Acanthamoeba Sequence Types and Allelic Variations in Isolates from Clinical and Different Environmental Sources in Italy

The genus Acanthamoeba comprises free-living amoebae distributed in a wide variety of environments. These amoebae are clinically significant, causing opportunistic infections in humans and other animals. Despite this, limited data on Acanthamoeba sequence types and alleles are available in Italy. In the present study, we analyzed all Acanthamoeba sequences deposited from Italy with new positive Acanthamoeba clinical samples from symptomatic AK cases, to provide an overview of the genetic variants’ spatial patterns from different sources within the Italian context. A total of 137 Acanthamoeba sequences were obtained. Six sequence types were identified: T2/6, T3, T4, T11, T13, and T15. Only T4 and T15 were found in both sources. The Acanthamoeba T4 sequence type was found to be the most prevalent in all regions, accounting for 73% (100/137) of the Italian samples analyzed. The T4 sequence type demonstrated significant allelic diversity, with 30 distinct alleles from clinical and/or environmental samples. These outcomes enabled a better understanding of the distribution of Acanthamoeba isolates throughout Italy, reaffirming its well-recognized ubiquity. Acanthamoeba isolates analysis from keratitis, together with the environmental strains monitoring, might provide important information on different genotypes spreading. This might be useful to define the transmission pathways of human keratitis across different epidemiological scales.


Introduction
The genus Acanthamoeba (Amoebozoa, Discosea) includes ubiquitous free-living amoebae (FLA) widely distributed in environments such as soil, fresh and marine water, and air.These amoebae have the potential to cause opportunistic infections in humans and other animals, like amoebic keratitis (AK) or granulomatous amoebic encephalitis (GAE).As recently reviewed [1], morphological approaches were primarily used in classifying species/taxa within the genus.Initially, Acanthamoeba was subdivided into three distinct groups (I-III) [2].However, the identification of species remained arduous, and the advent of molecular tools, especially methods of analysis based on DNA differences, provided new insights into the relationships between the isolates [3][4][5].
The classification of Acanthamoeba is now mainly based on the analysis of the nuclear SSU rRNA gene (18S rDNA) variability-only partially confirming the morphological classification-and 23 sequence types, also named genotypes (from T1 to T23), have been identified so far, phylogenetically grouped into four major lineages [6].In particular, the widely employed ASA.S1 region, a ~430 bp fragment matching the most variable segment of the 18S rDNA as proposed by Schroeder et al. [7], has proved to be a helpful fragment of the gene in the study of Acanthamoeba genetic variability, in the evolutionary relationships between the isolates, and in the definition of the different genotypes within the genus [1].
Along with the increased data on the sequence differences within this hypervariable segment, a different degree of intra-genotype variation was also observed among Acanthamoeba populations.Specifically, numerous studies have been conducted by examining polymorphisms in the DF3 region within the amplimer ASA.S1, yielding a sequence approximately 200-250 bases in size that includes most of the highly variable portions of this region, in order to better describe and categorize the genetic variation that exists in this region.These studies allowed the identification of numerous "alleles" within different sequence types, with the T4 genotype being the most variable.In addition to the T4, data on allele identification from a small number of genotypes-T3, T5, T11, T15, and supergroup T2/6-are now also available [1].The introduction of the 'allele' concept in 2002 by Booton et al. [8].marked a fundamental step in examining Acanthamoeba in both clinical and environmental contexts.
As stated, Acanthamoeba can act as an opportunistic parasite in humans, causing different infections, including granulomatous amoebic encephalitis, cutaneous lesions and sight-threatening amoebic keratitis.From a public health perspective, the molecular identification of Acanthamoeba isolates from clinical and environmental samples is therefore of great interest.Overall, the majority of isolates in clinical or environmental studies belong to the T4 genotype, which was further divided into seven main subtypes, labeled T4A to T4G based on phylogenetic analyses of 18S rRNA sequences for complete or almost complete sequences (with a length exceeding 2000 bases in length) [1,9].Recently, a new nuclear lineage within the Acanthamoeba T4 genotype, labeled T4H, has been described [10].
The clustering of Acanthamoeba isolates from clinical cases coupled with the tracking of strains in the environment through high-resolution genotyping at the sequence type and allele levels could provide insights into the spread of different genetic variants, which would be useful in better identifying the transmission pathways of human keratitis at different epidemiological scales.
In Italy, few and geographically fragmented data are available on Acanthamoeba sequence types distribution from clinical [11][12][13] and environmental samples [14][15][16][17].In the present study, sequence data retrieved from GenBank were compared with new positive Acanthamoeba clinical samples from symptomatic AK cases identified by phylogenetic analyses, to investigate the genetic variation within the ASA.S1 fragment in isolates from different sources in Italy.The specific objectives included: (i) adding new data on the distribution of Acanthamoeba sequence types from humans; (ii) conducting for the first time a comprehensive analysis of allelic variations in sequences obtained from all clinical and environmental isolates accessible in Italy; and (iii) providing a wide picture of the spatial patterns of genetic variants in isolates from various sources within the Italian context.

