A Novel Culturing Chip (cChip) Can Facilitate Culturing of Unculturable Bacteria From Aquatic Environment

Background: Culturing the unculturable microorganisms is an important aspect of microbiology. Once cultured the unculturable microorganisms can be a source of useful antibiotics, enzymes etc. Several studies have been conducted on culturing the unculturable microorganisms from soil. But little knowledge exists about culturing such bacteria from aquatic environment. Therefore, in this study we designed a novel culturing chip (cChip) to facilitate the growth of unculturable aquatic bacterial community. cChip was optimized for microbial growth using known bacteria in the lab, later, microbes from a freshwater lake were concentrated (instead of using the traditional dilution method) and inoculated in cChip before incubating the chip in simulated lake environment. Then further sub-culturing was done on laboratory media. The eld samples were also analyzed using traditional culturing and metagenomics for a comparison. Results: Metagenomics analysis showed that 832 microbial species were present in the samples belonging to ve different phyla, that is, Cyanobacteria, Proteobacteria, Actinobacteria, Bacteroidetes and Verrucomicrobia. However, traditional culturing yielded only 36 isolates that belonged to four different phyla, that is, Actinobacteria, Proteobacteria, Firmicutes and Bacteroidetes. Culturing through cChip yielded 154 isolates belonging to ve different phyla, that is, Proteobacteria, Actinobacteria, Bacteroidetes, Verrucomicrobia and Firmicutes. Out of these 154 cChip isolates, 45 were previously uncultured bacteria having a 16S rRNA gene similarity from 91.35 % to 98.7 % to their closest relatives according to NCBI GenBank. Conclusion: This study shows that culturing microorganisms in the cChip from aquatic environment using concentration process before performing the traditional petri plate culturing can result in the successful growth of unculturable bacteria. To the best of our knowledge this is the rst study conducted for successful exploration of unculturable aquatic microbial community. This study can have a signicant impact on our understanding of the techniques that can be applied for exploring the unexplored microbiome from diverse environments. We also hypothesize that certain modications in the manufacturing material, keeping the design and techniques of this study intact can result in the application of this technique from gut microbiome to extremophiles in the environment.


Introduction
Since the past many years, it is an established fact that not all the microorganisms present in the environment can be grown in the lab. The term "Great Plate Count Anomaly" refers to the fact that a much greater number of microbes can be observed under a microscope compared to the number grown on petri plates in the form of colonies [1,2]. These microorganisms were initially thought as dead cells but later studies showed that these are metabolically active cells within their environment even if these are not able to grow on laboratory media [3,4]. Our knowledge of advances in DNA sequencing technology additionally helped to understand that a signi cantly high amount of microorganisms exists in nature which could be identi ed through metagenomics technique even if those cannot be grown in the laboratory [5]. The term unculturable only refers to the fact that these microbes are yet to be cultured. The reason for these microbes not being able to grow in laboratory is mainly attributed to the lack of understanding of the conditions which are needed to be provided like pH, temperature, osmotic pressure as well as nutrients besides several other factors [6].
Unless a microorganism is grown in the laboratory using conventional culturing techniques, its characteristics cannot be explored. Nearly half of the commercially available pharmaceutical products have been produced naturally from microorganisms or are their derivatives [7]. The isolation of similar types of bacteria has led to achieve a limited diversity of natural product producers [8]. If the unculturable majority can be made to culture then this problem can be solved [9,10]. Different attempts have been made to grow microorganisms from aquatic and terrestrial habitats by changing the nutritional requirements with limited success [11][12][13][14]. Besides, scientists have tried to develop techniques for culturing microorganisms in simulated natural environment as well as within their real environment (in situ culturing) [15][16][17]. Among these techniques the use of Isolation Chip (iChip) remained a successful method for isolation of unculturable microorganisms from soil, as a new class of antibiotics called Teixobactine was discovered using it [16,18]. Aquatic microbial community also harbors useful microbes, for example, Thermus aquaticus, the source of Taq polymerase is one of the famous examples, therefore, culturing the unculturable microorganisms from aquatic environment can result in the growth and utilization of useful microorganisms in the laboratory. To the best of our knowledge, so far, no such study is available and there is little knowledge available about the unculturable aquatic microbial community. Another phenomenon that helps in culturing the microorganisms is co-culture in which the cells are made to grow in an environment where they can interact with each other and with environment [19]. Previously we hypothesized that combining iChip technology and co-culture technique may help in isolation of uncultivable microorganisms as well as antibiotic discovery [20].
In this study we have successfully developed a microbial culturing chip (cChip) for the isolation of unculturable microorganisms from aquatic environment. Microorganisms form the source were rst concentrated instead of using the conventional serial dilution, inoculated in the holes of cChip and then the cChips were incubated in simulated lake water before further screening of isolates after subsequent subcultures. According to our survey of literature this is the rst kind of chip for microbial culturing that can be used to obtain more aquatic unculturable microbial strains and where concentration instead of dilution is used for facilitating the growth of unculturable microorganisms.

