Introduction

The phylum Planctomycetota represents a phylogenetically diverse and ecologically significant group of bacteria, widely distributed across aquatic ecosystems. In addition to their prevalence in marine and freshwater environments, strains belonging to this phylum have also been identified on macroalgal surfaces1,2,3,4, in marine hydrothermal systems5,6,7, on marine artificial and biotic surfaces8,9,10,11,12, marine sediments13,14, soil15, and in association with various eukaryotic hosts such as jellyfish16, sponges17,18,19, and within wastewater systems20. A characteristic feature of Planctomycetota is their involvement in biofilm formation, as evidenced by their abundance in algal and water plant surface biofilms21,22, microbial mats23,24 and on the surfaces of both, wooden10 and synthetic plastic substrates8,11.

To date, more than one hundred species within the phylum have been characterized25,26, revealing distinctive biological traits, including complex intracellular compartmentalization27,28 and an unconventional mode of asymmetrical cell division via polar budding, which occurs in the absence of otherwise canonical bacterial divisome proteins25,29,30. Additionally, genomic analyses and empirical studies have identified the potential for secondary metabolite biosynthesis within Planctomycetota25, including the production of (bioactive)31 small molecules32,33,34, underscoring their potential ecological and biotechnological relevance.

In addition to their distinctive cell biology, members of the phylum exhibit a wide range of metabolic and ecological lifestyles. The phylum includes organisms capable of thriving under anoxic35 and oxic conditions25. Metabolic diversity within Planctomycetota spans autotrophic (planctomycetes performing anaerobic ammonium oxidation (anammox))36 and heterotrophic25 modes of nutrition, with certain taxa exhibiting predatory behavior specialized towards other bacteria37,38. In general, planctomycetes are characterized by low growth rates and are notoriously difficult to cultivate under laboratory conditions. During the successful isolation of more than 150 strains to date, a remarkable diversity of physiological traits and lifestyles has already been observed. This raises the compelling question of whether continued isolation efforts will further uncover previously unrecognized metabolic capabilities and ecological roles within the phylum. To unravel the full extent of Planctomycetota diversity and complexity, sustained efforts in cultivation and characterization remain essential.

To extend beyond the predominantly surface-associated habitats from which most isolates of the family Pirellulaceae have been recovered to date—including plant surfaces—a water sample was used as inoculum for isolation from the unsaturated zone of the terrestrial subsurface at the Hainich Critical Zone Exploratory (CZE) in Western Thuringia, Germany.

As of mid 2025, the family Pirellulaceae consists of 14 validly described genera and 41 validly described species. It was validly described in 202039, the type genus is Pirellula40 and the type species Pirellula staleyi40. Phylogenetically, the family is part of the phylum Planctomycetota, which consists of the four described classes Planctomycetia, Phycisphaerae, Candidatus Brocadiia, and Candidatus Uabimicrobiia, with the family of Pirellulaceae belonging to the class Planctomycetia and the order Pirellulales.

Materials and methods

Sampling site and sample characteristics

The percolate sample containing strain NA78T was taken from a drainage collector installed in a sub-horizontal borehole in weathered bedrock at the Hainich CZE of the Collaborative Research Center (CRC) AquaDiva41. At the forested Hainich low-mountain ridge (temperate climate) representing a groundwater recharge area of regional importance, these collectors provide further access to subsurface compartments of the Critical Zone, besides lysimeters (soil seepage)42 and monitoring wells (groundwater)43. At the Reckenbühl location (51.101910, 10.397129), the collector in ~ 0.65 cm depth collects percolate from fractured bioclastic limestone (Trochitenkalk Fm.) of weathered marine limestone-mudstone alternations (Triassic, Upper Muschelkalk), covered by Chromic Cambisol. More details on the sampling site and the drainage collector setup have been previously published42,43,44. The sample was collected on 11th March 2020, transported to the lab storage (4 °C), and represents bedrock percolate (~ 0.13 L) accumulated during 13 days of collection and dark storage under a mean air temperature of 3.5 °C (Reckenbühl weather station). Basic quality parameters of the Ca-HCO₃ type water are: pH: 6.6; EC25: 80.3 µS/cm; total organic carbon (TOC): 6.55 mg/L; dissolved organic carbon (DOC): 3.52 mg/L.