New Clinical Sample Collection and Molecular Characterization
Before conducting the analysis of the allelic variations in Acanthamoeba sequences from all clinical and environmental isolates accessible in Italy, the present work contributed to producing new original clinical sequences.In particular, the study involved corneal scraping specimens from symptomatic patients with suspected AK collected between 2015 and 2022 in the Unit of Parasitology of the Azienda Ospedaliera Universitaria Policlinico Tor Vergata (PTV) of Rome, Central Italy.To accurately identify the isolates and obtain information about Acanthamoeba sequence types and allelic variations, amplification of a fragment within the 18S rRNA gene containing the ASA.S1 region was performed.DNA was extracted from each sample using the Starlet extraction automate (SeeGene, Seoul, Republic of Korea) and used for molecular characterization.The PCR reaction mix encompassed a total volume of 25 µL, which contained 12.5 µL PCR master mix 2X (Promega, Milano, Italy), 5 µL of template DNA, and 0.5 µL of the primer pair JDP1 (5 ′ -GGCCCAGATCGTTTACCGTGAA) and JDP2 (5 ′ -TCTCACAAGCTGCTAGGGAGTCA-3 ′ ) to amplify a PCR product of ca.423 to 551 bp [7].The cycling conditions included one step at 96 • C for 2 min, followed by 35 cycles of denaturing for 1 min at 96 • C, annealing for 1 min at 60 • C, and extension for 1 min at 72 • C, with a final extension for 7 min at 72 • C, performed in a T100 PCR thermal cycler (BioRad, Segrate, Italy).Negative and positive controls were included in each batch of DNA extraction and PCR reaction.PCR products were examined by electrophoresis on 1% agarose gel stained by SYBR™ Safe DNA Gel Stain (Invitrogen™, Waltham, MA, USA).Amplicons were purified by the mi-PCR Purification Kit (Metabion GmbH, Steinkirchen, Germany) according to the manufacturer's instructions, and then sent to an independent laboratory for sequencing (Bio-Fab Research, Rome, Italy).Forward and reverse Sanger sequences from each PCR product were examined by DNA chromatograms using FinchTV 1.4 (Geospiza, Inc, Seattle, WA, USA), and the obtained consensus sequences were aligned by the latest version of MEGA 11 [18], with representative reference sequences used as provided on the website: "https://u.osu.edu/acanthamoeba/acanthamoeba-sequence-types/ (accessed on 15 December 2023)".To identify isolates at the sequence type level, a Maximum Likelihood (ML) phylogenetic tree was generated.The construction of the phylogenetic tree was conducted after testing for the best evolutionary models explaining the data [18].Bootstrapping with 1000 replicates was used to determine support for the genetic distances.

Data Deposition
The new clinical sequences obtained in the present study are available at NCBI Gen-Bank with the accession numbers PP126046-PP126065.