Designing of cChip:
In this study a culturing chip (cChip) was designed using Adobe Illustrator CC 2015 19.0.0 software. A manufacturer was consulted to make the device according to the design using Polypropylene plastic material. The design was optimized in terms of dimensions based on successful growth of known bacteria in the laboratory. The nal version of the cChip used is shown in gure 1. It is a polypropylene based cChip having total diameter of 5 cm with internal diameter of 3 cm containing the holes. Rest of 2 cm margins that does not contain holes are left for overall proper xation of membrane with glue. Each hole is of 3 mm diameter and distance between the holes is 5 mm, this distance is maintained for xing the membrane with glue here so that every hole is separated from the other and act as an entity. Thickness of this cChip is 50 mm.
Protocol for the Pre experiment The design was tested for successful growth of known microorganisms in the lab. The protocol of iChip as described by Berdy et al., 2017 was used with some modi cations [21]. 30 µl autoclaved Agar (BD TM Difco TM ) was inoculated in the holes of the cChip, 0.03 µm PCTE membrane Poretics TM (GVS Life Sciences) was used to cover the underside of the chip using glue (Momentive TM Sealant Rtv 108). Known bacteria E. coli was grown till the log phase, then 20 mL of nutrient broth containing log phase E. coli was taken and centrifuged for 15 min at 10,000 rpm to separate the pellet. This pellet was dissolved in 2 mL of 0.9% normal saline for making the inoculum. 10 µL of this inoculum was inoculated in the holes of cChip. Then other side of the chip was also covered using the membrane and glue as mentioned above and the cChips were kept in liquid nutrient broth containers to check if the nutrients diffused into the holes of the chip and E.coli was able to grow. Visible colonies in holes of the cChip showed that E. coli was successfully growing. Later log phase Rhodococcus inoculum was prepared like mentioned above and randomly mixed with E. coli inoculum. This mixture was inoculated in the cChip to check if both microbes were able to grow. Successful growth implied that our cChip would accommodate more than one bacterium to grow in aquatic environment. Different designs (not shown here) were tested for this purpose until the above-mentioned dimensions were selected based on the lab experiment results.

Field Experiment
A freshwater lake in Beijing was selected for the eld experiment. Samples were taken and were analyzed for exploring microbial diversity using three methods, cChip, traditional methods and metagenomics analysis.
cChip Inoculation: Water was brought in 5 L sterile containers in the lab. 50 mL water was ltered using 0.22 µm pore size MF-Millipore TM from Merck (Merck KGaA, Darmstadt, Germany) to retain all the microbes. The lter paper was washed using 2 mL of 0.9% saline and vortexed for detaching microbes into the saline. At rst 30 µL agar was added in each hole of cChip and its underside was sealed with PCTE using glue. Then 10 µL of inoculum prepared above was poured into each hole using micropipettes before sealing the top side of cChip. Then 5 L of water was put into an open container and all the prepared cChips were put into the container where these kept oating at the surface of water ( gure 2). The nutrients diffused from the bottom and microorganisms were able to grow on the top in each hole that were later seen with the naked eye as well as isolated. Water in the container was subsequently changed from the same lake after every week just for giving fresh nutrients to the microorganisms until the completion of experiment. After 8 weeks and 16 weeks of incubation the membranes from upper side and then lower side of the cChip were removed carefully and each hole was considered as the sample which was streaked on R2A agar (Coolabar, China) and FW70 media which was formulated according to Imazaki and Kobori 2010 [22].
Then subsequent subcultures from these media were performed for isolation of pure colonies of bacteria that were identi ed using 16S rRNA gene sequence analysis by the procedure mentioned in the later section. Figure 3 shows the complete experimental scheme.
Protocol used of traditional culturing method: Samples from the same source were serially diluted and different dilutions were spread on four culture media. These media used were, LB agar, nutrient agar, FW70 and R2A agar. 10 folds dilution (-1 dilution) was found to be most appropriate for analysis as different colonies were observed growing on the petri plates that were further subcultured, puri ed and identi ed using 16S rRNA gene sequencing described in the identi cation section. It was observed that at least eight to ten subcultures were minimally required for isolating a pure bacterial specie that could be further identi ed separately.

