ABSTRACT
We applied ribosomal DNA PCR-restriction fragment length polymorphism (RFLP) and mitochondrial DNA (mtDNA) RFLP analyses to 43 Acanthamoeba environmental isolates (KA/LH1 to KA/LH43) from contact lens storage cases in southwestern Korea. These isolates were compared to American Type Culture Collection strains and clinical isolates (KA/E1 to KA/E12) from patients with keratitis. Seven riboprint patterns were seen. To identify the species of the isolates, a phylogenetic tree was constructed based on the comparison of riboprint patterns with reference strains. Four types accounted for 39 of the isolates belonging to the A. castellanii complex. The most predominant (48.8%) type was A. castellanii KA/LH2 type, which had identical riboprint and mtDNA RFLP patterns to those of A. castellanii Castellani, KA/E3 and KA/E8. The riboprint and mtDNA RFLP patterns of the KA/LH7 (20.9%) type were identical to those of A. castellanii Ma, a corneal isolate from the United States. The riboprint and mtDNA RFLP patterns of the KA/LH1 (18.6%) type were the same as those of A. lugdunensis L3a, KA/E2, and KA/E12. The prevalent pattern for each type of Acanthamoeba in southwestern Korea was very different from that from southeastern Korea and Seoul, Korea. It is noteworthy that 38 (88.4%) out of 43 isolates from contact lens storage cases of the residents in southwestern Korea revealed mtDNA RFLP and riboprint patterns identical to those found for clinical isolates in our area. This indicates that most isolates from contact lens storage cases in the surveyed area are potential keratopathogens. More attention should be paid to the disinfection of contact lens storage cases to prevent possible amoebic keratitis.
Acanthamoeba spp., causative agents of the sight-threatening amoebic keratitis and the life-threatening granulomatous amoebic encephalitis, have ubiquitous distribution in human environments (35, 45). Amoebic keratitis was a rare corneal infection for 10 years after the first case report by Jones et al. (23). However, the number of reported cases has increased, apparently as a consequence of an association of the disease with contact lens wearing (36, 57). Amoebic keratitis has been continuously reported in Korea as well (unpublished data). Ecological studies (22, 30) showed that the contamination rate of contact lens paraphernalia by Acanthamoeba in Korea was much higher than rates reported in industrialized countries (24, 29, 56). In particular, the rate in southwestern Korea was twice that found in other provinces in Korea. The pathogenic potential of Acanthamoeba isolates from lens storage cases in Korea has yet to be determined. Because no simple and reliable in vivo or in vitro test exists, investigators have been using strain typing methods in order to indirectly assess the potential pathogenicity of the isolates (19, 49, 60).
Although the identification of Acanthamoeba at the genus level can be easily accomplished on the basis of the distinctive morphology of the trophozoites and cysts, there have been disputes over methods of species identification for Acanthamoeba. Pussard and Pons (42) classified Acanthamoeba into three groups according to the cyst size and morphological features. Group I consists of Acanthamoeba spp. with relatively large cysts, distinctly stellate endocysts, and smooth spherical ectocysts. Group II and group III Acanthamoeba spp. have smaller cysts (diameters less than 18 μm). Group II species have polygonal to stellate endocysts with irregular or wrinkled ectocysts, while the cysts of group III species have rounded or slightly angular endocysts with thinner and smooth or slightly wrinkled ectocysts. However, because of variable cyst morphology by culture conditions (53), species identification by morphology alone can hardly be possible (54).
Therefore, investigators have used several kinds of nonmorphological methods for the taxonomy of Acanthamoeba spp. Studies of isoenzyme patterns (11, 12, 14, 15, 55) suggested the other groupings of Acanthamoeba strains that are not consistent with previous species assignments based on morphological criteria. Kong et al. (28) and Chung et al. (6) reported intraspecific heterogeneity of the zymograms for several kinds of isoenzymes. Costas and Griffiths (13) considered that the variation among some Acanthamoeba species overlaps so extensively that the amoebae should be regarded as species complexes. Mitochondrial DNA (mtDNA) restriction fragment length polymorphism (RFLP) analysis has also been applied for taxonomy analysis of Acanthamoeba (2, 3, 58, 59). However, considering the profound interstrain diversity of the mtDNA RFLP and alloenzyme profiles, Kong et al. (28) and Chung et al. (6) suggested that both analyses should be used for the strain identification, differentiation, and characterization rather than species identification.