Database Development
For the purposes of this study, a search in PubMed Nucleotide of all Acanthamoeba sequences deposited from Italy was performed using a multistep strategy matching the following keywords: Acanthamoeba AND Italy OR Italian; FLA AND Italy OR Italian; and free-living amoeba AND Italy OR Italian.The full check of GenBank records was performed paying attention to the "country" and "isolation_source" features.This allowed the retrieval of all sequences referring to environmental and clinical samples described in Italy and suitable for the analysis conducted in this study.

Allelic Identification
Alleles variability is found within the 18S rRNA region of the sequences named ASA.S1.Hence, to properly identify the isolates at allele level, all Acanthamoeba sequences recovered by the search described above plus those obtained in the present study from new clinical specimens (see Section 2.1) were appropriately aligned and trimmed by both MEGA 11 and AliView [19], with alleles representative sequences provided as lists on the website: "https://u.osu.edu/acanthamoeba/alleles-within-sequence-types-2/(updated October 2023) (accessed on 15 December 2023)".

Identification of New Clinical Isolates and Sequence Types Analysis
In the present study, twenty Acanthamoeba isolates from suspected AK patients were obtained and successfully PCR-amplified.Good-quality sequences of ~405-bp length were achieved for all samples.The topology of the ML phylogenetic tree showed all the isolates analyzed here clustering in the composite T4 sequence type clade (Figure 1).
Through the search strategy applied as described in Section 2.3, coupled with the acquisition of 20 new clinical sequences from this study, a total of 137 Acanthamoeba sequences linked to isolates from Italy were obtained.Details for each sequence are presented in Table 1, encompassing all sequences related to clinical isolates, and in Table 2, including sequences derived from environmental samples.

Allele Identification and Analysis
To provide a more precise depiction of the variation degree present within the Acanthamoeba isolates, each sequence from all sequence types, excluding T13 (as allele analysis for this genotype is not yet available), was identified at the allele level, following the procedures outlined in Section 2.4.
As illustrated in Figure 4, allele analysis revealed the presence of a single variant within the sequence types T3 (allele T3/03) and T11 (allele T11/08), which both originated from AK cases.Three isolates from soil samples in Piedmont belonging to the T2/6 supergroup were identified as alleles T26A/01 (one isolate) and T26B/01 (two isolates).Two alleles, T15/01 and T15/02, were also identified within the T15 sequence type.The T15/01 allele was assigned to the majority of the T15 isolates (18/21; 85.7%) and obtained from both human and water sources, all from the Apulia Region.On the other hand, the T15/02 allele was derived from three hot spring water samples from Lazio (3/21; 14.3%).
In contrast, isolates grouped in the T4 sequence type exhibited a substantial allelic diversity, with 30 different alleles assigned to isolates from clinical and/or environmental samples.Noteworthy, three environmental isolates, one from thermal water in Lazio, one from tap water collected in Apulia, and one from soil in Sardinia, exhibited a mixed allele combination (T4/01-T4A + MT4/25-T4B; OT4/114-T4A + OT4/143-T4A; and T4/07-T4A + Two alleles, T15/01 and T15/02, were also identified within the T15 sequence type.The T15/01 allele was assigned to the majority of the T15 isolates (18/21; 85.7%) and obtained from both human and water sources, all from the Apulia Region.On the other hand, the T15/02 allele was derived from three hot spring water samples from Lazio (3/21; 14.3%).
In contrast, isolates grouped in the T4 sequence type exhibited a substantial allelic diversity, with 30 different alleles assigned to isolates from clinical and/or environmental samples.Noteworthy, three environmental isolates, one from thermal water in Lazio, one from tap water collected in Apulia, and one from soil in Sardinia, exhibited a mixed allele combination (T4/01-T4A + MT4/25-T4B; OT4/114-T4A + OT4/143-T4A; and T4/07-T4A + T4/16-T4A, respectively).These combinations were identified based on the polymorphisms reported in the FASTA sequences.
The most commonly identified allele was T4/01-T4A (22/100; 22%), which was predominantly present in AK specimens (20/80; 25%) and only encountered twice in environmental samples (2/20; 10%).The AK isolates were derived from various studies, including the present one, conducted on clinical samples from Lazio and Apulia.The two environmental samples were collected from thermal water sources in Lazio; one of the two samples showed a mixed allele combination.
The other two allele types founded most frequently were T4/13-T4A (9/100; 9%) and AKT4/22 T4A (8/100; 8%).The first, identified in clinical samples from Lazio (n = 6) and Lombardy (n = 2), has been also assigned to one isolate from tap water collected in the Lazio region.The second allele was identified in AK patients from Lombardy (n = 2), Apulia (n = 1), and Lazio (n = 2), and in ornamental fountain water and tap water from Apulia (n = 3).Three alleles, each with a frequency of 6%, have been identified: T4/09-T4B, predominantly observed in clinical samples; ZT4/24-T4D, largely assigned to environmental samples; and T4/06-T4B, exclusively found in AK isolates, including some obtained in the present study.The remaining 24 alleles, each with a frequency less than or equal to 5%, comprised 6 alleles shared between human and environmental samples, 14 alleles exclusively found in clinical isolates, and 4 alleles detected solely in environmental samples.
Lastly, for three Acanthamoeba isolates (two obtained from soil in Sardinia and one from AK in Lazio), it was not possible to assign the allele, as the sequence did not match with any deposited allele segment thus far.