Identi cation of Microorganisms:
After pure colonies of microorganisms were obtained those were subjected to colony PCR for the purpose of identi cation. All the isolates were subjected to colony PCR using universal primers. The primers used were forward primers 5 GAGAGTTTGATCCTGGCTCAG and reverse primers 5 CTACGGCTACCTTGTTACGA. Following conditions were used for colony PCR, step 1: initial denaturation 94°C for 10 minutes, step 2: denaturation 94°C for 20 seconds, step 3: annealing 56°C for 30 seconds, step 4: extension 72°C for 1 minute 20 seconds, step 5: nal extension: 72°C for 10 minutes. A total reaction of 25 cycles was performed. The sequences were blasted against the NCBI database for identi cation. According to the literature, a cut off value of 98.7% 16S rRNA similarity was used for recognizing a new bacterial specie and 94.5% was used for recognizing a new bacterial genus [23].

Metagenomics analysis:
Lake water was ltered using 0.22µ lter papers and a triplicate of lter papers was subjected to metagenomic analysis to check for the molecular evidence of the available microbial diversity. Genomic DNA was extracted using PowerSoil DNA Isolation Kit (MoBio Laboratories, Carlsbad, CA). DNA concentration was tested using Qubit and PCR was conducted by using primer set speci c to V3-V4 regions of 16S rRNA gene, that is, 338f (5′-ACTC CTACGGGAGGCAGCAG-3′) and 806r (5′-GGACTACHVGGGTWTCTAAT-3′) [24]. The analyses of samples were done by using Illumina highthroughput sequencing platform (Allwegene, Beijing, China). According to the unique barcode of each sample sequencing reads were assigned before trimming the barcodes and primers [25,26]. From the original DNA fragments, pairs of reads were merged using FLASH (version 1.2.10) [27,28]. Then raw tags were ltered using QIIME (v1.8.0) using default settings and chimeric sequences were removed according to the previous studies [29]. Low quality fragments and the sequences that were less than 200 bp were removed by using Mothur program in order to minimize the effects of random sequencing errors and later the data generated was analyzed using QIIME (v1.8.0) [30]. Alpha and beta diversity were analyzed using Custom Perl scripts. Those sequences which had ≥97% were assigned the same operational taxonomic units (OTUs). To minimize the sequencing depth effect on indices the OUT table was rari ed. The Mothur program was used to generate the Simpson diversity index, Shannon diversity index and rarefaction curves [31]. Heat map analysis was conducted using R (version 3.1.1) for the 20 most abundant genera present in each sample under study. Then Venn diagram was created using InteractiVenn [32]. SILVA database (Version 128) was used for the alignment of 16S rRNA gene. Then taxonomy analysis was conducted using Ribosomal Database Project (RDP) Classi er at 70% con dence interval [33]. The sequences obtained from Illumina MiSeq sequencing from the triplicate samples were deposited in the Sequence Read Archive (SRA) database of National Center of Biotechnology Information under the accession numbers (SRR12044250, SRR12044251, SRR12044252).