Isolation and cultivation

Strain NA78T was isolated using a standardized limnic planctomycete medium (limnic M1 medium; M1H NAG AFW) consisting of artificial freshwater (AFW), 4-(2-hydroxyethyl)1-piperazineethanesulfonic acid (HEPES) and N-acetylglucosamine (NAG) as carbon and nitrogen source45. 50 µL percolate water sample was spread on limnic M1 plates (solidified with 15 g/L agar) supplemented with 100 mg/L ampicillin, 20 mg/L cycloheximide, and 20 mL/L nystatin to avoid growth of fast-growing bacteria or fungi44. Plates were incubated under oxic, dark conditions at 18 °C. Plates were checked regularly to investigate colony growth. Colonies appeared after a few weeks of incubation and were re-streaked to receive axenic cultures. Once cultures appeared to be axenic, a 16S rRNA gene sequence-based identification was performed following a previously published protocol46. After isolation and identification, strain NA78T was cultivated on limnic M1 plates and liquid cultures without supplemented antibiotics.

DNA extraction, genome sequencing, and genome assembly

Genomic DNA isolation, long-read sequencing using Oxford Nanopore technology, genome assembly and polishing using Illumina short reads including post-processing of the assembly was carried out as previously described20 with the recently reported modifications used for strain SH66447. Illumina sequencing was performed by Eurofins Genomics (Ebersberg, Germany). The Galaxy web-based platform and the server available under the public domain usegalaxy.eu48 were used for parts of the bioinformatic workflow as previously reported20. The used materials and tools are provided in Table S1.

Nucleotide sequence accession numbers

Nucleotide sequences of the novel strain are deposited in the NCBI GenBank database under accession numbers PV476534 (16S rRNA gene sequence) and CP187419 (genome).

Phylogenetic analysis

Maximum likelihood phylogenetic trees based on 16S rRNA gene sequences and multi-locus sequence analysis (MLSA) were calculated as described44,49,50,51. The 16S rRNA gene sequence-based tree included the sequences of all validly published members of the phylum Planctomycetota (as of May 2025) including three members of sister phyla from the PVC (Planctomycetota-Verrucomicrobiota-Chlamydiota) superphylum, namely Lentisphaera araneosa HTCC2155 (NCBI acc. no. AJ229235), Kiritimatiella glycovorans L21-Fru-AB (acc. no. NR_146840), and Opitutus terrae PB90-1 (NCBI acc. no. AJ229235) as outgroup. The MLSA-based tree was calculated based on the genomes of all current members of the family Pirellulaceae as previously described44. The genomes of three species belonging to the family Lacipirellulaceae, namely Aeoliella mucimassa Pan181T (NCBI acc. no. CP036278.1), Bythopirellula goksoeyrii Pr1dT (NCBI acc. no. CP042913.1), and Lacipirellula parvula PX69T (NCBI acc. no. AP021861.1) served as outgroup. Additional phylogenetic markers, including average amino acid identities (AAI), average nucleotide identities (ANI), percentage of conserved proteins (POCP), and partial rpoB gene sequence similarity were calculated as described previously44. To further support species delineation, digital DNA–DNA hybridization (dDDH) values were determined with the Genome-to-Genome Distance Calculator (GGDC) hosted by the DSMZ using formula 252,53,54.

To assess the presence of strain NA78T in local groundwater and seepage, its 16S rRNA gene sequence was compared with publicly available bacterial 16S rRNA gene datasets previously obtained from groundwater and seepage samples collected at the Hainich CZE55. Details of the procedures for amplicon sequence variants (ASVs) generation were previously described55. Briefly, after primer removal, the DADA2 package (v1.26) in R was used to infer ASVs, and taxonomic assignment was performed using the SILVA reference database (v138.1). ASVs classified within the phylum Planctomycetota were further analyzed by performing pairwise alignments with the 16 SrRNA gene sequence of strain NA78T using the “pairwiseAlignment” function in the Biostrings package (V 2.66). The four most closely related ASVs were subsequently subjected to BLAST analysis to determine their percent identity to strain NA78T.

