Loss of dsRNA-based gene silencing in Entamoeba histolytica: implications for approaches to genetic analysis

Abstract

The ability to regulate gene expression in the protozoan parasite Entamoeba histolytica is critical in determining gene function. We previously published that expression of dsRNA specific to E. histolytica serine threonine isoleucine rich protein (EhSTIRP) resulted in reduction of gene expression [MacFarlane, R.C., Singh, U., 2007. Identification of an Entamoeba histolytica serine, threonine, isoleucine, rich protein with roles in adhesion and cytotoxicity. Eukaryotic Cell 6, 2139-2146]. However, after approximately one year of continuous drug selection, the expression of EhSTIRP reverted to wild-type levels. We confirmed that the parasites (i) contained the appropriate dsRNA plasmid, (ii) were not contaminated with other plasmids, (iii) the drug selectable marker was functional, and (iv) sequenced the dsRNA portion of the construct. This work suggests that in E. histolytica long term cultivation of parasites expressing dsRNA can lead to the loss of dsRNA based silencing through the selection of "RNAi" negative parasites. Thus, users of the dsRNA silencing approach should proceed with caution and regularly confirm gene down regulation. The development and use of constructs for inducible expression of dsRNA may help alleviate this potential problem.

Figure 1. Expression of EhSTIRP no longer downregulated in dsRNA (+)/silencing (−) parasites despite neomycin expression and putative cleavage of dsRNA

(A) Expression of EhSTIRP is shown by Northern blot analysis for nontransfected parasites (no trx), control transfected, dsRNA (+)/silencing (+) parasites (6 months after transfection) and dsRNA (+)/silencing (−) parasites (1 year after transfection). Actin and methylene blue stained ribosomal RNA are shown as loading controls. (B) Expression of the neomycin resistance gene (neo) is shown by Northern blot analysis for nontransfected parasites (no trx), control transfected, dsRNA (+)/silencing (+) and dsRNA (+)/silencing (−) parasites. Methylene blue stained ribosomal RNA is shown as loading control. (C) RNA from nontransfected parasites (no trx), control transfected or dsRNA(+)/silencing (−) parasites were probed for the presence the control or dsRNA cassette using the 3′ gene fragment as a probe. Control transfected parasites at 48ug/ml G418 exhibited robust expression of the 5′ gene fragment/stuffer region/3′ gene fragment, while the dsRNA (+)/silencing (−) parasites contained very little of the 3′ gene fragment/stuffer region/3′ gene fragment transcript despite expression of the neomycin resistance gene (Figure 1B). Stable transfectants were obtained with pJST4-EhSTIRP dsRNA and pJST4-EhSTIRP control constructs as previously described (MacFarlane, 2007) and maintained at 48μ/ml G418. Northern blots were performed as previously described (MacFarlane, 2007). Representative images from multiple blots are shown.

Figure 1. Expression of EhSTIRP no longer downregulated in dsRNA (+)/silencing (−) parasites despite neomycin expression and putative cleavage of dsRNA

(A) Expression of EhSTIRP is shown by Northern blot analysis for nontransfected parasites (no trx), control transfected, dsRNA (+)/silencing (+) parasites (6 months after transfection) and dsRNA (+)/silencing (−) parasites (1 year after transfection). Actin and methylene blue stained ribosomal RNA are shown as loading controls. (B) Expression of the neomycin resistance gene (neo) is shown by Northern blot analysis for nontransfected parasites (no trx), control transfected, dsRNA (+)/silencing (+) and dsRNA (+)/silencing (−) parasites. Methylene blue stained ribosomal RNA is shown as loading control. (C) RNA from nontransfected parasites (no trx), control transfected or dsRNA(+)/silencing (−) parasites were probed for the presence the control or dsRNA cassette using the 3′ gene fragment as a probe. Control transfected parasites at 48ug/ml G418 exhibited robust expression of the 5′ gene fragment/stuffer region/3′ gene fragment, while the dsRNA (+)/silencing (−) parasites contained very little of the 3′ gene fragment/stuffer region/3′ gene fragment transcript despite expression of the neomycin resistance gene (Figure 1B). Stable transfectants were obtained with pJST4-EhSTIRP dsRNA and pJST4-EhSTIRP control constructs as previously described (MacFarlane, 2007) and maintained at 48μ/ml G418. Northern blots were performed as previously described (MacFarlane, 2007). Representative images from multiple blots are shown.

