A genetic screen for improved plasmid segregation reveals a role for Rep20 in the interaction of Plasmodium falciparum chromosomes

Abstract

Bacterial plasmids introduced into the human malaria parasite Plasmodium falciparum replicate well but are poorly segregated during mitosis. In this paper, we screened a random P.falciparum genomic library in order to identify sequences that overcome this segregation defect. Using this approach, we selected for parasites that harbor a unique 21 bp repeat sequence known as Rep20. Rep20 is one of six different repeats found in the subtelomeric regions of all P.falciparum chromosomes but which is not found in other eukaryotes or in other plasmodia. Using a number of approaches, we demonstrate that Rep20 sequences lead to dramatically improved episomal maintenance by promoting plasmid segregation between daughter merozoites. We show that Rep20(+), but not Rep20(-), plasmids co-localize with terminal chromosomal clusters, indicating that Rep20 mediates plasmid tethering to chromosomes, a mechanism that explains the improved segregation phenotype. This study implicates a direct role for Rep20 in the physical association of chromosome ends, which is a process that facilitates the generation of diversity in the terminally located P.falciparum virulence genes.

Fig. 1. Screening of a random genomic library for improved episomal maintenance in P.falciparum selects for a Rep20-containing plasmid. The plasmid library pHH/DRlib was constructed by ligation of random 0.5–2.0 kb P.falciparum inserts into the EcoRV (E) site of pHH1 (Reed et al., 2000). pHH/DRlib plasmids contain a marker to confer WR99210 resistance to parasites (Fidock and Wellems, 1997) (hDHFR cassette) and a partial fragment of the P.berghei DHFR-TS 3′ UTR (Pb). DNA from the same pHH/DRlib preparation was transfected directly into E.coli and into P.falciparum on two separate occasions (Pf #1 and Pf #2). Following the establishment of drug-resistant populations, plasmids from Pf #1 and Pf #2 transformants were recovered in E.coli. In all three instances (indicated by arrows), plasmid DNA was extracted from 18 randomly chosen E.coli colonies and digested with a combination of BamHI (Ba) and NotI (N) to release inserts. Plasmids profiles were examined on ethidium bromide-stained gels (upper panels) that were subsequently transferred to nitrocellulose for hybridization with a Rep20 probe (bottom panels). DNA markers loaded in the left hand lane of each stained gel are indicated in kb. Released fragments of >1.2 kb represent the presence of a P.falciparum insert. The asterisk indicates the Rep20⁺ plasmid used in further studies, termed pHH/DR1.4.

Fig. 2. Rep20 allows rapid establishment of transfected P.falciparum lines. (A) Comparison of the ability of Rep20⁺ and parental Rep20^– plasmids (pHHΔE and pHH/DRlib) to establish drug-resistant populations. Rep20⁺ plasmids included derivatives of pHH/DR1.4 with the Rep20 sequence truncated as shown. Within each experiment an identical number of parasites from the same parasite population were transfected and the individual transformed populations were cultured under identical conditions. The number of days required for the cultures to reach 1% parasitemia post-transfection is indicated. (B) Verification of individual transformed lines by recovery in E.coli of plasmids in DpnI-treated DNA. The restriction profile of plasmid DNA extracted from three randomly chosen E.coli colonies (designated 1, 2 and 3) was compared to that of the originally transformed plasmid (P). DNA markers in kb (M) are indicated on the left. ND, not determined. (C) Additional Rep20⁺ plasmids were derived by transfer of Rep20 sequences into the NotI (N) site of pHHC* to derive pHHC*/DR1.4 and pHHC*/DR0.28. Another Rep20⁺ plasmid was derived by insertion of a 0.5 kb Rep20 sequence from a different P.falciparum line (3D7) into pHHMC* to derive pHHMC*/3R0.5. These Rep20⁺ plasmids and their parental Rep20^– controls were analyzed for their ability to establish drug-resistant parasite lines as described above. (D) Verification of individual transfected lines derived by plasmid recovery in E.coli as described above.

Fig. 3. Rep20⁺ plasmids show improved retention compared with Rep20^– plasmids in transfected P.falciparum parasites. Drug pressure was removed from synchronous parallel cultures of pHHC* (open squares) and pHHC*/DR1.4 (open circles) transfected lines at day 0. DNA was extracted at different time points from these cultures, treated overnight with DpnI before digestion with various enzyme combinations for Southern blot analysis. Relative plasmid copy number was determined at each time point by phosphoimager analysis of plasmid (P) and endogenous (E) hybridization signals on Southern blots from three different enzyme/probe combinations. The error bars represent the standard deviations from these three calculations. The inset shows a representative blot of BglII–NsiI–EcoRI-restricted genomic DNA from day 0 and 21 cultures hybridized to a combined pGEM/MSP-1 probe. This illustrates the complete loss of plasmids from both transfected lines after 21 days of culture in the absence of drug.

Fig. 4. FISH analysis confirms that Rep20⁺ plasmids are segregated more efficiently than Rep20^– plasmids in transfected P.falciparum parasites. (A) Representative multinucleated schizonts from pHHC*/DR1.4 (Rep20⁺) and pHHC* (Rep20^–) transfected populations by hybridization to a plasmid probe. (B) Quantitation of the number of plasmid-positive nuclei in schizonts that possessed at least one plasmid signal. The number of nuclei counted in each population is shown.

Fig. 5. Rep20⁺ plasmids physically associate with P.falciparum telomeric clusters. Smears of trophozoite-stage parasites from Rep20^– pHHC* (A) and Rep20⁺ pHHC*/DR1.4 (B) transfected populations, prepared from day 0 cultures shown in Figure 3, were analyzed by DAPI staining and by FISH using TARE 3 (Figueiredo et al., 2000), and plasmid backbone probes used to detect telomeric clusters and transfected plasmid, respectively. The two rows in each panel show two different representative fields.

Fig. 6. Proposed model demonstrating a mechanism by which Rep20 mediates plasmid tethering and plays a role in telomeric cluster formation. The subtelomeric region of P.falciparum chromosomes extends ∼60 kb from the telomere and is comprised of six different non-coding TAREs and a coding region containing virulence-associated genes that are organized in a conserved arrangement on different chromosomes. The physical association of Rep20-containing plasmids (which exist as covalently linked concatamers; O’Donnell et al., 2001) with telomeric clusters demonstrated in this paper presumably occurs via an interaction with Rep20-binding proteins. These proteins, and perhaps others that bind to different TAREs (question marks), would cross-link the subtelomeric regions of P.falciparum chromosomes and promote the formation/stabilization of the cluster. This alignment of chromosome ends favors ectopic recombination and hence the generation of diversity in the neighboring virulence genes, most particularly var (Freitas-Junior et al., 2000). Truncated chromosomes that have spontaneously lost their subtelomeric sequences, but not their telomere tracts, which are in fact amplified, do not associate with telomeric clusters, consistent with a role for TAREs in cluster stabilization (Scherf et al., 2001; Figueiredo et al., 2002). These truncated chromosomes remain anchored to the nuclear membrane (NM) by a different mechanism that probably resembles peripheral nuclear membrane tethering seen in yeast (Tham and Zakian, 2000; Scherf et al., 2001).