Micrococcus luteus Isolated From a Fish Tank Filter

Ariana Casey

BIOL 342 — Microbiology


The human race predominantly lives on land and breathes in the air surrounding them, not giving much thought to the bodies of water that cover the majority of the earth’s surface.   In all these environments exists a common organism — bacteria.   One must look past the human —past what the naked eye can see- to understand the bacteria.   Many species exist, inhabiting a multitude of niches in the world, from the inside of soil clumps and the bottom of the ocean —devoid of oxygen- to the skin on our hands and the surfaces of plants —with aerobic respiration to sustain them.

With so many different species of microorganisms, it would be a shame to disregard the ones that live in water simply because it’s not the environment that people exist in.   With the human body and the surrounding environment being so dependent on water, is would make sense to examine this substance to find what microbes exist there.   Among one of the many uses for water that people have, the one that seemed it would have a good measure of microbiota was a fish tank filter.

The objective of the following research was to isolate a bacteria from a sampled location and to determine through physiological and genetic testing what bacteria was isolated from the sampling location.  Being that a fish tank seems a rich environment for microbiota, it is hypothesized that the bacteria will be unique, with the ability to live aerobically or anaerobically.



The chosen location for sampling was the filter of a fish tank.   To properly sample, the filter was first removed from the water.   After removal from the fish tank, the filter was swabbed with a cotton swab, which was then streaked over the surface of the TSA plate (Lab Handout 1).   The plate was kept at room temperature for six days before being brought into lab.   At this point, a colony was pulled with a sterilized metal loop and put onto another TSA plate using a quad streak (Lab Handout 2).   As the bacteria grew, individual colonies were used to further isolate the bacteria through additional quad streaks.

At the fourth quad streak, the culture was determined to be fairly pure through microscopic observation (Lab Handout 4).  At this time, the culture was deemed pure enough to be physiologically and genetically tested to determine bacterial strain.


One objective of obtaining a pure culture through quad streaking was to observe the colony morphology of the cells in isolate (Lab Handout 2).  To determine cell size, shape, arrangement, and wall-type, gram-staining was performed (Lab Handout 4).     Fluid thioglycollate test was performed and incubated at room temperature to determine oxygen class (Lab Handout 6).   Presence of cytochrome C oxidase was tested using an oxidase test strip with an indicator dye (Lab Handout 6).   A catalase test was used to determine if the strain of bacteria has the enzyme catalase (Lab Handout 6).   To determine which sugars the bacteria could ferment, as well as to determine other physiological characteristics which could help to characterize the isolated bacteria, an API 20E test strip was used (Lab Handout 6).   Later, due to the results with API 20E being semi-inconclusive, and with more information on the isolated bacteria, the API staph strip was used to test the bacteria.

To determine if the isolated bacteria had the ability to grow in the presence of methylene blue, and further to see if it could ferment lactose and/or sucrose, the isolate was plated onto Eosin Methylene Blue (EMB), following lab 6 protocol.   For determining if the isolate had the ability to grow in the presence of crystal violet and bile salts, and to further determine capability to ferment lactose, the isolate was plated on MacConkey (MAC) Agar (Lab Handout 6).

Additionally, to further determine the identity of the isolated bacteria, its susceptibility or resistance to various antibiotics was tested via the protocol of Lab Handout 9.  The isolated bacteria was suspended in TSB and then plated with a cotton swab onto two TSA plates, onto which antibiotic discs were placed evenly apart.  To determine susceptibility to antibiotics, zone diameter was measured once a bacterial lawn had time to form (approximately 48 hours past inoculation).  The tested antibiotics were Amakicin, Cefaperazone, Vancomycin, Trimethoprim, Oxacillin, Clindamycine, Piperacillin, and Gentamicin.


An isolated pure culture bacteria colony was suspended into tryptic soy broth (TSB).   After a week, this TSB suspension of bacteria was used for genomic DNA extraction (Lab Handout 5).     DNA extraction was done using the PowerSoil DNA kit (MoBio), which was sequenced using Illumina MiSeq technology at the University of Alaska Fairbanks’s DNA Core Lab (Lab Handout 5).   The data procured from the Illumina Miseq sequence analysis will then be computed using BaseSpace (Lab Handout 7).  Assembly of the genome was done using the SPAdes Genome Assembler application (Lab Handout 7).  Using the Kraken Metagenomics application and the assembled genome, a taxonomic classification of the bacterial isolate was determined (Lab Handout 7).  The final application used on the data procured from the DNA Core Lab sequencing was the Prokka Genome Annotation to determine potential functional genes of the genes (Lab Handout 7).