Results
Before proceeding to actual experiment using environmental sample, the cChip was optimized to check if bacteria were able to grow in the chip or not. We call this stage as pre-experiment. Results are described as under. Pre-experiment: In this experiment, known bacteria, that is, E. coli (log phase) was inoculated in the holes of the cChip at a concentration of 10 µL and the chips were kept in nutrient broth. After 24 h of incubation at 37°C the growth of E. coli in the holes of chip was visible. This experiment resulted in the successful optimization of the parameters required for successful growth of a bacteria in aquatic environment.
Later a mixed solution of log phase E. coli (that has cream color growth) and log phase Rhodococcus sp.
(that has red color growth) were inoculated in the holes of the cChip and it was incubated in nutrient broth. Successful appearance of cream and red color colonies in the holes of the cChip concluded that its design was optimized successfully for the growth and isolation of more than one type of bacteria from aquatic environment.
Field Experiment: A freshwater lake was selected to carry out the eld experiment. Water samples were analyzed for exploring microbial diversity using three methods. Metagenomics analysis of the samples was performed to explore the molecular evidence of the presence of microbial biodiversity. Traditional culturing method was used to culture the samples on four different media to explore the culturable bacterial diversity of the samples. Finally, inoculation in the cChip was done and it was incubated in the simulated lake water environment to check if it could facilitate the growth of unculturable bacteria from the lake water. As a result, a comparison of three methods in achieving successful cultivation of uncultivable bacteria from the aquatic environment was drawn.
Metagenomics Analysis: Results of metagenomics analysis showed that microorganisms from at least ve different bacterial phyla were present in the samples. The raw data from the triplicate samples (M1, M2 and M3) has been submitted to the NCBI SRA database under the accession numbers (SRR12044250, SRR12044251, SRR12044252). Total 832 microbial species were present in the lake water ltrate that included microbes belonging to at least ve phyla, that is, Cyanobacteria, Proteobacteria, Actinobacteria, Bacteroidetes, Verrucomicrobia. A total of 198 bacterial genera were present according to molecular analysis. It was observed that in the triplicate samples tested for the presence of microbes metagenomically, Cyanobacteria were present in the highest abundance, followed by Proteobacteria, Actinobacteria, Bacteroidetes and Verrucomicrobia.
Traditional Culturing: The samples were cultured after performing 10 folds (-1) serial dilution on four culture media, that is, FW70, R2A agar, LB media and Nutrient agar. Total 36 isolates belonging to four different phyla were successfully cultured on these four media. 9 isolates were obtained from culturing on FW70 media, 13 isolates from nutrient agar, 6 isolates each from culturing on R2A agar and LB media. Interestingly bacteria belonging to Firmicutes were obtained using traditional culturing, however, those were not obtained using metagenomics analysis. Similarly, metagenomics showed representatives of Verrucomicrobia that were not obtained by traditional culturing. In total, three novel bacterial species were obtained using traditional method (all from culturing on FW70 media). As these species were obtained from cChip as well so these will be mentioned in the later section to avoid any repetition. In terms of the diversity of phyla obtained by using traditional culturing, Actinobacteria were the highest in number followed by Proteobacteria, Firmicutes and Bacteroidetes.
cChip Culturing: Samples from the lake water were concentrated before inoculation. The retentate on lter paper was washed in 5 mL normal saline and vortexed thoroughly to obtain concentrated inoculum and 10 µL from this inoculum was inoculated in the holes of cChip. Then incubation of four and eight weeks was carried out in simulated lake water environment before further culturing on two laboratory media, FW70 and R2A agar. This process resulted in successful culturing of 154 bacterial isolates belonging to ve different phyla. These were Proteobacteria, Actinobacteria, Bacteroidetes, Verrucomicrobia and Firmicutes.
Interestingly, four of these phyla were shown in metagenomics analysis also, while one phylum, that is, Firmicutes was not shown in metagenomics analysis. Besides, representatives of this phylum were able to grow in traditional culturing method. Isolates obtained from cChip culturing belonged to 79 genera in terms of diversity, 21 genera from traditional culturing and 206 genera from metagenomics analysis.
Venn diagram in gure 5 shows the number of individual and overlapping genera obtained from all the three methods. Relative abundance of the representative species belonging to different phyla obtained from cChip culturing was in the order, Proteobacteria Actinobacteria Firmicutes Bacteroidetes Verrucomicorbia. Figure 4 shows the relative abundance of microorganisms in terms of phyla obtained from three methods used in this study.
By using cChip culturing, 45 unculturable bacteria were successfully isolated that included one genus. A 16S rRNA gene sequence comparison showed that these isolates had a similarity index of 91.35% to 98.7% to their closest relatives according to the NCBI GenBank. 3 of the 45 isolates were also found using the traditional culturing that will be mentioned in the table 1 as under:

Discussion
In this study a new chip (cChip) was designed for in situ culturing of the unculturable microbial community from aquatic environment after using a process of concentration of microbes instead of traditional serial dilution. Many studies have been conducted previously for the exploration of microbial diversity from the environment using different approaches [15,16,[37][38][39][40][41][42][43]. All the approaches used a process of dilution for culturing of environmental samples and scientists were able to successfully isolate those bacteria that were previously not cultured. In this study we have used an approach of concentration instead of dilution to trap all the microorganisms from an environmental sample inside the chambers of one device. Franklin et al., 2001 concluded in a study that dilution can result in the elimination of rare organisms present in the sample that may result in the drop of overall diversity [44], therefore, we used the process of concentration to try to retain all the microorganisms in the chambers of cChip. We hypothesize that this approach provided a chance to unculturable microorganisms to thrive inside the chambers of cChip thus facilitating their growth in laboratory in further steps of our experiments.
In this study our metagenomics analysis did not show any member of phylum Firmicutes from the samples while representatives of this phylum were grown successfully in cChip and traditional culturing. It has been reported previously that there can be fraction of genomes that may remain undetectable using culture independent approaches because of many factors especially in case of rmicutes, the endospore formers remain undetected by using metagenomics [45]. Moreover, in our study, traditional culturing after serial dilution resulted in the successful growth of 36 bacterial species including three novel bacteria while cChip successfully isolated 154 bacterial species out of which 45 novel bacterial species including one genus was successfully cultured in the laboratory. The three novel species obtained from traditional culturing were also successfully cultured using the cChip. Therefore, we hypothesize that if microorganisms from an aquatic environment are concentrated in the holes of cChip and grown in simulated environment before streaking on laboratory media, the unculturable microorganisms can be successfully grown and isolated on laboratory media.
Domestication of microorganisms is another important phenomenon [16]. This is a process in which microorganisms are grown within their environment for some period before streaking them on laboratory media. Allain and Querellou 2009 concluded that transition of microorganisms from non-growing to growing stage on laboratory media, termed adaptation to laboratory media, can be a very slow process requiring lot of time [38]. Nichols et al., 2010 [16] described this process for soil microorganisms in a study where chips were placed in soil for some weeks and then microbes from the holes were streaked on laboratory media. Similarly, we trapped microbes in the holes of our cChip for a period of 8 weeks and 16 weeks (2 months and 4 months) before streaking them on laboratory media, therefore, after streaking, we were able to successfully isolate and grow many bacterial species including novel bacteria. A successful growth of 154 bacterial species including 45 novel bacterial species as compared to 36 bacteria including only three novel bacteria obtained from traditional culturing implies that a direct streaking of microbes from the holes of chip after two months of in situ incubation in simulated environment can result in successful adaptation of unculturable bacteria to grow on laboratory media. Normally, an increased length of incubation period is provided when microorganisms are grown on the petri dishes. Different studies show that an incubation of 4 weeks to 8 weeks can result in the isolation of novel microorganisms [17,46], however, in our study, although subsequent subcultures are required and petri plates having mixed cultures have to be incubated at room temperature for uptil 4 weeks but all the pure cultures have been isolated with an average incubation of 2 weeks and maximum incubation period of 3 weeks. Therefore, the use of our cChip can facilitate the growth of previously unculturable microorganisms in a comparatively less time as compared to the previously reported methods. To the best of our knowledge this is the rst study showing the process of in situ culturing in aquatic environment and all the previous studies have been conducted primarily on soil or marine sediments.
In our study, the cChip contains holes of 3 mm diameter which can easily hold more than one type of microorganisms and those can be seen with the naked eye. We would refer to the diffusion chambers designed by Kaeberlein et al. (2002) here, where scientists were able to successfully isolate novel bacterial species using these chambers that trapped many microbes at a single time. Although the process according to them was very laborious as many microbes were able to grow in the diffusion chambers [15], that had to be separated after many subcultures, still novel microbes were isolated using diffusion chambers. Similarly, in our technique we were able to isolate previously uncultured microbes besides the process being laborious. Later Nicholes et al., 2010 developed iChip technology where miniature chambers of the iChip were able to grow nearly one type of microorganism in the chip that were isolated after streaking from the respective holes [16]. Our designed cChip has holes of the diameter greater than iChip and less than the diameter used in diffusion chambers. This feature better adapts our cChip for in situ culturing in the aquatic environment although still the process becomes a bit laborious because many microbes grow in each hole of the cChip that are separated using conventional subcultures but it eventually leads to successful growth of unculturable microorganisms.
Previously we hypothesized that a combination of iChip technology and coculture technique can result in the isolation of novel microorganisms which can ultimately lead to potential antibiotic discovery [20]. In this study, helper species may have played their role in the growth of those bacteria that are previously unculturable. As the diameter of each hole can trap many microbes at a time, therefore there is every possibility that microbes may support each other's growth by co-culture phenomenon. Besides, we use concentration of microbes after retaining them on the lter paper, therefore, there is a possibility of the presence of helper species which can be a facilitating factor for growth of those microorganisms that have been previously uncultured. Different studies have proven in the past years that the presence of some microorganisms and sometimes a small distance between the two microorganisms like less than 2µm facilitates the growth of microorganisms in co-culture [47][48][49]. We would like to explore these synergistic relationships in future.
The undiscovered microbial diversity can be a potential source of secondary metabolites especially antibiotics, for example, Teixobactine, a new class of antibiotics have been discovered using the iChip technology where bacterium isolated from soil using this technique was found to be a potential source of this antibiotic [18]. Besides microorganisms have been a potential source of several useful compounds [7,50]. Therefore, culturing the unculturable bacterial diversity in the lab can help in the use of microbes to our advantage by exploring their potential of production of useful substances.