The sequence of the small-subunit (ssu) rRNA gene is very useful as molecular data for phylogeny and taxonomy (5, 21, 33, 52). Stothard et al. (52) studied 18S (ssu) rRNA gene phylogeny of Acanthamoeba and classified 53 isolates into 12 sequence types. However, when the number of isolates to be studied is large, sequencing of the 2,300-bp rRNA gene is too labor-intensive, time-consuming, and expensive for identification or characterization of Acanthamoeba isolates. Thus, a simpler and less expensive method is needed. Riboprinting, the examination of restriction enzyme site polymorphism of ssu rRNA-coding DNA (rDNA) amplified by PCR, is a simple, inexpensive, and timely method and has recently been used to establish the taxonomic relationships among Entamoeba spp. (9, 10), among Naegleria spp. (8, 16), and among Acanthamoeba spp. (5, 27). Chung et al. (5) were the first to apply riboprinting to the subgenus classification of Acanthamoeba. Their results coincided well with those of rRNA sequencing performed by Stothard et al. (52). Acanthamoeba taxonomists generally agreed with the classification that Chung et al. (5) had suggested (T. J. Byers, personal communication). By either rDNA sequencing or riboprinting, most clinical isolates of morphological group II were shown to belong to a clade (T4 by Stothard et al., A. castellanii subgroup by Chung et al.) containing A. castellanii Castellani. Stothard et al. (52) suggested that various species in T4 all might be reclassified as A. castellanii because the sequence type includes the type strain for that species.
In the present study, we analyzed 43 Acanthamoeba isolates from contact lens storage cases of the residents of southwestern Korea using riboprinting and mtDNA RFLP analyses and evaluated the potential keratopathogenicities of these isolates.
MATERIALS AND METHODS
Acanthamoeba isolation and axenic culture.In order to isolate Acanthamoeba from contaminated contact lens storage cases, the cases were opened under aseptic conditions. The centrifuged sediment from the solution in a case was aseptically smeared on a nonnutrient agar plate covered with heat-treated Escherichia coli (60). A sterile cotton ball was then rubbed over the internal surface of the lens case and placed on a different agar plate covered with heat-treated E. coli. The plates were incubated at 25°C for 1 week and examined daily for the presence and growth of Acanthamoeba under an inverted microscope.
The Acanthamoeba cysts encysted on the plate were grouped by their size and morphological features according to the criteria of Pussard and Pons (42).
A piece of agar plate (0.5 by 1 cm) covered with the cysts of a clone was treated with 0.1 N HCl for 24 h for axenization and washed with glass-distilled water three times. The piece of agar plate with many cysts was placed in Proteose Peptone-yeast extract-glucose medium (60) at 25°C.
Reference strains.Twenty-three Acanthamoeba strains (Table 1) were obtained from the American Type Culture Collection (ATCC) (Manassas, Va.). Twelve clinical isolates (KA/E1 to KA/E12) from Korean keratitis patients were also used as reference strains.
List of Acanthamoeba reference strains obtained from the ATCC
Riboprinting: PCR-RFLP of the ssu rDNA.Chromosomal DNA of each clone was extracted as described by Kong and Chung (27). Amoebae harvested at the end of the logarithmic growth phase were washed with cold phosphate-buffered saline three times and boiled in 0.1 ml of 0.1 N NaOH for 3 min. The supernatant was collected after centrifugation at 800 × g for 2 min at room temperature and mixed with 0.2 ml of glass-distilled water. An equal volume of phenol was added to the solution, and this mixture was vortexed for 1 min. The mixture was centrifuged at 15,000 rpm for 5 min at 4°C twice. The resulting aqueous phase was centrifuged at 12,000 rpm for 5 min again with 300 μl of phenol-chloroform (1:1) solution. The nuclear DNA was precipitated by adding 600 μl of cold absolute ethanol and 30 μl of 3 M sodium acetate (pH 5.2), incubating this mixture at −70°C for 15 min, and then centrifuging it at 15,000 rpm for 20 min at 4°C. The sediment was washed with 70% ethanol, vacuum dried, and then dissolved in 30 to 50 μl of glass-distilled water. The DNA was stored at −20°C until used.
