This is what I've dreamed of for so long! Testing for microbes in our tanks!

[AquaBiomics]

Just received my Aquabiome test results for a 11+ year old non-filtered (no mechanical/chemical filtration, just live rock/live sand) 12g nano. Couple of my takeaways:

1. Diversity & balance higher than 50th percentile of tested reef tanks
2. Balance Score (Correlation with Typical Abundance) shows similarities and differences compared to the 'typical' reef tank
3. Huge relative abundance of Pelagibacteraceae (Gram-negative, rod-shaped, free-living Bacteria (Alphaproteobacteria), aerobic & chemoheterotrophic, previously called SAR11, thought to be the most abundant bacterial group in the ocean worldwide. Well-adapted for life in the low-nutrient waters of the open ocean. Require reduced sulfur compounds, glycine, and dissolved organic carbon for growth). Distant runner ups: Hyphomicrobiaceae (Gram-negative Bacteria (Alphaproteobacteria), mostly rod-shaped, some free-living, Mostly aerobic & chemoheterotrophic, some photoheterotrophic, Extremely diverse, widely distributed and highly abundant in marine habitats including open ocean, sediments, and algal biofilms. Degrade sulfur-containing compounds (e.g. sulfite, DMSP). Many use methylated amines (MA) as primary nitrogen source) & Rhodobacteraceae (Gram-negative Bacteria (Alphaproteobacteria), mostly rod-shaped, some free-living, Mostly aerobic & chemoheterotrophic, some photoheterotrophic, Extremely diverse, widely distributed and highly abundant in marine habitats including open ocean, sediments, and algal biofilms. Degrade sulfur-containing compounds (e.g. sulfite, DMSP). Many use methylated amines (MA) as primary nitrogen source)
4. Typical ammonia−oxidizing microbes (with the exception of Nitrososphaeraceae (0.00026), which apparently is not typically registered in samples).
5. Nitrite−oxidizing Nitrospiraceae lower than typical
6. No Cyanobacteria species found
7. No Fish pathogen species found
8. No Coral pathogen species found

From the test reports I've seen online, there appears to be a general trend of decreasing diversity with age. This tank actually has a higher diversity (66%) compared to the 50th percentile. This is interesting and perhaps a bit surprising because the only significant additional bacteria are from occasional new specimens (~ once or twice a year, perhaps, on average) and weekly feeding of live earthworms from the compost heap. The system has never been subjected to any commercial 'fix-it' products (Vibrant, Chemi-clean, etc.) and hasn't had a melt-down (knocks on wood). Both the live rock and most of the live sand are over 20 years old (came from a previous long-lived aquarium). Since the sand bed had a very light dusting of coca-cola colored brown algae in places on the sand bed I suspected cyano, but no Cyano in the report.

Any thoughts, Eli?


Ralph.

Hi Ralph,

Nice summary.

My perspective on your nitrifying microbes is this was a really healthy nitrifying community. Ammonia-oxidizing microbes were both abundant (upper half of the typical range) and diverse (3 groups) in your sample. Nitrite oxidizing microbes were pretty typical in abundance and diversity (very few samples show anything *but* Nitrospiraceae, like yours). I see your results as entirely positive here.

As far as where how have you maintained diversity? If you'd asked me whether weekly feedings of live earthworms affected the microbiome I'd have guessed probably not. But as far as I know you're the only one in the DB doing that... so it seems hard to ignore the possibility thats contributing. More and more, as I discuss clients results with them, its becoming clear that there is a huge variety of different foods and feeding practices out there. Live earthworms is certainly a new one for the list!

PaulB famously feeds live stuff to his tank regularly and also had quite high diversity for his tank age...

Your preservation of old live rock and live sand is an important factor to consider too. I know many of the tanks in the DB were not started with years to decades old live rock and sand.... so this may also be a relevant difference in your tank's history.

 

[AquaBiomics]

What was the difference between the 'swab' and the 'water'. was there any ammonia reducing bacteria in the free water? Thanks.

An important question for our discussions about where various microbes live in the tank.

Its not a part of the standard output but I did a little digging. Since Ralph has shared his results I'll use his results as an example to answer the question.