To further analyze the results, the first node of the contigs.fasta file was pulled from the SPAdes Genome Assembly application and ran through the Nucleotide BLAST (Basic Local Alignment Search Tool) application provided by the U.S. National Library of Medicine.



The pure, isolated bacteria grew in yellow-pigmented, shiny, smooth, small, and roung colonies.  Gram-staining revealed the bacteria to Gram-positive in nature, with cocci approximately 1.5 um in diameter.  The cells were usually found in tetrads with the smallest aggregations being diplococci, though often the cells would clump further.   The fluid thioglycollate test determined the bacteria to be an obligate aerobe.   Hydrogen peroxide resulted in formation of bubbles, testing positive on the catalase test.   The oxidase test strip turned a light grey-purple in the presence of the bacterial suspension, testing positive on the oxidase test.   The API 20E tests came back predominantly negative, with the only reliable positive being the test for presence of gelatinase.   Additionally, the VP turned a slight pink within the 10 minutes after VP1 and VP2 reagents were both added.  This was inconclusive as to whether the test was positive or negative.  With the API Staph strip, there were also all negatives within 48 hours of starting the strip.   After addition of the VP and NIT reagents (there was no access to PAL reagents, so that test is inconclusive), the VP turned slight pink after the 10 minutes.  Using apiweb as a resource, I determined this to be a positive reaction.

Antibiotic testing results (Table 1) reveal this isolate to be fairly susceptible to antibiotics.

Table 1: Antibiotic Zone Diameters and Susceptibility
Antibiotic Zone Diameter (mm) Susceptible, Intermediate, or Resistant
Amakicin *radius: 22 Susceptible
Cefaperazone *radius: 18 Susceptible
Vancomycin 29 Susceptible
Trimethoprim 15 Intermediate
Oxacillin 15** Resistant**
Clindamycine 36 Susceptible
Piperacillin 50 Susceptible
Gentamicin 36 Susceptible
*Radius was used where the zones overlapped and the diameter was impossible to measure due to overlap.  The radius of each zone was greater than the determined diameter susceptibility range.

**Plated oxacillin disc was contaminated with a piece of agar from another lab mate’s bacterial isolate, and so would be inconclusive, but my bacterial isolate was used by another student that lost their isolate in a lab clean-up.  This student also used the Oxacillin disc and had a zone diameter in the resistance range.



Basespace results indicate that the bacteria is likely Micrococcus luteus NCTC 2655 with 75.83% of the reads classified and 98.59% of the 73.91% analyzed reads were classified to the species level.   This is further supported by the BLAST results, which indicated a 97% identity match to Micrococcus luteus NCTC 2665.   The SPAdes Genome Assembler data revealed a genomic length of 2,615,526 base pairs with a GC content of 72.41%.   The bacteria also had 53 tRNAs and 2365 coding regions.   Probable functional genes were discovered through use of the Prokka Genome Annotation application.  Putative acetyl Co-A acetyltransferase, putative oxidoreductase, and NADH dehydrogenase-like protein SA0941 were all probable functional genes found in the genome of the isolate.


The bacteria isolated from the fish tank filter was not a unique microbe, as was expected, but a ubiquitous microbe found in many niches.  Additionally, the bacteria was surprisingly an obligate aerobe, which was unexpected as with an environment comprised mainly of water.  The ability of the bacteria to survive in such an environment may be due to the ability of M. luteus to become quite dormant and awaken with the use of the Rpf gene (Mukamolova et al 2002; Young et al 2010).

Identification of the bacterial isolate through the genetic data alone is alright in cases where the said bacteria cannot be plated and isolated.  In the case of this study, looking at both the physiological and genetic aspects of the bacteria for being able to identify the bacteria as Micrococcus luteus NCTC 2665 strain is important.


Micrococcus luteus NCTC 2665 “Fleming Strain’ have a distinct yellow-pigmentation to them, likely an adaptation to protect against radiation (Young et al 2010).  The bacteria is quite ubiquitous, in that it is usually found on mammalian skin, but can also be found in soil as well as fresh and marine water (Akayli et al 2016; Rickard et al 2002; Young et al 2010).  This aligns well with the sampling location being a fish tank filter, as the bacteria are found in water, plus the filter is something often handled by people, thus either could have been the source of the M. luteus at this particular location.  The bacteria is known to be coccoid, gram-positive, and form mainly tetrads (Akayli et al 2016).  With the tetrads, clumping happens, which makes sense as M. luteus are observed to create coaggregates in fresh water with some other bacteria to form a biofilm (Rickard et al 2002).  Additionally, M. luteus tests positive for both catalase and oxidase, all of which agrees with the testing done for the isolated bacteria.