Discussion
Over the past decades, an increasing number of studies on Acanthamoeba distribution and genetic diversity have been conducted worldwide [5,7,8,15,[20][21][22].These studies primarily contributed to advancements regarding classification and phylogenetic relationships between Acanthamoeba isolates, but they also provided insight into different aspects of ecology and spatial distribution of these organisms, useful for determining potential threats to human health.In fact, despite being free-living organisms, several amoebae within the genus Acanthamoeba exhibit an "amphizoic" behavior.This means they are capable of transforming into an endozoic parasitic stage, causing serious and sometimes fatal diseases in humans [23].Transmission to humans or other animals typically begins through exposure to contaminated water or soil containing the trophozoite stage [24].Therefore, a comprehensive knowledge of the distribution of this organism in both environmental and clinical samples is crucial for accurately assessing the risk of infection.
In Italy, since 2008, studies on the molecular characterization of Acanthamoeba allowed the identification of various genotypes in isolates from different sources.However, there are no investigations that simultaneously analyzed isolates of both clinical and environmental origin within the same geographic area.Due to the limited and fragmented data so far available, this study is designed to serve as a comprehensive large-scale survey regarding sequence type distribution and allelic variations of Acanthamoeba in isolates from various sources and different geographic locations across Italy, aiming to give a more comprehensive overview of their diffusion and to offer insight into the potential transmission dynamics of Acanthamoeba within the country.
As summarized in Tables 1 and 2, the presence of Acanthamoeba in Italy has been documented in clinical specimens and/or in different environments with six sequence types, T2/6 (supergroup), T3, T4, T11, T13, and T15 detected so far.It is noteworthy that all these genotypes have been described to cause AK worldwide [25].
The twenty new clinical samples identified in this study were all assigned to T4, thus providing further support to the evidence that the T4 genotype plays a prominent role in the epidemiology of human infections, mainly AK.Taking into account these new clinical cases, as evidenced in Tables 1 and 2, the T4 genotype represents 73% (100/137) of the counted Italian samples, 58.4% (80/137) from humans, and 14.6% (20/137) from the environment, distributed across all the examined regions.This finding is in agreement with data from the literature, emphasizing that the majority of isolates in clinical or environmental studies belong to the T4 sequence type [1,7,20,21].This genotype, also defined as A. castellanii complex, encompasses several morphological species, including the type species of the genus A. castellanii.High intra-genotypic variation is therefore observed, leading to its subdivision into eight main subtypes and to the definition of a high number of alleles accepted to date (https://u.osu.edu/acanthamoeba/alleles-within-sequence-types-2/;updated October 2023) (accessed on 15 December 2023).
The high level of polymorphism here detected within the T4 sequence type, as evidenced by the identification of thirty different alleles with different patterns of SNPs, is therefore not unexpected.Among these, twelve alleles were documented to be shared between clinical and environmental samples.The most frequently observed was T4/01-T4A (22%).In the literature, the T4/01-T4A allele type was initially identified in Acanthamoeba keratitis patients in Hong Kong [8].Some years later, it has also been found in samples collected from drinking water treatment plants in Spain [22].However, in contrast to Italian data, this allele is detected with low frequency worldwide, as recently reviewed [1].The other two frequently encountered allele types in both human and environmental isolates were T4/13 and AKT4/22.Both of these alleles have been reported in over 100 deposited isolates worldwide, with AKT4/22 being the most prevalent, found in more than 240 isolates from many countries [1].
Notably, twenty-four low-frequency T4 allele types, each occurring at less than or equal to 5% frequency, dominate the distribution in Italy.They were predominantly observed as single cases in AK patients or, in four cases, assigned to mixed Acanthamoeba samples from grassland and rice field soil and from tap water.As previously mentioned [1,21], these ambiguous sequencing reads (mixed) may derive from a multi-cyst/multi-trophozoite extraction, representing an unrecognized mixture of isolates.Additionally, this ambiguity could be attributed to intracellular polymorphism in the 18S rRNA gene.
