Hairpin dsRNA does not trigger RNA interference in Candida albicans cells

Abstract

RNA interference/silencing mechanisms triggered by double-stranded RNA (dsRNA) have been described in many eukaryotes, including fungi. These mechanisms have in common small RNA molecules (siRNAs or microRNAs) originating from dsRNAs that, together with the effector protein Argonaute, mediate silencing. The genome of the fungal pathogen Candida albicans harbours a well-conserved Argonaute and a non-canonical Dicer, essential members of silencing pathways. Prototypical siRNAs are detected as members of the C. albicans transcriptome, which is potential evidence of RNA interference/silencing pathways in this organism. Surprisingly, expression of a dsRNA a hairpin ADE2 dsRNA molecule to interfere with the endogenous ADE2 mRNA did not result in down-regulation of the message or produce adenine auxotrophic strains. Cell free assays showed that the hairpin dsRNA was a substrate for the putative C. albicans Dicer, discounting the possibility that the nature of the dsRNA trigger affects silencing functionality. Our results suggested that unknown cellular events govern the functionality of siRNAs originating from transgenes in RNA interference/silencing pathways in C. albicans.

Figure 1

Plasmid construct for expression of hairpin ADE2 dsRNA within cells to interfere with the ADE2 message. (a) Schematic of the ADE2 hairpin construct. ADE2 sequences from nt +1 to +414 were cloned in opposite orientation flanking 100 bp of yEGFP3, downstream of the enolase promoter (ENO1p) and with the 3′ untranslated region of HWP1 (3′ HWP1) in pENO1GFP3. Arrows indicate 5′ to 3′ of ADE2. Relevant restriction endonuclease sites are shown above the construct. (b) Endogenous ADE2 mRNA levels are not affected in a strain bearing the dsRNA plasmid construct. Top panel: northern blot analysis of the wild-type strain SC5314, lane 1; parental strain (CAI4), lane 2; and EAGA (CAI4 transformed with pEAGA), lane 3; probed for ADE2 sequences with a ³²P-labelled PCR amplicon. Bottom panel: equal loading of total RNA was assessed by ethidium bromide staining of rRNA bands separated in standard formaldehyde gels used in northern blot analysis

Figure 2

Detection of ADE2 dsRNA by RT-PCR. Total RNA from EAGA (lanes 2, 3, 6 and 7) and the parental strain, CAI4 (lanes 4, 5, 8 and 9), was reverse-transcribed and used in PCRs with a single oligonucleotide of ADE2 sequences (ADE2-35; lanes 2, 3, 4, and 5). A PCR amplicon was detected only in reactions with RT templates from EAGA in addition to a faster migrating band probably due to secondary structure (hairpin) of the inverted repeat DNA product (lane 2). Controls: no RT reactions to exclude DNA contamination (lanes 3, 5, 7 and 9), and the amplification of an actin amplicon (lanes 6, 7, 8 and 9) that served as positive control for the RT-PCR. M, DNA size markers (HyperLadder IV, Bioline). The migration of the 1000 and 500 bp bands are indicated on the left

Figure 3

Cell extract processing of blunt and hairpin dsRNA templates. A hairpin (ADE2-GFP-ADE2) or blunt (ADE2 sequences only) radiolabelled substrate was incubated with SC5314 cell extracts, and the product resolved in a 15% acrylamide, 7 M urea gel. Lanes 1, 2, 5 and 6 were incubated with the blunt dsRNA; lanes 3, 4, 7 and 8 were incubated with the hairpin dsRNA. Odd-numbered lanes lacked cell extract. The RNase reaction is dependent upon ATP hydrolysis (4 °C reactions). The migration of RNA size markers is indicated on the left (Decade Markers, Ambion/Applied Biosystems)

Figure 4

Expression analysis of C. albicans Argonaute (CaAGO, orf19.2903). (a) Total RNA from SC5314 was probed with radiolabelled CaAGO and ACT1 sequences by northern blotting. (b) Detection of CaAGO mRNA by RT-PCR. Amplification of a region of ACT1 served as positive control for the RT reactions. Reactions lacking RT (–RT) controlled for DNA contamination of the samples. M, DNA size markers (1 kb Plus DNA Ladder, Invitrogen). The identity of the size markers in base pairs are indicated on the left