A. castellanii (ATCC 30011) was obtained from the American Type Culture Collection. Acanthamoeba trophozoites were cultured axenically in PYG medium, which contains proteose peptone 0.75% (w/v), yeast extract 0.75% (w/v), and glucose 1.5% (w/v) at 25℃ in a Sanyo incubator (San Diego, California, USA). Acanthamoeba cysts were induced in encystation media (0.1 M KCl, 0.008 M MgSO₄, 0.0004 M CaCl₂, and 0.02 M 2-amino-2-methyl-1,3-propanediol, pH 9.0) for 3 days [10]. Mature cysts were counted under a light microscope after treatment with 0.5% SDS to calculate encystation ratios [17].

Total RNA was purified using TRIzol Reagent (Gibco BRL, Rockville, Maryland, USA), and cDNA synthesis was conducted using a RevertAidTM First Strand cDNA Synthesis Kit (Fermentas, Hanover, Indiana, USA). Real-time PCR was performed using the ABI PRISM® 7000 sequence detection system (Applied Biosystems, Foster City, California, USA), using the default thermocycler program for all genes [9] and sense and anti-sense primers (sense 5'-TCATCTACATGTTCTGCGCCC and antisense 5'-CGATCCAGTTGTTGAGCATGC for cellulose synthase, and sense 5'-AGTACGAGATGCTCCTCAACCG and antisense 5'-TCTCGACAA GACGAAGAGGTCC for xylose isomerase). The cDNA sequence information of xylose isomerase was obtained from differentially expressed gene analysis [7]. All reaction mixtures were made using Sybr Premix Ex Taq (Takara, Otsu, Shiga, Japan). The 18S ribosomal DNA was used as the reference gene [9]. Real-time PCR was performed to determine relative gene expressions using the 2-ΔΔCT method, and experiments were performed in triplicate.

Small interfering RNA (siRNA) targeting cellulose synthase of A. castellaniii was synthesized by Sigma-Proligo (Boulder, Colorado, USA), based on its cDNA sequence. The siRNA duplex with sense (5'-GAUCGAGUACUUCAACAUCdTdT) and anti-sense (5'-GAUGUUGAAGUACUCGAUCdTdT) against cellulose synthase were used. The siRNA (4 µg) was transfected into A. castellanii trophozoites at a cell density of 4×10⁵ per well as previously described [4]. As a negative control, siRNA with a scrambled sequence (Ambion, Austin, Texas, USA) absent in Acanthamoeba was used.

Acanthamoeba were incubated in encystation media for 3 days with or without siRNA against cellulose synthase. After the induction of encystation, amoeba pellets were resuspended in 2.5% calcofluor white staining solution and incubated for 20 min at room temperature [13]. After washing with PBS, samples were observed under a fluorescence microscope.

Acanthamoeba were prefixed with 4% glutaraldehyde and post-fixed with 1% osmium tetroxide. Fixed cells were dehydrated with ethyl alcohol, treated with propylene oxide-resin, and embedded in resin. Ultrathin sections were cut and stained with uranyl acetate and lead citrate [9]. Samples were observed under a Hitachi H-7000 transmission electron microscope (Hitachi, Tokyo, Japan).

Data are expressed as the mean±SEM. Statistical significance was analyzed by an unpaired Student's t-test. A P-value of <0.01 was interpreted as statistically significant.

Gene silencing by siRNA was used to determine the involvement of cellulose synthase (CS) in encystation of Acanthamoeba. The mRNA level of cellulose synthase was highly expressed at day 3 after induction of encystation (Fig. 1A-). After transfection of CS-siRNA, a significant decrease in the mRNA expression of cellulose synthase was observed at day 3 after induction of encystation (Fig. 1A-), showing that the encystation of Acanthamoeba was significantly inhibited by silencing of cellulose synthase (Fig. 1B). Scrambled negative siRNA had no effects on Acanthamoeba encystation.

To determine the effects of gene silencing of cellulose synthase (CS) on encystation of Acanthamoeba, calcofluor white staining and transmission electron microscopy were performed. As shown in Fig. 2A, calcofluor staining was not observed in trophozoites. Interestingly, at day 3 post-induction, 2 different patterns of calcofluor positive materials were observed on cyst surfaces (Fig. 2B). Young cysts had a single layer of calcofluor stained, whereas mature cysts exhibited a double layer of calcofluor stained cells. An endocyst, containing a larger amount of cellulose than an ectocyst, showed a strong staining reaction (Fig. 2B, arrows). In CS-siRNA transfected amoeba, single layered young cysts predominated at day 3 post-induction, whereas in non-transfected amoeba, double layered mature cysts were observed (Fig. 2C).

To confirm the inhibition of cyst maturation by CS-siRNA, normal mature cysts and CS-siRNA transfected cysts were observed under a transmission electron microscope (TEM). After normal induction of encystation, Acanthamoeba formed mature cysts with double cyst walls (Fig. 3A). Ectocysts were deeply wrinkled and electron-dense, whereas smooth-organized endocysts were closely associated with the inner cellular membrane. In contrast, CS-siRNA transfected Acanthamoeba showed significant impairment of cyst morphology. Cells were shrunken and lost contact with cyst walls, which lost their compactness, and ectocysts were lightly stained with few or no wrinkles. Furthermore, endocysts were not observed. The majority of CS-siRNA transfected Acanthamoeba remained as single walled young cysts or died (Fig. 3B).

Real-time PCR was conducted to examine the effects of CS-siRNA on expression of xylose isomerase in Acanthamoeba. Silencing of cellulose synthase during the encystation of Acanthamoeba had no effects on expression of xylose isomerase (Fig. 4). In fact, the expression pattern of xylose isomerase in CS-siRNA transfected amoeba was similar to that in normal untransfected controls.