The sequence of primers for PCR ssu rDNA (P3, 5′-CCGAATTCGTCGACAACTGTTGATCCTGCCAGT-3′; P4, 5′-GGATCCAAGCTTGATCCTTC TGCAGGTTCACCTAC-3′) are designed to hybridize at the highly conserved sequences of the extreme 5′ (P3) and 3′ (P4) termini of eukaryotic ssu rDNA (1, 5). PCR consisted of 1 min at 94°C, 1 min at 58°C, and 2 min at 72°C in a thermal cycler (PE480; Perkin-Elmer Cetus). After 30 cycles, 10 min of extension time was given. After amplification, the PCR products of 43 Acanthamoeba isolates were checked by electrophoresis in a 1.5% agarose gel at 4 V/cm for 1.5 h. The amplified ssu rDNA was examined by digestion with 10 restriction endonucleases (MspI, HaeIII, HhaI, HinfI, DdeI, TaqI, Tru9I, MboI, RsaI, and Sau96I; Poscochem, Seongnam, Korea) that have recognition sequences of four nucleotides for 2 h. The digested DNA was electrophoresed in a 2.5% agarose gel (3 parts agarose, 1 part Nusieve). When the enzyme digested the DNA into small and similarly sized fragments, 15% polyacrylamide gel with Tris-borate-EDTA buffer was used to obtain clearer and more-accurate data of fragments. The ethidium bromide-stained gel was examined and photographed under a UV transilluminator.
Sequence divergence estimates were calculated by the Nei and Li equation (40) from the fragment comigration data set which was obtained by comparison of the riboprints of 43 isolates and all reference strains (5). A phylogenetic tree was constructed by the unweighted pair group method with arithmetic average using a computer program (PHYLIP, version 3.5) (7, 18).
mtDNA extraction and RFLP.mtDNA of Acanthamoeba isolates was extracted by the method described by Yagita and Endo (59). Briefly, Acanthamoeba trophozoites washed with phosphate-buffered saline (pH 7.4) were suspended in 100 μl of chilled TEG buffer (25 mM Tris-HCl, 10 mM EDTA, 50 mM glucose; pH 8.0) and incubated on ice for 5 min. Amoebae were lysed by adding 200 μl of chilled fresh 1% sodium dodecyl sulfate solution in 0.2 N NaOH and then incubated subsequently on ice for 5 min. Chilled 3 M potassium acetate buffer was mixed with the suspension and incubated on ice for 15 min. Degraded cellular proteins and genomic DNA were extracted with equal volumes of phenol-chloroform by centrifugation at 12,000 rpm at 4°C for 5 min. mtDNA in supernatant was precipitated by adding 1.0 ml of absolute ethanol and 40 μl of 3 M sodium acetate solution and by incubating at −70°C for 15 min. After centrifugation at 15,000 rpm for 20 min at 4°C, the precipitated DNA was washed with 70% chilled ethanol. The DNA sediment was vacuum dried and dissolved in 15 to 25 μl of TE buffer (5 mM Tris-HCl, pH 8.0; 1 mM EDTA) and stored at −20°C until used. mtDNAs of 43 Acanthamoeba isolates were digested with restriction enzymes that have recognition sequences of six nucleotides at 37°C for 2 h (sometimes overnight) in a 20-μl reaction volume with the buffers specified for each restriction enzyme (EcoRI, BglII, XbaI, SalI, ScaI, ClaI, HpaI, and PvuII; Poscochem). Digested DNA was electrophoresed in a 0.7% agarose gel and stained with ethidium bromide. The RFLP of mtDNA of 43 strains was observed and photographed under a UV transilluminator. The HindIII-digested λ phage was used as a size marker.
RESULTS
Morphology of Acanthamoeba isolates.Among 43 isolates analyzed in this study, 42 isolates belonged to the morphological group II of Pussard and Pons (42). In their system, the cyst has a polygonal endocyst and an irregularly wrinkled thick ectocyst, with the cyst diameter ranging from 11.8 to 16.5 μm and the number of arms ranging from 3.8 to 6.2. One other isolate (KA/LH35) belonged to group III, which features round endocysts closely attached to wrinkled ectocysts.
