Deciphering fact from artifact when using reporter assays to investigate the roles of host factors on L1 retrotransposition

Abstract

Background: The Long INterspersed Element-1 (L1, LINE-1) is the only autonomous mobile DNA element in humans and has generated as much as half of the genome. Due to increasing clinical interest in the roles of L1 in cancer, embryogenesis and neuronal development, it has become a priority to understand L1-host interactions and identify host factors required for its activity. Apropos to this, we recently reported that L1 retrotransposition in HeLa cells requires phosphorylation of the L1 protein ORF1p at motifs targeted by host cell proline-directed protein kinases (PDPKs), which include the family of mitogen-activated protein kinases (MAPKs). Using two engineered L1 reporter assays, we continued our investigation into the roles of MAPKs in L1 activity.

Results: We found that the MAPK p38δ phosphorylated ORF1p on three of its four PDPK motifs required for L1 activity. In addition, we found that a constitutively active p38δ mutant appeared to promote L1 retrotransposition in HeLa cells. However, despite the consistency of these findings with our earlier work, we identified some technical concerns regarding the experimental methodology. Specifically, we found that exogenous expression of p38δ appeared to affect at least one heterologous promoter in an engineered L1 reporter, as well as generate opposing effects on two different reporters. We also show that two commercially available non-targeting control (NTC) siRNAs elicit drastically different effects on the apparent retrotransposition reported by both L1 assays, which raises concerns about the use of NTCs as normalizing controls.

Conclusions: Engineered L1 reporter assays have been invaluable for determining the functions and critical residues of L1 open reading frames, as well as elucidating many aspects of L1 replication. However, our results suggest that caution is required when interpreting data obtained from L1 reporters used in conjunction with exogenous gene expression or siRNA.

Keywords: HSV-TK; Host factor; L1; LINE-1; Promoter; Renilla; Reporter; SV40; p38.

Fig. 1

The MAPK p38δ phosphorylates ORF1p on S/T-P motifs required for L1 retrotransposition. a ORF1p-WT or S/T-P mutants (200 μM), purified from E. coli, were incubated with 85 nM activated p38δ-WT (top) or the constitutively active p38δ mutant F324S (bottom) in the presence of [γ-³²P]-ATP; bands on autoradiogram show ³²P incorporation into ORF1p. ORF1p mutants are S18A/S27A/T203G/T213G (AAGG), S18A/S27A (AA), T203G/T213G (GG), S27A/T203G/T213G (SAGG), S18A/T203G/T213G (ASGG), S18A/S27A/T213G (AATG) and S18A/S27A/T203G (AAGT). b ORF1p-WT was incubated with activated p38δ-WT, p38δ-F324S, an inactive p38δ mutant D176A, or no kinase in reactions as described in (a). c A Coomassie-stained gel shows each ORF1p construct (approximately 100 ng) purified from E. coli.

Fig. 2

Schematic of L1 reporter plasmids. All reporters contain a full-length L1 element with 5′ and 3′ UTRs (orange), ORF1 (pink), intergenic region (gray), ORF2 (blue) and a retrotransposition reporter (yellow) interrupted by an artificial intron (purple) with splice donor (SD) and acceptor (SA) sites. In JM101, L1 is driven by the CMV promoter (green), and in pYX017 by the hybrid CAG promoter (green). pYX014 contains only the native L1 promoter in the 5′UTR, and pYX015 is identical to pYX014 except for two missense mutations (R261A/R262A) [38] in ORF1p, rendering pYX015 incompetent for retrotransposition. The reporter in JM101 is an mneo cassette driven by the SV40 promoter (green) located within the 3′ UTR. The pYX017, pYX014 and pYX015 constructs contain a Firefly luciferase reporter (Fluc), also driven by SV40 (green), as well as a gene for Renilla luciferase (Rluc; aqua) driven by the HSV-TK promoter (green)