Analysis of genome-encoded features

Genome completeness and contamination as well as coding densities were analyzed with checkM v.1.2.356. Secondary metabolite-associated biosynthetic gene clusters were predicted with antiSMASH version 857. The reconstruction of the pangenome was performed with anvi’o v.858. To analyze the presence/absence of representative canonical cell division genes within the genus Anatilimnocola, the genomes of Anatilimnocola floriformis PX40T (GenBank accession number GCA_024256385.1) and Anatilimnocola aggregata ETA_A8T (GenBank accession number GCA_007747655.1) were downloaded from NCBI. Together with the genome of strain NA78T (GenBank accession number GCA_049819085.1), the genomes were analyzed using the run-kegg-kofam function from anvi’o 7.159 to annotate the genomes based on KEGG orthology (KO) numbers. The presence/absence of the KO numbers of FtsA (K03590) and FtsZ (K03531) was then checked in the three genomes.

Physiological tests

Temperature and pH optima for growth were determined by streaking strain NA78T on limnic M1 agar plates and by inoculating liquid limnic M1 medium, respectively. To evaluate temperature-dependent growth, cultures were incubated at 4, 10, 18, 21, 24, 28, 32, 37, and 42 °C. For each temperature, 150 µL of a culture adjusted to an optical density at 600 nm (OD600) of 0.05 was plated in duplicates. Plates were incubated for up to five weeks at the respective temperatures, with regular monitoring for visible growth. After five weeks of incubation, growth was documented by photography.

The optimal pH for growth was determined by preparing limnic M1 medium with different pH values using either 2-(N-morpholino)ethanesulfonic acid (MES), HEPES, or N-cyclohexyl-2-aminoethanesulfonic acid (CHES) as buffering agent with concentrations of 100 mM. The pH was adjusted using HCl or KOH. The pH values 5.0, 6.0 (MES), 7.0, 7.5, 8.0 (HEPES), 9.0 and 10.0 (CHES) were tested. The incubation took place at 21 °C with a total volume of 5 mL culture in biological duplicates, as this temperature supported robust growth during initial screening and allowed reliable assessment of pH-dependent growth. Anaerobic growth (fermentation and anaerobic respiration) was tested by flushing liquid limnic M1 medium with dinitrogen gas for several minutes and transferring it to sterile, anoxic vessels containing 100% dinitrogen gas. To test for fermentation, plain anoxic limnic M1 medium was inoculated to an OD600 of 0.05. Anaerobic respiration was tested by adding the alternative electron acceptors sodium nitrate (3 mM), sodium fumarate (10 mM), and sodium sulphate (10 mM) separately to the plain, anoxic M1 medium that was inoculated to an OD600 of 0.05. To verify growth, oxic control vessels with an OD600 inoculum of 0.05 were regularly compared to anoxic vessels. Inoculated anoxic vessels were transferred to an anaerobic chamber (BugBox Ax, Baker Company) to avoid oxygen influx. The OD600 was measured on a regular basis for at least two months. The incubation took place at 18 °C.

Phenotypic analysis and cell size determination

Phase contrast (PhC) and differential interference contrast (DIC) images of cells mounted on a 1% (w/v) agarose cushion were acquired with a Nikon Eclipse Ti2 setup as described previously24. Images for display were adjusted with the brightness and contrast function of FIJI and scale bars were added in FIJI60. The cell size was analyzed as described20. Briefly, native Nikon files were converted to tiff files using FIJI, images were transferred to BacStalk61, segmentation settings were adjusted to 20 pixels for cell size and 15 pixels for minimum cell size, stalk detection was unticked. Cells segmented by BacStalk were manually checked for segmentation errors. The obtained data was transferred to Excel and then uploaded to SuperPlotsOfData62 for plotting. Colony images of strain NA78T were taken after incubating a loop of culture from a cryo stock on limnic M1 agar plates.

