been identified using clinical E. histolytica isolates in which cysts are present, including identification of developmentally regulated transcripts and a stage-specific transcription factor which is important in regulating encystation [3, 4]. Although E. histolytica recent clinical isolates have cysts, regulated, high-level encystation has not been possible in this species despite significant efforts [2, 5]. Thus, research into this vital aspect of parasite biology has been hampered by the lack of a good in vitro model for E. histolytica encystation.
In the absence of such a model, researchers have utilized a related species, Entamoeba invadens, which infects reptiles and causes invasive disease similar to E. histolytica infection in humans . In E. invadens, both encystation and excystation can be performed in vitro with high efficiency . Encystation is triggered by glucose starvation and osmotic shock, and excystation is initiated by incubation in rich media with the addition of bile salts, which mimics the environment in the small intestine [8-10]. However, the utility of this model system is compromised by a dearth of genetic tools to manipulate gene expression. Here, we report development of a method for transfection of E. invadens, which can be used both for transient expression of genes of interest and to generate stable transfectant parasite lines. These advances will allow analyses of genes in E. invadens at the molecular level and facilitate genetic dissection of the developmental cascade.
To create the transient transfection expression vector, we wanted to use promoters of genes that have high expression in both trophozoites and cysts. The E. invadens enolase gene (EIN_093390) was chosen and RT-PCR confirmed high gene expression in E. invadens trophozoites, and continued expression during encystation up to 48 hours (data not shown). To generate the vector (pEi-eno-Luc), the 5' and 3' flanking regions of the E. invadens enolase gene were amplified from genomic DNA by PCR and cloned into pBluescript SK- (Agilent). For the 5' promoter region, a 480bp segment (including a 12bp portion of the enolase coding region) was amplified, and cloned into BamHI/HindIII sites. For the 3' downstream regulatory region, a 484bp fragment was used and cloned using XhoI/KpnI. Finally, firefly luciferase was amplified from the TP4i vector  and inserted between the enolase flanking sequences using the XhoI and HindIII sites (Figure 1A). For use in stable transfection, the pEi-eno-Neo construct was generated by replacing the luciferase coding region with the coding region for neomycin phosphotransferase, which had been PCR amplified from pKT-3M (Figure 1B). Sequences for all cloning primers are listed in Table 1.
E. invadens strain IP-1 was maintained in LYI-S-2  at 25°C . In order to determine optimal conditions for electroporation, transient transfection assays were performed with pEi-eno-Luc. We first attempted transfection following the previously published protocols for E. histolytica , however, we did not achieve significant levels of luciferase expression with this method. Hence, we adapted the electroporation protocol from a previously published method for transfection of T. brucei , a parasite which, like E. invadens, grows at 25°C. For each trial, one 25cm2 flask of log-phase trophozoites was iced, harvested, and resuspended in 0.4ml ZM buffer (132mM NaCl, 8mMKCl, 8mM Na2PO4, 1.5mM KH2PO4, 0.5mM Magnesium acetate, 0.09mM CaCl2) at RT, and transferred to a 2mm gap cuvette. Parasites were electroporated with the pEi-eno-luc vector using 1.5kV and 25μF, giving an average time constant of ~0.2ms. After transfection, the trophozoites were transferred to LYI-S-2 media in 15ml glass tubes, and left to recover at 25°C. Transfection efficiency was assayed by measuring luciferase activity with the Promega Luciferase Assay kit following the manufacturers instructions. A number of different electroporation parameters were attempted, including differing amounts of DNA (30μg and 60μg) and numbers of pulses (one versus two pulses). Luciferase activity was measured at 17 and 24 hours post electroporation, with the highest activity at 17h post transfection, with considerable loss of activity by 24h. As in E. histolytica, increasing the amount of DNA and using 2 pulses increased the transfection efficiency (data not shown) . The highest luciferase activity detected (using 60μg DNA, 2 pulses, and measured at 17h post transfection) was >3000 fold over lysate from untransfected cells, indicating that transfection was successful and allowing us to proceed with attempts at generating stable transfectants.
Before attempting to make stable transgenic lines, we first tested the growth of E. invadens IP-1 in neomycin (G418) and hygromycin in order to determine selection conditions. We performed killing curves with neomycin (from 0.1μg/ml to 100μg/ml). Trophozoites were grown under standard culture conditions and drug was added to log phase parasites. At 24-hour intervals, tubes were examined for living, adherent parasites. E. histolytica HM-1:IMSS parasites were used as a control and were susceptible to 1μg/ml G418. In contrast, E. invadens IP-1 was much more resistant to neomycin, killing 100% of parasites only at 50μg/ml G418. This strain was even more resistant to hygromycin, requiring up 100μg/ml for complete killing (data not shown).