Figure 1. Expression of EhSTIRP no longer downregulated in dsRNA (+)/silencing (−) parasites despite neomycin expression and putative cleavage of dsRNA

(A) Expression of EhSTIRP is shown by Northern blot analysis for nontransfected parasites (no trx), control transfected, dsRNA (+)/silencing (+) parasites (6 months after transfection) and dsRNA (+)/silencing (−) parasites (1 year after transfection). Actin and methylene blue stained ribosomal RNA are shown as loading controls. (B) Expression of the neomycin resistance gene (neo) is shown by Northern blot analysis for nontransfected parasites (no trx), control transfected, dsRNA (+)/silencing (+) and dsRNA (+)/silencing (−) parasites. Methylene blue stained ribosomal RNA is shown as loading control. (C) RNA from nontransfected parasites (no trx), control transfected or dsRNA(+)/silencing (−) parasites were probed for the presence the control or dsRNA cassette using the 3′ gene fragment as a probe. Control transfected parasites at 48ug/ml G418 exhibited robust expression of the 5′ gene fragment/stuffer region/3′ gene fragment, while the dsRNA (+)/silencing (−) parasites contained very little of the 3′ gene fragment/stuffer region/3′ gene fragment transcript despite expression of the neomycin resistance gene (Figure 1B). Stable transfectants were obtained with pJST4-EhSTIRP dsRNA and pJST4-EhSTIRP control constructs as previously described (MacFarlane, 2007) and maintained at 48μ/ml G418. Northern blots were performed as previously described (MacFarlane, 2007). Representative images from multiple blots are shown.

Figure 2. Schematic of dsRNA/control constructs and PCR analysis

(A) The constructs encoding for a dsRNA hairpin or the control (non-annealing) transcript are shown. In the dsRNA construct two 395bp segments from the 3′ region of the gene family (separated by a non-specific stuffer region of 500bp) were cloned in opposing orientation. The 395bp segment was amplified from the 3′ end of EhSTIRP in a region of high similarity amongst all family members and would be predicted to down regulate the entire EhSTIRP gene family. The control construct contained the 395bp region from the 3′ end of the gene and 350bp from the 5′ end of EhSTIRP1 and thus was incapable of forming dsRNA. Both the insert and the selectable marker Neo were placed under the control of the EhActin1 5′ and 3′ regulatory regions. Arrows indicate the location of the primers used for PCR based identification of the constructs. (B) PCR amplified products were run on a 1 % agarose gel. Lane 1 = PCR product amplified from dsRNA(+)/silencing (−) gDNA using stuffer forward and 3′ forward primers. Lane 2 = PCR product amplified from dsRNA(+)/silencing (−) gDNA using stuffer forward and 3′ reverse primers. Lane 3 = PCR product amplified from dsRNA(+)/silencing (−) gDNA using stuffer reverse and 3′ forward primers. (C) PCR amplified products were run on a 1% agarose gel. Lane 1 = PCR product amplified from dsRNA(+)/silencing (−) gDNA using 3′ forward and 5′ reverse primers. Lane 2 = PCR product amplified from control gDNA using 3′ forward and 5′ reverse primers. Lane 3 = negative control (PCR product amplified using all primers sets but no gDNA template). Polymerase chain reaction was performed as previously described (Shah, et al., 2005). Briefly, gDNA from transfected parasites was used as a template and primers corresponding to portions of the dsRNA construct or the control construct were used. The primers used in this study are listed in Table 1.

Figure 2. Schematic of dsRNA/control constructs and PCR analysis

(A) The constructs encoding for a dsRNA hairpin or the control (non-annealing) transcript are shown. In the dsRNA construct two 395bp segments from the 3′ region of the gene family (separated by a non-specific stuffer region of 500bp) were cloned in opposing orientation. The 395bp segment was amplified from the 3′ end of EhSTIRP in a region of high similarity amongst all family members and would be predicted to down regulate the entire EhSTIRP gene family. The control construct contained the 395bp region from the 3′ end of the gene and 350bp from the 5′ end of EhSTIRP1 and thus was incapable of forming dsRNA. Both the insert and the selectable marker Neo were placed under the control of the EhActin1 5′ and 3′ regulatory regions. Arrows indicate the location of the primers used for PCR based identification of the constructs. (B) PCR amplified products were run on a 1 % agarose gel. Lane 1 = PCR product amplified from dsRNA(+)/silencing (−) gDNA using stuffer forward and 3′ forward primers. Lane 2 = PCR product amplified from dsRNA(+)/silencing (−) gDNA using stuffer forward and 3′ reverse primers. Lane 3 = PCR product amplified from dsRNA(+)/silencing (−) gDNA using stuffer reverse and 3′ forward primers. (C) PCR amplified products were run on a 1% agarose gel. Lane 1 = PCR product amplified from dsRNA(+)/silencing (−) gDNA using 3′ forward and 5′ reverse primers. Lane 2 = PCR product amplified from control gDNA using 3′ forward and 5′ reverse primers. Lane 3 = negative control (PCR product amplified using all primers sets but no gDNA template). Polymerase chain reaction was performed as previously described (Shah, et al., 2005). Briefly, gDNA from transfected parasites was used as a template and primers corresponding to portions of the dsRNA construct or the control construct were used. The primers used in this study are listed in Table 1.