Fluorescence microscopy

Cells were stained with 4′,6-diamidino-2-phenylindole (DAPI) and Synaptored as described earlier63. Briefly, 3 µL DAPI (500 µg/mL) and 1 µL Synaptored (1 mg/mL) were added to 500 µL culture and cells were stained for 10 min and washed two times with limnic M1 medium. Cells were then immobilized on an agarose cushion (1% agarose in water, w/v). Images (z-stacks) were obtained with the same setup described above, but using an Orca-flash 4.0 camera, a DAPI filter (Semrock; DAPI-1160B), and a Texas Red longpass filter (AHF; 560/40x, 600 DC, 610 LP). Obtained images were transferred to FIJI and individual images from the z-stack were selected. Brightness and contrast were adjusted and DAPI and Synaptored images were merged before adding scale bars.

Results and discussion

Isolation of strain NA78T

Strain NA78T was isolated on the 4th of May 2020, 54 days after samples were taken, using limnic M1 agar plates. As ca. 60% of the so far isolated planctomycetal strains appear pink due to carotenoid biosynthesis20,64, we normally tend to screen for pinkish colonies. However, since strain NA78T was isolated from the subsurface, which is characterized by the absence of light, and other subsurface planctomycetes appear to be more whitish to beige44, not only pinkish colonies were checked. To ensure culture purity, individual colonies were sub-cultured three consecutive times on limnic M1 agar plates. The axenic status was subsequently confirmed through amplification and sequencing of the 16S rRNA gene.

Phylogenetic characterization

To identify strain NA78T, the obtained partial sequence of its 16S rRNA gene was compared using Blastn and revealed phylogenetic relationship to A. aggregata ETA_A8T (MK559973.1) as current closest neighbor. An identity of 98.5% over the analyzed length of 740 bp was obtained that falls below the species threshold of 98.7%65. Thus, strain NA78T was selected for in-depth characterization including genome sequencing. After the genome sequence was obtained, the phylogenetic position of strain NA78T was analyzed further employing maximum-likelihood phylogenetic tree calculation based on full length 16S rRNA gene sequences (Fig. 1A) and MLSA-based tree construction (Fig. 1B). Within both trees and as expected from the Blastn pre-screening, strain NA78T clusters with the two described members of the genus Anatilimnocola.

Fig. 1
Fig. 1
Full size image

Phylogenetic position of the novel strain NA78T within maximum likelihood phylogenetic trees. Maximum likelihood phylogenetic trees based on 16S rRNA gene sequences (A) and MLSA (B) were constructed using strains belonging to the phylum Planctomycetota (A) or to the family Pirellulaceae (B). Outgroups are described in the Materials and methods section. Strain NA78T clusters within the genus Anatilimnocola in the family Pirellulaceae with its closest relative being Anatilimnocola aggregata ETA_A8T. Phylogenetic trees were visualized with iTOL v6. The scale bar indicates the number of substitutions per nucleotide (A) or amino acid (B) position. Bootstrap values from 1000 (A) or 500 re-samplings (B) are shown at the nodes (in %).

To further evidence the classification as novel species within the genus, the phylogenetic markers AAI, ANI, POCP, partial rpoB sequence similarity (gene coding for the β-subunit of RNA polymerase), and digital DNA-DNA hybridization (dDDH) were assessed (Fig. 2). Strain NA78T yielded AAI values of 80.4% and 71.7% during comparison with A. aggregata ETA_A8T and A. floriformis PX40T, respectively. With an AAI value between 65% and 95% a strain is assigned to a novel species66 which is in agreement with the obtained values. The calculation of ANI values revealed identities of 80.6% (A. aggregata) and 78.5% (A. floriformis), both being far below the given species threshold of 95–96%67, further supporting the delineation of strain NA78T from the described Anatilimnocola species. The placement of strain NA78T as novel species within the genus Anatilimnocola is further strengthened with dDDH values of 20.7% and 21.9% to its closest relatives Anatilimnocola floriformis PX40T and Anatilimnocola aggregata ETA_A8T, respectively, being well below the species threshold of 70% and probabilities for DDH ≥ 70% of 0%52,68. The POCP comparison gave values of 74.6% and 69.1% to A. aggregata ETA_A8T and A. floriformis PX40T, respectively. Since the POCP is typically used for the delineation of genera, the values only support the relationship on the level of the same genus (> 50% for members of the same genus)69. The comparison of a ca. 1300 bp partial sequence of the rpoB of strain NA78T yielded similarity values of 89.2% and 89.0% to A. aggregata ETA_A8T and A. floriformis PX40T, respectively, that fall below the species threshold of 96.3%, but above the genus threshold range of 75.5–78.0% used for the family Pirellulaceae70.