Analysis of ssu rDNA PCR-RFLP.Figure 1 presents PCR-amplified full-length rDNA and RFLP patterns obtained by using three kinds of restriction enzymes. The size of the PCR products was approximately 2,300 bp in all isolates, meaning no intron was present in the ssu rDNAs of 43 environmental isolates. Considering all patterns by eight kinds of restriction enzymes (some data not shown), a total of seven different rDNA RFLP types emerged. Using the phylogenetic tree of the isolates with reference strains based on the estimated sequence divergence, 41 isolates were identified as A. castellanii complex, and 1 was identified as A. polyphaga (Fig. 2), but the remaining isolate has yet to be studied. Among the 41 isolates belonging to A. castellanii complex, 8 isolates (KA/LH1 type) showed the same rDNA PCR-RFLP pattern as A. lugdunensis L3a. Twenty-one isolates (KA/LH2 type) showed patterns of rDNA PCR-RFLP very similar to those of A. castellanii Castellani, the type strain of the type species. Nine isolates (KA/LH7 type) were identical to A. castellanii Ma in their rDNA PCR-RFLP pattern. The pattern of one isolate (KA/LH32 type) was identical to that of A. castellanii Castellani. An isolate (KA/LH35 type) with morphological characteristics of group III showed a unique rDNA PCR-RFLP pattern. Two isolates (KA/LH39 type), although they belonged to A. castellanii complex, showed a rDNA PCR-RFLP pattern somewhat different from that of the Castellani strain. One isolate (KA/LH8 type) clustered with the A. rhysodes clade.
Agarose gel electrophoretic pattern of PCR products from seven different types of Acanthamoeba and their polyacrylamide gel electrophoretic restriction fragment patterns. Lanes M, HindIII-digested λ phage DNA or 100-bp ladder (DNA size standards).
Phylogenetic tree of seven different types and 23 reference strains of Acanthamoeba based on the rDNA PCR-RFLP analyses obtained by using the unweighted pair group method with arithmetic average and a computer program (PHYLIP, version 3.5).
mtDNA restriction phenotypes.Figure 3 shows agarose gel electrophoretic patterns of EcoRI digests of mtDNA from 43 Acanthamoeba isolates. They were divided into seven different types according to their patterns. Although KA/LH2 and KA/LH7 types appeared to have almost the same patterns by EcoRI digestion, as seen in Fig. 3, the fragment sizes were slightly different. Moreover, they showed very different RFLP patterns with other restriction enzymes, such as BglII (Fig. 4A). The most predominant type was KA/LH2, and 21 isolates showed the same mtDNA RFLP pattern. Nine isolates were of the KA/LH7 type, and eight isolates were of the KA/LH1 type. Two isolates, KA/LH39 and KA/LH40, had the same mtDNA RFLP pattern as each other. The mtDNA RFLP pattern of KA/LH32 was very similar to that of KA/LH2 but slightly different. The other two isolates, KA/LH8 and KA/LH35, had unique mtDNA RFLP patterns. A total of seven different restriction phenotypes by BglII enzymes emerged for 43 isolates, as shown in Fig. 4A. The other three kinds of restriction enzymes we tested showed the same patterns (data not shown).
Agarose gel electrophoretic restriction fragment patterns by EcoRI of mtDNA of 43 environmental Acanthamoeba isolates. Lanes M, HindIII-digested λ phage DNA (DNA size standard).
Agarose gel electrophoretic restriction fragment patterns of mtDNA by BglII restriction enzyme. Seven different types of 43 environmental Acanthamoeba isolates (A), three types of environmental isolates and associated reference strains of Acanthamoeba (B), and two types of environmental isolates and associated clinical isolates of Acanthamoeba from keratitis patients in Korea (C) were compared. Lanes: M, HindIII-digested λ phage DNA (DNA size standard); L3a, A. lugdunensis L3a; Castellani, A. castellanii Castellani; Ma, A. castellanii Ma.
DISCUSSION
In the present study, 43 Acanthamoeba isolates, originating from contact lens storage cases of residents in southwestern Korea, were morphologically and genetically analyzed based on riboprinting and mtDNA RFLP analysis. Except for one isolate (KA/LH35), all belonged to the morphological group II (42). These isolates may be morphologically assigned to the A. castellanii or A. polyphaga complexes. This indicates that Acanthamoeba of morphological group II may be predominant in contact lens storage cases in Korea. Based on riboprints, 41 isolates were genetically very closely related to A. castellanii Castellani, and 1 was closely related to A. polyphaga. The remaining 1 isolate showed a unique riboprint. This coincided well with the results of morphological observation and indicated that amoebae closely related to A. castellanii may be predominant in contact lens storage cases in Korea.
Although rDNA sequencing or riboprinting is promising for subgenus classification of Acanthamoeba (5, 19, 52), it is not simple to identify isolates at the species level, especially strains in group II, to which most clinical and environmental isolates belong. Stothard et al. (52) suggested that various species in T4 might be reclassified as A. castellanii because the sequence type includes the type strain for that species. However, the molecular characteristics of strains in T4 were quite different among strains, and many species which belonged to this clade have already been assigned. We would rather assign this subgroup (T4 sequence type) as a species complex (60), as Costas and Griffiths suggested (13).