Among clinical reports, the only no-AK sample derived from a case of disseminated Acanthamoeba infection in an immunocompromised patient was attributed to the allele OT4/39.This allele, labeled since 2015, has significant importance as it identifies the type isolate for the genus Acanthamoeba (A. castellanii ATCC 30011) [1].Unfortunately, no further information is available on this case as the data are unpublished.
The second most abundant genotype described in Italy was T15.Like T4, this sequence type was found in both clinical and environmental samples in two regions, Lazio and Apulia.Until 2009, the T15 genotype had only been detected in a few isolates attributed to the species A. jacobsi, isolated from different environments [5].The first case of keratitis associated with this genotype was identified in Italy [11], thus enlarging the spectrum of Acanthamoeba sequence types capable of causing pathogenic effects in humans.Currently, the T15 genotype represents the third most involved genotype in keratitis cases after T4 and T3 [25] and has a notable worldwide environmental presence, as recently reviewed [26].Remarkably, the majority of T15 Acanthamoeba isolates occurred in Italy in both clinical and environmental isolates from Apulia and shared the identical allele sequence T15/01.This allele is reported to have the highest frequency (56%) among the alleles found within T15 sequences in the DNA databases:"https: //u.osu.edu/acanthamoeba/alleles-within-sequence-types-t15/;updated October 2023 (accessed on 15 December 2023)".Conversely, three isolates from Lazio thermal water, showing identical sequences to each other, were identified as T15/02.It is worth noting that the almost complete Rns sequence obtained subsequently for one of these isolates (Pool-4-37; Accession number KY513796) revealed a group I intron in the 18S ribosomal region, expanding the recognized genotype of Acanthamoeba with nuclear introns [27].
The remaining four sequence types, T2/6, T3, T11, and T13, were reported with lower frequency in various Italian regions, either in clinical or environmental isolates.
Sequence type T3, reported at 8%, was identified in 11 ocular isolates, all originating from Apulia and presenting the same allele T3/03.In the recent systematic review of Acanthamoeba in keratitis by Diehl et al. [25], T3 represents the second most prevalent genotype detected worldwide.It was detected for the first time by Gast et al. [4] by a strain identified in the marine amoeba A. griffini [28] and described primarily as a human pathogen associated with AK in a patient from the United Kingdom by Ledee et al. [29].In Italy, sequence type T3 was also previously described in one patient presenting with a wide corneal ulcer in Tuscany [30].However, since no deposited sequence is available, this Acanthamoeba isolate was excluded from our analysis.
Finally, two sequence types have been sporadically identified in Italy.The first, the T13 sequence type, refers to a single isolate found in one environmental sample from Sardinia but also isolated by the same authors in two samples from soils at high altitudes in Tibet [15].This genotype was only encountered once in a case of amoebic keratitis in South Africa [31] and seems to be a rarely identified form globally [1].The second one was the T11 sequence type, reported in the present study in one AK patient from Lazio.This genotype, closely related to T3, is also found with a low prevalence (2.07%) in AK cases from Europe, America, and Asia [25].
The results obtained in the present study allowed a better understanding of the extensive circulation of Acanthamoeba sequence types in Italy, confirming the well-established nature of these organisms as ubiquitous protozoans.
The identification of isolates from both humans and the environment at the allele level seems to suggest that, despite the genetic diversity observed within genotypes, this variability does not appear necessarily correlated to a differential isolates' ability to act as potential sources of infection for humans.This consideration is particularly applicable to isolates belonging to the sequence type T4, which, along with their specific properties (e.g., greater virulence and transmissibility) are the most widespread in natural habitats, thus having a higher likelihood of coming into contact with humans.
As for the other sequence types and alleles reported in Italy so far, the results here obtained indicate a widespread presence of less frequent or rare variants of Acanthamoeba isolates in the environment and in clinical cases.The real biogeographic distribution and consequent potential contribution to human infections of each of these genotypes in Italy remains unclear and can only be elucidated through the analysis of a more extensive and suitably diverse set of samples from both environmental sources and clinical cases.
Studying genetic variations within and among Acanthamoeba will help in formulating hypotheses about potential host-amoeba interactions, identifying possible pathways for the transmission of these free-living organisms, and exploring pathogenic aspects associated with distinct sequence types.