Fig. 2
Fig. 2
Full size image

Analysis of phylogenetic markers. Strain NA78T was compared to its closest relatives Anatilimnocola floriformis PX40T and Anatilimnocola aggregata ETA_A8T using the phylogenetic markers 16S rRNA gene sequence similarity (16S), average nucleotide identity (ANI), digital DNA-DNA hybridization (dDDH), average amino acid identity (AAI), percentage of conserved proteins (POCP), and partial rpoB gene sequence identity.

With all phylogenetic markers including 16S rRNA gene sequence identity falling below the respective species threshold values, the classification of strain NA78T as the member of a novel species within the genus Anatilimnocola is supported.

Comparison of genomic features and analysis of secondary metabolic functions

The comparison of genomic features included strain NA78T and the type strains of the two closely related Anatilimnocola species (Table 1). The latter two have a very similar genome size of ca. 9.0 Mbps whereas the genome of strain NA78T is one Mbp smaller. Plasmids are probably not present in the current genus (the genome sequence of A. floriformis PX40T is not yet complete, hence the information is tentative.). The DNA G + C content is around 58% and differs less than 1 percentage point for the compared genomes. Overall similarities of the genome content are reflected in similar numbers of tRNA genes (67–72 genes), single copies each of 5S, 16S and 23S rRNA genes, coding densities of 87–89% and relative numbers of hypothetical proteins between 25 and 31% of the overall number of protein-coding genes. Pangenome reconstruction yielded a total of 10,599 gene clusters, of which 1350 turned out to be singletons in strain NA78T (Fig. 3). Within the set of singletons, 391 genes could be automatically annotated (Table S2). The list contains putative regulatory proteins (protein kinases), DNA-binding or -modifying enzymes (nucleases, helicases, transposases, transcriptional regulators, proteins involved in DNA repair, etc.), periplasmic proteins, alternative sigma factors and others, such as transport proteins. The core genome shared by all three strains consisted of 4778 genes that include encoded proteins with essential functions (replication, transcription and translation, primary metabolism) and were not analyzed in further detail at this stage.

Table 1 Comparison of genomic features and secondary metabolite-associated biosynthetic gene clusters.
Full size table
Fig. 3
Fig. 3
Full size image

Pangenome based on the analyzed Anatilimnocola genomes. Each open circle represents the pangenome of all strains but is colored darker when the gene is present in the respective genome. The matrix in the upper right corner indicates the degree of relationship based on average nucleotide identity values (pale orange: < 70% to bright orange: 100%). The group of singleton genes of strain NA78T is indicated by an asterisk.

The prediction of secondary metabolite-associated biosynthetic gene clusters (BGCs) using antiSMASH 8 yielded between 6 and 8 hits per genome (Table 1), which corresponds to slightly less than one BGC per Mbp. Most of the BGCs are putatively involved in the biosynthesis of terpenoids. Minor conserved clusters encode putative non-ribosomal peptide synthetase (NRPS)-like enzymes. A single type I polyketide synthase cluster was found in A. aggregata ETA_A8T, while A. floriformis PX40T harbors a putative type III polyketide synthase cluster. Strain NA78T lacks any polyketide synthases. Polyketide synthases are key enzymes involved in the biosynthesis of diverse potentially bioactive secondary metabolites71,72; therefore, their presence within a genome is often considered indicative of a strain’s potential to produce biologically active compounds and suggests further investigation. In planctomycetes, type III polyketide synthases have been associated with the production of long-chain alkylresorcinols, compounds that may contribute to biological activities71,73. Consequently, the detection of putative type I or type III polyketide synthase genes in a novel planctomycetal strain may suggest the capacity for producing such bioactive molecules. The absence of these genes in strain NA78ᵀ, however, indicates a low potential for the biosynthesis of polyketide-derived metabolites and suggests that any antimicrobial activity linked to such pathways is unlikely.