Conclusion
From this study we have concluded that concentrating bacteria from aquatic environment inside the chambers of our designed cChip and using environment as a medium can help in the exploration of aquatic unculturable bacterial diversity. These microbes ultimately adapt to grow on the lab media. Moreover, concentration instead of dilution reduces the chances of rare microbes to be eliminated at the initial stages of culturing. Although the process is laborious but a combination of these two techniques can lead to successful exploration of unculturable bacterial diversity. Once a microbe is successfully cultured, there is every possibility that it may be a source of useful compounds like enzymes, antibiotics, etc., besides, it can help in the overall understanding of microbes comprising a particular environment. We also hypothesize that certain modi cations in our cChip, like manufacturing material while keeping the rest of parameters and protocols the same can also lead towards successful exploration of microbiome from other environments as well. To the best of our knowledge this is the rst study that leads to successful exploration of aquatic microbial diversity by using is situ culturing and where concentration instead of traditional dilution s used. This study will provide a new direction in the quest of exploration of microbiome in diverse environments.

Declarations
Ethical approval: In this study only environmental samples were collected, and no human subjects were involved.

Consent for publication:
All the authors have given their consent for publication of this study.

Con ict of interest:
There is no con ict of interest about publication of this manuscript.
Availability of data and materials: The data presented in this manuscript is available in the NCBI Sequence Read Archives under the accession numbers: SRR12044250, SRR12044251, SRR12044252, besides all the NCBI GenBank submissions have been made under the accession numbers that have been written in Table 1 in the manuscript. Funding: There are no funding sources available for this study.
Author's contribution: Adil Farooq Lodhi and Ying Zhang contributed equally to this work and should be considered as rst coauthors, Maria Adil helped in the study design and compilation of data, Yulin Deng provided with all the     The evolutionary history was inferred using the Neighbor-Joining method [34]. The optimal tree with the sum of branch length = 2.87394367 is shown. The evolutionary distances were computed using the Kimura 2-parameter method [35] and are in the units of the number of base substitutions per site. The rate variation among sites was modeled with a gamma distribution (shape parameter = 1). This analysis involved 128 nucleotide sequences. All positions containing gaps and missing data were eliminated