Based on the riboprints and mtDNA RFLP analysis in the present study, the A. castellanii KA/LH2 type was the most predominant (48.8%) type of Acanthamoeba from contact lens storage cases in southwestern Korea (Table 2). It is noteworthy that the riboprint and mtDNA RFLP patterns of the KA/LH2 type were identical to those of A. castellanii Castellani, KA/E3, and KA/E8 (Fig. 4B and C)—the type strain and two clinical isolates from infected corneas of Korean patients, respectively. The predominance of the Castellani type of Acanthamoeba in southwestern Korea is unique compared to the previous survey results in southeastern and Seoul, Korea. The predominance of the Castellani type was 3.6% in the southeastern area (31) and 7.7% in Seoul (60).
Isolates of Acanthamoeba spp. from contact lens storage cases and RFLP phenotypes of their rDNA and mt-DNA
Acanthamoeba castellanii KA/LH7 (20.9%) and KA/LH1 (18.6%) types were the second and third most prevalent types of Acanthamoeba in the surveyed area. The riboprint and mtDNA RFLP patterns of the KA/LH7 type were identical to those of A. castellanii Ma (Fig. 4B), a clinical isolate from the United States (34). The Ma type has never been isolated in southeastern Korea (31) but was relatively prevalent (20.5%) in Seoul (60). The riboprint and mtDNA RFLP patterns of the KA/LH1 type were the same as those of A. castellanii L3a, KA/E2, and KA/E12 (Fig. 4B and C). The KA/LH1 type was reported to be the most prevalent type of Acanthamoeba in southeastern Korea (64.3%) and in Seoul (51.3%). Two amoebic keratitis patients from Seoul and Pusan were recently reported to be infected with this type (unpublished data).
The prevalent pattern for each type of Acanthamoeba in southwestern Korea was very different from those in the southeastern area. However, it was interesting that the pattern in Seoul was intermediate between those in the southeastern and southwestern areas, because the citizens in Seoul are composed of people from various provinces, including provinces in the southeastern and southwestern areas. The prevalence pattern can vary from nation to nation. Kilvington et al. (25) reported that isolates with the mtDNA RFLP type of A. castellanii Jones (formerly A. polyphaga or A. lugdunensis) were demonstrated the most frequently among clinical isolates of Acanthamoeba in Europe. According to Gautom et al. (20), isolates with the mtDNA restriction phenotype of the A. castellanii Castellani type were demonstrated the most frequently from clinical and environmental sources in the United States. In Japan, isolates with mtDNA RFLP patterns of the A. castellanii Ma type were demonstrated the most frequently from contact lens paraphernalia and infected corneas (T. Endo, personal communication).
The riboprint pattern of the KA/LH32 type was the same as that of the Castellani, KA/E3, and KA/E8 types, but the KA/LH32 type could not be regarded as a clone of the Castellani, KA/E3, or KA/E8 types but was regarded as a closely related clone, because the mtDNA RFLP pattern of the KA/LH32 type was slightly different from those of the reference strains. This finding can be explained by the fact that mtDNA evolves faster than nuclear DNA. Therefore, the mitochondrial genome can be used for phylogenetic study of closely related Acanthamoeba isolates. The mtDNA RFLP pattern of KA/LH32 was almost identical to that of JAC/E1, an isolate from a Japanese keratitis patient (58).
The most important fact in the present study is that 38 (88.4%) out of 43 isolates from contact lens storage cases of the residents of southwestern Korea revealed mtDNA RFLP and riboprint patterns identical to those of clinical isolates of Acanthamoeba. This indicates that most isolates in this study are potential keratopathogens. Four of the remaining five isolates were new types.
In contrast to the high contamination rate of contact lens paraphernalia by potentially keratopathogenic Acanthamoeba spp. in Korea, the incidence of Acanthamoeba keratitis so far is low. It is suggested that the combination of wearing a contact lens contaminated with keratopathogenic Acanthamoeba and minor injury of the corneal epithelium may be the predisposing factor in Acanthamoeba keratitis among contact lens wearers. At any rate, more attention should be paid to the disinfection of contact lens storage cases to prevent possible amoebic keratitis.
ACKNOWLEDGMENTS
This work was supported by a grant (HMP-97-2-0029 for the 1997, Good Health R&D Project) from the Ministry of Health and Welfare, Republic of Korea.
FOOTNOTES
- Received 6 September 2001.
- Returned for modification 8 October 2001.
- Accepted 2 December 2001.
- Copyright © 2002 American Society for Microbiology