Physiological analyses and in situ occurrence

Compared to its closest relatives A. aggregata ETA_A8T and A. floriformis PX40T, which both were isolated from freshwater45,74, strain NA78T was isolated from a subsurface habitat. While both habitats share the characteristic of freshwater, environmental and abiotic conditions are drastically distinct between them. Shared by strain NA78T and its closest relatives A. aggregata ETA_A8T and A. floriformis PX40T is the ability to grow on glucose as carbon source45,74. The temperature range for growth of strain NA78T was tested by inoculating the strain on limnic M1 medium under oxic conditions. Testing revealed the ability of strain NA78T to grow between 18 °C and 24 °C with an optimum at 18 °C (Fig. 4). No growth was detected at 10 °C or above 24 °C, indicating that the lower temperature limit lies between 10 °C and 18 °C. The mesophilic nature of strain NA78T is shared with its closest relatives A. aggregata ETA_A8T and A. floriformis PX40T with growth ranges (and optimum growth temperature) of 15–33 °C (30 °C) and 15–30 °C (22–25 °C), respectively45,74.

Fig. 4
Fig. 4
Full size image

Determination of pH and temperature ranges for growth. Strain NA78T thrives within a temperature range of 18 to 24 °C and pH 6.0–9.0, with an optimal growth temperature of 18 °C and pH of 7.5. No growth was observed below 18 °C or above 32 °C. At 28 °C, slow growth was detected in liquid culture, but not on limnic M1 plates.

To test the strain’s pH range for growth, cells were inoculated into liquid limnic M1 medium and incubated at 21 °C. Testing revealed a tolerated external pH value of 6.0 to 9.0 with an optimum at pH 7.5, rendering strain NA78T a neutrophile. Its closest relatives A. aggregata ETA_A8T and A. floriformis PX40T show growth (and optimum pH for growth) between pH 5.0 to pH 10.0 (pH 8.0) and pH 5.0 to pH 7.5 (pH 6.5–7.0), respectively45,74, thereby also being neutrophilic strains. Thus, no obvious adaptions to the subsurface habitat could be deduced from the temperature optima while the optimal pH for growth of strain NA78T matches with that of its habitat.

To test the capacity of strain NA78T to grow under anoxic conditions, cells were inoculated into plain anoxic liquid limnic M1 medium and anoxic liquid limnic M1 medium supplemented with 10 mM sodium sulphate/fumarate or 3 mM sodium nitrate. After eight weeks of incubation neither in the fermentation nor in the anaerobic respiration testing vessels growth could be observed. Hence, strain NA78T is characterized as strictly aerobic planctomycete. This aligns with the aerobic lifestyle of its closest relatives45,74.

To investigate the potential in situ occurrence of strain NA78T within its native environment, its 16S rRNA gene sequence was further compared to an ASV database derived from seepage water, water percolating the unsaturated zone, and groundwater samples collected at Hainich CZE55(Table S3). The four most closely matching ASVs – ASV_3449, ASV_512701, ASV_73205 (all from groundwater), and ASV_87558 (from seepage) – showed sequence similarities of 92.5%, 92.5%, 91.5%, and 91.8%, respectively, with a consistent query cover of 57%. The relative abundance of these ASVs in the corresponding microbial communities were up to 0.3%, 0.15%, 0.01%, and 0.01%, respectively. Taxonomically, all ASVs belong to the family Pirellulaceae but differed at the genus levels. Given the moderate sequence identity and absence of a high-confidence match, strain NA78T was not detected in the environmental amplicon data. This likely reflects its low abundance in the native subsurface habitat, limitations in sequencing depth and resolution, or could point towards the possibility that this strain was washed into the subsurface from upper layers of soil where it might be more abundant.