The result of being an obligate aerobe from the fluid thioglycollate test also falls in line with the literature on M. luteus (Young et al 2010).  The physical test results of the API Staph test, when placed into the apiweb resource, come back with a 99.1% identity as a Micrococcus species, with an oxidase positive and yellow colony morphology leaning the test results towards Micrococcus luteus.


With both the BaseSpace and BLAST results identifying the strain as M. luteus, especially with all the physiological evidence, it would be hard to refute that the isolated strain is, in fact, M. luteus NCTC 2665.  Beyond this, additional genetic evidence points to the identification of the strain obtained as M. luteus.  For one, the GC content of the isolated strain was 72.41%, while it is 73% for NCTC 2665 (Young et al 2010).  Strain NCTC 2665 has 2,501,097 bp, encodes 2,403 (of 2,458 total) genes to proteins, and has 48 tRNAs, which is all comparable to the isolated bacteria’s 2,615,526 bp, 2,365 coding genes, and 53tRNAs (Young et al 2010).  Of the genes, the oxidoreductase, CoA, and NADH-like protein mentioned beforehand are all important to M. luteus.  Cytochrome c oxidoreductase is a component of the respiratory chain and CoA is part of the citric acid cycle (Young et al 2010).

In discussing genes, it is important to relate some of them to physiological aspects that are seen.  A lack of glucokinase, for instance, results in M. luteus being unable to grow with glucose alone as a carbon source, which is what we witnessed through the API test strips having negative results for fermentation of glucose (Young et al 2010).  Additionally, the high susceptibility of M. luteus to antibiotics is possibly due to lack of the gene wblC, as it is a pleiotropic regulator in other bacteria that helps with resistance towards antibiotics (Young et al 2010).

Additional Discussion

The strain represented in this research is the “Fleming strain’ NCTC 2665, which is still known as Micrococcus luteus, but another strain of M. luteus —primarily ATCC 9341- has been reclassified as Kocuria rhizophila (Tang and Gillevet 2003).  In fact, the NCTC 2665 strain had, “recently’ (in accordance to a 2010 paper) been declared taxonomically separate from K. rhizophila (Young et al 2010).

Strain ATCC 9341 is different in a few ways, some being that it is oxidase negative compared to the oxidase positive nature of NCTC 2665, as well as that the strain ATCC 9431 forms acids from glucose and fructose, neither of which were positive tests on the API test strips performed, meaning that my bacteria isolate does not confer with the traits presented by the Micrococcus luteus ATCC 9431 strain that was reclassified as Kocuria rhizophila (Tang and Gillevet 2003).  It, indeed, appears that the isolated bacteria follows more closely with the “Fleming strain’ NCTC 2655 of M. luteus which has not been reclassified to Kocuria rhizophila.



Akayli, T., Albayrak, G., Ürkü, Ç, Çanak, Ö, & Yörük, E. (2015). Characterization of Micrococcus luteus and Bacillus marisflavi Recovered from Common Dentex (Dentex dentex) Larviculture System. Mediterranean Marine Science, 17(1). doi:10.12681/mms.1322

Mukamolova, G. V., Turapov, O. A., Young, D. I., Kaprelyants, A. S., Kell, D. B., & Young, M. (2002). A family of autocrine growth factors in Mycobacterium tuberculosis. Molecular Microbiology, 46(3), 623-635. doi:10.1046/j.1365-2958.2002.03184.x

Rickard, A., McBain, A., Ledder, R., Handley, P., & Gilbert, P. (2003). Coaggregation between freshwater bacteria within biofilm and planktonic communities. FEMS Microbiology Letters, 220(1), 133. doi:10.1016/S0378-1097(03)00094-6

Tang, J. S. and Gillevet, P. M. (2003). Reclassification of ATCC 9341 from Micrococcus luteus to Kocuria rhizophila. International Journal Of Systematic And Evolutionary Microbiology, 53(4), 995-997. doi:10.1099/ijs.0.02372-0

Young, M., Artsatbanov, V., Beller, H. R., Chandra, G., Chater, K. F., Dover, L. G., . . . Greenblatt, C. L. (2009). Genome Sequence of the Fleming Strain of Micrococcus luteus, a Simple Free-Living Actinobacterium. Journal of Bacteriology, 192(3), 841-860. doi:10.1128/jb.01254-09