Small fragment homologous replacement: evaluation of factors influencing modification efficiency in an eukaryotic assay system

Abstract

Homologous Replacement is used to modify specific gene sequences of chromosomal DNA in a process referred to as "Small Fragment Homologous Replacement", where DNA fragments replace genomic target resulting in specific sequence changes. To optimize the efficiency of this process, we developed a reporter based assay system where the replacement frequency is quantified by cytofluorimetric analysis following restoration of a stably integrated mutated eGFP gene in the genome of SV-40 immortalized mouse embryonic fibroblasts (MEF-SV-40). To obtain the highest correction frequency with this system, several parameters were considered: fragment synthesis and concentration, cell cycle phase and methylation status of both fragment and recipient genome. In addition, different drugs were employed to test their ability to improve technique efficiency. SFHR-mediated genomic modification resulted to be stably transmitted for several cell generations and confirmed at transcript and genomic levels. Modification efficiency was estimated in a range of 0.01-0.5%, further increasing when PARP-1 repair pathway was inhibited. In this study, for the first time SFHR efficiency issue was systematically approached and in part addressed, therefore opening new potential therapeutic ex-vivo applications.

Figure 1. Experimental design for SDF and cell clone generation.

A) SDF sequence is homologous to the entire wild type eGFP coding sequence. SDF-PCR-WT, 876 bp long was generated by PCR amplification with primer pair 1F/1R (Table 1). SDF-DIG-WT, 752 bp long, was obtained by HindIII and XhoI digestion of pCR-2.1 vector. C/T transition, responsible of fluorescence switching off, is showed. B) Sequencing analysis showing wild type (WT; top panel) and mutated (Mut; bottom panel) pCEP4-eGFP in C1 and D1 cell clones, respectively. Arrows indicate the modified base (C→T). C) FACS density plot of C1 (WT; top) and D1 (Mut; bottom) respectively. D) pCEP4-eGFP copy number determination for each cell clone.

Figure 2. Amount and type of transfected SDF.

A) Correction efficiencies after transfecting different amounts of SDF-PCR-WT in D1 cells. Positive events are used to determine the overall modification efficiency respect to D1 control cells transfected with a SDF homologous to mutated eGFP sequence (CTR). B) Different kind of SDFs were tested in D1 cells: double (ds-SDF-PCR-WT 12×10⁶ SDF/cell) or single strand (ss-SDF-PCR-WT 12×10⁶ SDF/cell) PCR fragments and fragment obtained by enzymatic digestion (ds-SDF-DIG-WT 12×10⁶ SDF/cell) were compared to cells transfected with SDF homologous to mutated eGFP sequence (CTR). For representative FACS dot plots see Fig. S3 and S4.

Figure 3. Modification efficiencies obtained testing different concentrations of mimosine, thymidine, vinblastine and SDFs with different superimposed methylation patterns on D1 cells.

A) For each drug the concentration that gives the highest percentage of synchronized cells and the lowest cell death (highlighted in grey) was selected. B) Correction efficiencies after transfection in different cell cycle phases. A SDF homologous to mutated eGFP sequence was used as control (CTR). Gene modification efficiency was enhanced when cells are synchronized in G₂/M phase (*p = 0.0001 respect to CTR and +p = 0.0001 respect to unsynchronized cells, Fig. S5). C) Differently in vitro methylated SDFs were tested to assess methylation involvement in gene modification efficiency. SDF-PCR-WT gave the highest efficiency of modification (*p resulted to be significant when compared to all treatments; specifically p = 0.002 respect to Dam+, p = 0.01 respect to SssI+, p = 0.008 respect to Dam+/SssI+, and p = 0.009 respect to SDF-DIG-WT). For representative FACS dot plots see Fig. S5 and S6.

Figure 4. Molecular analyses of sorted D1 cells.

A) Modification efficiency in D1 cells transfected with 12×10⁶ SDF-PCR-WT/cell. Positive cells (0.5%) were sorted and soon after reanalyzed (right panel) to asses population purity (>99%). B) PCR/RFLP analysis design. C) Amplicon is generated using RFLP primer pair. Cells transfected with mutated SDF represent our control (D1 CTR, lane 2). All amplification products were digested with BtsI, except lane 4. Restriction patterns of Sorted positive D1 clone (lane 1) and of parental eGFP C1 cells (lane 5) were identical. No restriction bands were present in D1 CTR (lane 2) and in sorted negative cells (lane 3). M is ladder 50 bp. D) Sequence analysis of D1 cells (sorted positive, sorted negative and CTR). The site-specific T-to-C conversion was present only in sorted positive cells.

Figure 5. Southern blot analysis.

A) Probe design. A 566 bp probe was used, recognizing a region of eGFP gene. Dashed box correspond to pCEP4-eGFP locus integrated within genomic DNA. BtsI site recovery highlight the correction of the eGFP gene. After SalI/BtsI genomic DNA digestion, two different restriction pattern can be obtained, according to the presence/absence of BtsI restriction site. B) Southern blot. A 1111 bp band was obtained only in cells in which BtsI site is present (D1 sorted positive and C1 clone).

Figure 6. eGFP expression increased after 5-Aza-2′-Deoxycytidine treatment.

A) Bright field (upper row) and fluorescent (bottom row) images of D1 sorted corrected cells at different experimental time (scale bar: 150 µm). B) eGFP expression, analyzed by Real Time PCR, after 24 h and 48 h of treatment with 0.5 µM 5-Aza-2′-Deoxycytidine respect to untreated (0 h) (*p = 0.002); untreated cells, at 24 h and 48 h, usually showed a decreasing relative eGFP expression (data not shown).

Figure 7. HpaII and AciI methylation analyses of integrated eGFP in C1 and D1 clones.

A) Experimental design showing the amplicon regions and their length within eGFP locus integrated in genomic DNA. HpaII and AciI site are indicated. B) Densitometric analyses of parental C1 clone methylation pattern on eGFP⁺ more positive, eGFP⁺ less positive and eGFP⁻ cells (see also Fig. S9B and Fig. S10B). ANOVA test gave a statistical significance of p<0.001 and p<0.005 respectively for HpaII and AciI panels. C) Densitometric analysis of methylation pattern of D1 SFHR-modified clone on both fluorescent and non fluorescent cells (see also Fig. S9C and S10C). ANOVA test gave a statistical significance of p<0.001 for both panels.

Figure 8. Relative modification efficiency in D1 cells transfected with SDF-PCR-WT and treated with α-Amanitin, 1,5-Isoquinolinediol and KU-55933.

Transfected samples were analyzed three days after transfection (3 days; black columns) or in parallel treated 24 hours after transfection with 0.5 µM of 5-Aza-dC for 48 hours (1 day+2 days 5-Aza-dC; white columns). No statistically significant differences were observed at 3 days (black bars) respect to untreated cells (SDF-PCR-WT). Demethylating effect of 5-Aza-dC increased eGFP detection in all samples (white columns) in a statistically significant manner (Δ p = 0.003; +p = 0.01; • p = 0.0007). 5-Aza-dC addition also disclosed the effect of 1,5-Isoquinolinediol on SDF-mediated correction in a statically significant manner respect either to cells not treated with 5-Aza-dC (**p = 0.0002) and to the cells transfected with SDF-PCR-WT in which no drug was added (*p = 0.003). Dashed lines refers to modification efficiency observed in cells without addition of any drug but treated by 5-Aza-dC. Results are from mean values of three independent experiments and are reported as relative modification efficiencies in respect to control without drugs (SDF-PCR-WT).