Phenotypic characterization

After having determined the optimal cultivation conditions, the morphological traits of strain NA78T were investigated. On a macroscopic scale, strain NA78T appears in the form of white to beige-colored colonies on limnic M1 agar plates (Fig. 5A; Table 2), which aligns with our expectation of a subsurface isolate without the need of UV protecting pigments, usually common in surface planctomycetes75,76. The colonies are round and form entire margins, colonies appear shiny and reflect light (Fig. 5A). Liquid cultures are beige to whitish in color and only occasionally small macroscopic aggregates can be observed.

Fig. 5
Fig. 5
Full size image

Colony and cell morphology, and cell size of strain NA78T. (A) Image of round colonies with entire margins displaying a white to beige color. (B) Phase contrast (PhC) and differential interference contrast (DIC) images of cells of the novel isolate. The scale bar represents 2 μm. (C) DIC microscopic image of cell aggregates. The scale bar represents 5 μm. (D) Cell sizes determined in three replicates (150 cells each) indicated by three different colors; larger circles indicate the mean values of each replicate. The cell size is approx. 1.8 ± 0.3 × 1.1 ± 0.2 μm, supporting the rice-grain shape of the cells.

Table 2 Comparison of phenotypic and physiological characteristics of strain NA78T to its closest relatives Anatilimnocola aggregata ETA_A8T and Anatilimnocola floriformis PX40T, n.o.: not observed, n.d.: not determined.
Full size table

On a microscopic scale, cells are rice grain-shaped (Fig. 5B); A. aggregata ETA_A8T and A. floriformis PX40T share this cell morphology45,74. In most of our observations, single cells were observed, but occasionally cells formed microscopic aggregates (Fig. 5C). We could not observe aggregates in form of rosettes, which separates this strain from its current closest neighbors45,74. Individual cells of strain NA78T displayed a mean length of 1.8 ± 0.3 μm and a mean width of 1.1 ± 0.2 μm (Fig. 5D). This observation renders strain NA78T cells the shortest in length, but ranked second in width, since A. aggregata ETA_A8T and A. floriformis PX40T have a mean length and width of 2.0 ± 0.3 × 1.4 ± 0.2 µm and 1.9 ± 0.5 × 0.7 ± 0.2 µm, respectively45,74. As observed for all members of the family Pirellulaceae70,77,78, including the two closest relatives45,74, cells of strain NA78T divide asymmetrically via polar budding. This mode of cell division appears to function without most of the canonical cell division machinery30,79. Similar to other members of the phylum Planctomycetota25, we could not identify the otherwise canonical cell division genes ftsZ and ftsA in the genomes of the compared type strains A. aggregata ETA_A8T, A. floriformis PX40T, and strain NA78T.

Strain NA78T possesses comparably few membrane invaginations and a condensed nucleoid

Members of the phylum Planctomycetota are known for their ability to form invaginations of their cytoplasmic membrane, thereby creating an enlarged periplasmic space. This phenotype is one of the distinctive characteristics of Gemmata obscuriglobus (family Gemmataceae) and was also investigated in the past for Planctopirus limnophila (family Planctomycetaceae) upon its involvement in polysaccharide uptake27. Recently, we were able to extend this observation to the family Isosphaeraceae by investigating this feature in Kueselia aquadivae44. Since strain NA78T is a member of the family Pirellulaceae we wanted to test the strain’s ability to form cytoplasmic invaginations (Fig. 6) by staining the cells with Synaptored, a membrane-staining dye. In contrast to G. obscuriglobus, P. limnophila, and K. aquadivae, invaginations of strain NA78T could be observed in only a small number of cells. Observed invaginations appeared smaller in size and we were unable to find invaginations spanning through the cytoplasmic space. Nevertheless, strain NA78T appears to have the ability to form (small) invaginations. However, without analyzing species from the families Lacipirellulaceae and Thermoguttaceae it remains speculative if this feature can be attributed to the order Pirellulales and the class Planctomycetia as a whole. Additionally, untypically for many bacterial model specimen, planctomycetal cells often possess a condensed nucleoid27,44,80, which can be visualized by DAPI staining. Similar to the previous observations, stained cells of strain NA78T displayed a highly condensed nucleoid in planktonic cells as well as in cells being part of aggregates (Fig. 6).