Fig. 3

Effects of p38δ on two different L1 reporter assays. a Top rows show duplicate wells of Giemsa-stained G418-resistant colonies resulting from transfection of the L1 reporter JM101 in the presence of pcDNA mammalian expression vectors for: empty vector (EV), p38δ-WT (WT), p38δ-F324S (FS) or p38δ-D176A (DA). Bottom row shows the effect of each pcDNA expression vector on cell growth. The right panel indicates fluorescence intensities obtained from cotransfection of EGFP with each indicated p38δ construct or empty vector; results from duplicate wells are shown. b Relative Fluc/Rluc luminescence ratios obtained from lysates of HeLa cells transfected with the L1 reporter plasmid pYX015 or pYX017 in the presence of indicated pcDNA mammalian expression vectors. Three biological replicates are shown for each experimental condition; error bars represent the SEM from two technical replicates (defined as two distinct samples taken from each biological sample). The graph at right shows the average of three biological replicates shown separately in the left panel; error bars indicate the SEM, n = 3 biological replicates. c Individual luminescence values are shown for Fluc (blue) and Rluc (red) used to calculate the Fluc/Rluc ratios from pYX017 in (b); technical replicates are side-by-side; biological replicates are indicated in subscript. d Mean Fluc and Rluc luminescence values were derived by first averaging the technical replicates for each biological sample (n = 2), and then averaging the resulting values of each biological replicate; error bars represent the SEM of biological replicates, n = 3

Fig. 4

p38δ increases Fluc independent of a heterologous promoter. a Duplicate wells containing G418-resistant colonies resulting from transfection of HeLa cells with the L1 reporter JM101 in the presence of pcDNA mammalian expression vectors for: empty vector (EV), p38δ-WT (WT) or p38δ-F3324S (FS). b Mean Fluc (left) and Rluc (right) luminescence values obtained from lysates of HeLa cells transfected with the L1 reporter plasmid pYX014 in the presence of indicated pcDNA mammalian expression vectors. Averages were derived from raw data shown in (c) by first averaging technical replicates for each biological sample (n = 3), and averaging biological replicates; error bars represent SEM of biological samples, n = 2. c Individual luminescence values are shown for Fluc (blue) and Rluc (red) used to calculate averages in (b); technical replicates are side-by-side; biological replicates are indicated with subscripts

Fig. 5

MKK3b_2E and pcDNA-MKK6_2E increase Rluc luminescence. a Mean Fluc (left) and Rluc (right) luminescence values obtained from lysates of HeLa cells transfected with the L1 reporter plasmid pYX015 or pYX017 in the presence of pcDNA-MKK3b_2E (M3) or pcDNA-MKK6_2E (M6). Averages were derived from data shown in (b) by first averaging technical replicates for each biological sample (n = 2), then using this value to average biological replicates; error bars represent SEM of biological samples, n = 3. b Individual luminescence values are shown for Fluc (blue) and Rluc (red) obtained from lysates transfected with pYX015 or pYX017 and the indicated pcDNA expression vectors; technical replicates are side-by-side; biological replicates are indicated with subscripts. c Wells show effects on cell growth in response to expression of pcDNA-MKK3b_2E (M3) or pcDNA-MKK6_2E (M6)

Fig. 6

NTC siRNAs have differential effects on L1 reporter assays. a Wells show G418-resistant colonies resulting from transfection of the L1 reporter JM101 in the presence of no siRNA (mock, with transfection reagent only) or 10 nM NTC #3 siRNA. Graph at right shows EGFP fluorescence from cells pretreated with 10 nM NTC #3 siRNA or mock (M); results from duplicate wells are shown. b Top row shows G418-resistant colonies resulting from the transfection of the L1 reporter JM101 in the presence or absence of 25 nM of indicated siRNA; bottom row shows effect of 25 nM of indicated siRNA on cell growth. c Mean Fluc (left) and Rluc (second from right) luminescence values obtained from lysates of HeLa cells transfected with the L1 reporter pYX017 in the presence of no siRNA (M) or 25 nM NTC #3 or NTC #5; averages were derived from data shown in (d) by first averaging technical replicates for each biological sample (n = 2), then using this value to average biological replicates; error bars represent SEM of biological samples, n = 3; average Fluc/Rluc ratios (third from right) are also shown. d Individual luminescence values are shown for Fluc (blue) and Rluc (red) obtained from lysates of HeLa cells transfected with pYX017 and the indicated siRNA; technical replicates are side-by-side; biological replicates are indicated with subscripts