Figure 1 .
Figure 1.Phylogenetic tree based on Acanthamoeba 18S rRNA partial sequences generated in the present study (represented in bold) and reference sequences representing different Acanthamoeba sequence types (Tn) selected as indicated in "https://u.osu.edu/acanthamoeba/acanthamoeba-sequence-types/(accessed on 15 December 2023)" and retrieved from GenBank.Analysis was inferred by using the Maximum likelihood method (ML).Genetic distances were calculated using the Kimura 2-parameter model + G. Numbers on the tree nodes indicate bootstrap values >40%.

Figure 1 .
Figure 1.Phylogenetic tree based on Acanthamoeba 18S rRNA partial sequences generated in the present study (represented in bold) and reference sequences representing different Acanthamoeba sequence types (Tn) selected as indicated in "https://u.osu.edu/acanthamoeba/acanthamoebasequence-types/(accessed on 15 December 2023)" and retrieved from GenBank.Analysis was inferred by using the Maximum likelihood method (ML).Genetic distances were calculated using the Kimura 2-parameter model + G. Numbers on the tree nodes indicate bootstrap values >40%.

Figure 2 .Table 3 .
Figure 2. Sequence type distribution of Acanthamoeba isolates in clinical and environmental samples from Italy.Table 3. Sequence type distribution of Acanthamoeba in clinical and environmental samples from Italy based on the present study or retrieved from GenBank.Source T2/6 T3 T4 T11 T13 T15 Reference

Figure 2 .
Figure 2. Sequence type distribution of Acanthamoeba isolates in clinical and environmental samples from Italy.

Figure 3 .
Figure 3. Geographical distribution of Acanthamoeba genotypes in Italy.Both clinical and environmental isolates are represented.Values indicate the number of sequence types calculated for each region.

Figure 3 .
Figure 3. Geographical distribution of Acanthamoeba genotypes in Italy.Both clinical and environmental isolates are represented.Values indicate the number of sequence types calculated for each region.

Table 1 .
Acanthamoeba database of clinical sequences obtained through a multistep strategy search in GenBank.Isolates from present study are included.Accession number, genotype, allele, isolate, and region of sampling are indicated.Results are presented in order by genotype, according to the region." indicates as above.

Table 1 .
Acanthamoeba database of clinical sequences obtained through a multistep strategy search in GenBank.Isolates from present study are included.Accession number, genotype, allele, isolate, and region of sampling are indicated.Results are presented in order by genotype, according to the region." indicates as above.

Table 2 .
Acanthamoeba database of environmental sequences obtained through a multistep strategy search in GenBank.Accession number, genotype, allele, isolate, region of sampling, and source are indicated.Results are presented in order by genotype, according to the region." indicates as above.

Table 3 .
Sequence type distribution of Acanthamoeba in clinical and environmental samples from Italy based on the present study or retrieved from GenBank.