Fig. 6
Fig. 6
Full size image

Strain NA78T displays few cytoplasmic invaginations and a condensed nucleoid. Single cells as well as cells in aggregates show only few cytoplasmic invaginations (Synaptored, arrowheads). Strain NA78T possesses a condensed nucleoid (DAPI). Scale bars represent 2 μm.

Habitats of isolated family Pirellulaceae strains

The family Pirellulaceae is a diverse group of planctomycetes, constituted by 14 validly published genera and 41 validly published species at the time of writing. Strains isolated from marine habitats span over the six genera Blastopirellula, Crateriforma, Novipirellula, Rhodopirellula, Rosistilla, and Stieleria6,70,77,81,82,83. But not only abiotic surfaces and water is inhabited by these microorganisms. Planctomycetes assigned to the family Pirellulaceae have also been isolated from marine phototrophs, like marine macroalgae, such as the six genera Bremerella, Mariniblastus, Rhodopirellula, Roseimaritima, Rosistilla, and Rubripirellula1,2,3,4,78,82. However, strains belonging to the family have also been isolated from jellyfish such as the two genera Aureliella and Novipirellula16,70 and from a sponge (a member from the genus Crateriforma)19. Furthermore, isolates belonging to the genera Blastopirellula and Bremerella were obtained from marine hydrothermal vents and volcanic sites5,84. Besides marine habitats75 strains were obtained from coastal brackish habitats, such as from the genera Lignipirellula, Roseimaritima, and Stieleria10,64,85. Pirellulaceae from the genera Rhodopirellula and Rubripirellula and Anatilimnocola were even found in freshwater8,40,45,74,86. In contrast, strain NA78T was obtained from a subsurface habitat. It is not yet clear whether it thrives in the subsurface environment or was washed into the subsurface. However, from amplicon data we know that members of the family Pirellulaceae are for example extremely abundant in the attached fraction of limnic cyanobacterial blooms45. Thus, strain NA78T might have dwelled originally in a lake or surface habitat and was washed into the subsurface.

Conclusion

With members of the planctomycetal family Pirellulaceae already being isolated from a broad diversity of habitats ranging from marine to freshwater and soil habitats, strain NA78T found in shallow bedrock adds another habitat to the spectrum, the terrestrial subsurface. The novel strain was isolated from percolate waters in fractured limestone of the shallow subsurface of the Hainich Critical Zone Exploratory in central Germany. The polyphasic approach including genomic, phylogenetic, morphological, and physiological characterization supports the position of strain NA78T as the type strain of a novel species of the genus Anatilimnocola within the phylum Planctomycetota, class Planctomycetia, order Pirellulales, family Pirellulaceae. Strain NA78T thereby represents the third described species of this genus. As isolated within the AquaDiva project, we propose the name Anatilimnocola aquadivae sp. nov. for the novel taxon.

Description of Anatilimnocola aquadivae sp. nov.

aquadivae (a.qua.di’vae. N.L. gen. n. aquadivae, of the AquaDiva project).

Cells form round colonies with a white to beige color and entire margins. Cells are rice-grain shaped with a size of approx. 1.8 ± 0.3 × 1.1 ± 0.2 μm, have a condensed nucleoid and divide asymmetrically by polar budding. Growth of the strictly aerobic type strain was observed between 18 °C and 24 °C, with 18 °C being the optimum temperature, and a pH of 6.0 to 9.0, with optimal growth at pH 7.5. The type strain is NA78T (= CECT 30429T = STH00992T, the STH number refers to the Jena Microbial Resource collection JMRC). It has a genome size of 7,974,895 bp, a DNA G + C content of 58.2% and lacks extrachromosomal elements. The type strain was isolated from a percolate sample, collected by a drainage collector within the groundwater recharge area of the Hainich Critical Zone Exploratory.