Stable transfectants were first generated using the pEi-eno-Neo construct. Electroporation was performed with 50μg plasmid following the protocol outlined above with 2 pulses of 1.5kV. Drug selection was begun at 24h post transfection. Interestingly, despite our finding that 50μg/ml G418 was required for killing untransfected amoeba, multiple attempts to select at that concentration did not succeed, with 100% lethality of the parasites. We hypothesized that this was possibly due to increased sensitivity of the parasites after electroporation and thus attempted selecting for stable transfectants in a graded manner. We were successful in selecting for stable transgenic parasites by culturing in 20μg/ml G418, beginning at 24h after electroporation and continuing until >90% of parasites had died off (approximately 1 week). This was followed by reduction of drug concentration to 5μg/ml for ~2 weeks, until cells had gone through two passages, after which drug concentration was slowly increased to 60μg/ml. Mock-transfected parasites (electroporated with no DNA) subjected to the identical treatment all died and did not recover despite the graded drug selection, indicating that parasites which survived in the experimental sample were not drug resistant mutants but had maintained the plasmid and Neo cassette. To confirm that we had true transgenic strains, we performed a Northern blot for expression of Neo (Figure 2A). RNA was isolated from untransfected E. invadens IP-1 trophozoites, and from pEi-eno-Neo transfectants growing at 30μg/ml and 60μg/ml G418, using Trizol reagent (Invitrogen). Purified RNA (10μg) was run on a 1% agarose gel, transferred to nitrocellulose, and hybridized with a [α-32P] labeled PCR probe generated against Neo. Expression of Neo was seen at both 30μg/ml and 60μg/ml G418, although with higher expression at the higher drug concentration, indicating a higher copy number consistent with episomal maintenance of the plasmid, as has been seen previously with the stable transfection of E. histolytica , and in a recently published method for E. invadens transfection . As expected, no Neo expression was seen in untransfected trophozoites. Expression of the endogenous enolase gene was probed as a loading control and demonstrated equivalent levels of RNA in all lanes (Figure 2A). Sequences for primers used for probe amplification are listed in Table 1.
In order to overexpress genes in E. invadens, we developed the construct pEi-ck-Myc (Figure 1C). This plasmid was derived from pEi-eno-Neo by introduction of the flanking regions from the casein kinase II (ckII) gene (EIN_083650). This gene, like enolase, has high expression in both trophozoites and cysts, as demonstrated by RT-PCR (data not shown). A 489bp segment from the ckII 5' regulatory region (including 15bp of coding region) was amplified from genomic DNA and cloned into pEi-eno-Neo using NotI/XbaI. A 445bp segment of the ckII 3' flanking regulatory region was cloned into the vector with SacI/SacII. A sequence coding for a 3xMyc tag was amplified by PCR from the plasmid pKT-3M , and added by cloning into the NotI and SacII sites. This Myc-tag region includes the additional cloning sites NheI and AvrII, for C-terminal and N-terminal tagging, respectively (Figure 1C). Finally, as the addition of these sequences resulted in an in-frame stop codon, the plasmid was modified by site-directed mutagenesis using the QuikChangeII kit (Agilent) to remove the stop codon and allow expression of the gene of interest.
A Myb domain containing protein (EIN_241140) was chosen for overexpression studies and cloned into this construct. This gene was chosen based on homology to a gene previously identified as upregulated in E. histolytica cysts and which serves as a potential regulator of stage conversion [3, 4]. In addition, this gene has low expression in E. invadens trophozoites and thus would allow us to determine whether our construct facilitated its overexpression (data not shown). The full-length coding region was amplified by PCR from E. invadens genomic DNA, and cloned into the AvrII site of pEi-ck-Myc, which should result in an N-terminal fusion of the 3xMyc tag.
Stable transfectants were generated from this construct, using the protocol outlined above. When the ck-Myb parasites were growing to confluence at 40μg/ml G418, RNA was extracted with Trizol reagent, and used to test for gene expression. This RNA, as well as RNA extracted from untransfected trophozoites, was used to generate cDNA using previously published methods . RT-PCR was performed using a forward primer that hybridizes to the Myc tag, and a reverse primer that hybridizes to the EiMyb coding region. This primer pair should only amplify transcript from transfected cells, and should not detect the endogenous Myb transcript. We identified a PCR product in the cDNA from the ck-Myb overexpressing cells and not from the untransfected controls (Figure 2B). Actin (EIN_093130) was used as a loading control, and showed equivalent signal from both samples. Control samples (with no reverse transcriptase) as expected gave no PCR product. In addition, we performed a Northern blot using 20μg of RNA following the procedures outlined above. An oligonucleotide probe directed at the Myc tag, plus an additional 12bp targeted to hybridize with the portion of ckII coding region that was included as a fusion, was labeled with [γ-32P] ATP and hybridized overnight. We detected signal at the appropriate size in RNA from the ck-Myb parasites but not for the control parasites (Figure 2C). Enolase was used as a loading control. Sequences for primers used are listed in Table 1.
Our establishment of a system for transient and stable transfection in E. invadens is an important step in developing tools that will facilitate studying the genetic and molecular basis of stage conversion in Entamoeba. A similar method was recently developed by another group, which also used electroporation and graded selection for stable transfectants [16, 18]. Some differences in the approaches include our use of electroporation conditions for Trypanosomes whereas the group of Bhattacharya used conditions more similar to those used for E. histolytica, with some modifications including a higher V/cm. However, many similar findings emerge from both studies, including the apparent episomal nature of the plasmid (similar also to what is noted in E. histolytica), the higher resistance of E. invadens to neomycin, and the relatively slow selection process obtaining stable transfectants (a time frame that is significantly longer than that in E. histolytica). These advances in genetic manipulation of E. invadens will enable molecular characterization of proteins that are potential regulators of encystation and will be an important use of the available genome sequence information. Many studies, including analysis of promoter activity, protein localization, effect of gene overexpression and genetic manipulation (such as regulated expression, dominant-negative approaches, or antisense) can now be attempted. In addition, the ability to generate stably transfected parasites lays the foundation for the development of RNAi-based techniques for gene knockdown, which have been used successfully in E. histolytica [19, 20], and which will contribute greatly to our understanding of the role targeted genes play in amebic development. The ability to genetically manipulate E. invadens will greatly enhance the study of Entamoeba development.
We acknowledge all members of the Singh lab for helpful comments and suggestions, especially Hussein Alramini, Vy Tran, Hanbang Zhang, and Justine Pompey. The work was supported by NIH grants (AI088042 and AI094887) to US.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.