Ammonia oxidizers:
Nitrosomonadaceae -- both, 3.8-fold higher in biofilm
Nitrososphaeraceae -- only detected in biofilm
Cenarchaeaceae -- both, 5.1-fold higher in biofilm

Nitrite oxidizers:
Nitrospiraceae -- both, 2.6-fold higher in biofilm

This matches my general impression across multiple samples, although again this tank has relatively high levels of nitrifying microbes overall compared to the range I've tested. In some tanks neither group is detected. In many tanks, ammonia oxidizers are low and nitrite oxidizers below detection. I believe what is happening is the two groups vary together across different tanks, with nitrite oxidizers always at lower levels.

 

[MnFish1]

An important question for our discussions about where various microbes live in the tank.

Its not a part of the standard output but I did a little digging. Since Ralph has shared his results I'll use his results as an example to answer the question.

Ammonia oxidizers:
Nitrosomonadaceae -- both, 3.8-fold higher in biofilm
Nitrososphaeraceae -- only detected in biofilm
Cenarchaeaceae -- both, 5.1-fold higher in biofilm

Nitrite oxidizers:
Nitrospiraceae -- both, 2.6-fold higher in biofilm

This matches my general impression across multiple samples, although again this tank has relatively high levels of nitrifying microbes overall compared to the range I've tested. In some tanks neither group is detected. In many tanks, ammonia oxidizers are low and nitrite oxidizers below detection. I believe what is happening is the two groups vary together across different tanks, with nitrite oxidizers always at lower levels.

Fyi - please separate my samples... As it will be instructive if you can

 

[Nano sapiens]

Hi Ralph,

Nice summary.

My perspective on your nitrifying microbes is this was a really healthy nitrifying community. Ammonia-oxidizing microbes were both abundant (upper half of the typical range) and diverse (3 groups) in your sample. Nitrite oxidizing microbes were pretty typical in abundance and diversity (very few samples show anything *but* Nitrospiraceae, like yours). I see your results as entirely positive here.

As far as where how have you maintained diversity? If you'd asked me whether weekly feedings of live earthworms affected the microbiome I'd have guessed probably not. But as far as I know you're the only one in the DB doing that... so it seems hard to ignore the possibility thats contributing. More and more, as I discuss clients results with them, its becoming clear that there is a huge variety of different foods and feeding practices out there. Live earthworms is certainly a new one for the list!

PaulB famously feeds live stuff to his tank regularly and also had quite high diversity for his tank age...

Your preservation of old live rock and live sand is an important factor to consider too. I know many of the tanks in the DB were not started with years to decades old live rock and sand.... so this may also be a relevant difference in your tank's history.

Thanks, Eli, for the informative reply. Will respond to your email with details later on tonight...

 

[Scrubber_steve]

An important question for our discussions about where various microbes live in the tank.

Its not a part of the standard output but I did a little digging. Since Ralph has shared his results I'll use his results as an example to answer the question.

Ammonia oxidizers:
Nitrosomonadaceae -- both, 3.8-fold higher in biofilm
Nitrososphaeraceae -- only detected in biofilm
Cenarchaeaceae -- both, 5.1-fold higher in biofilm

Nitrite oxidizers:
Nitrospiraceae -- both, 2.6-fold higher in biofilm

This matches my general impression across multiple samples, although again this tank has relatively high levels of nitrifying microbes overall compared to the range I've tested. In some tanks neither group is detected. In many tanks, ammonia oxidizers are low and nitrite oxidizers below detection. I believe what is happening is the two groups vary together across different tanks, with nitrite oxidizers always at lower levels.

Sorry for the ignorant question - is it possible at all for (living) de-nitrifying bacteria to show up in a water sample?

 

[AquaBiomics]

Sorry for the ignorant question - is it possible at all for (living) de-nitrifying bacteria to show up in a water sample?

Not ignorant at all, I have often wished it were easier to include this information.

While these are mostly surface associated its becoming increasingly clear that surface associated groups show up in the water samples. I have no reason to think these will be any different.

The major obstacle IMO is that the range of microbes that engage in denitrification is so broad. Instead of a short list of families we'd be facing a long list of genera and species... which brings its own challenges since a single genetic marker is not always able to classify a microbe to the species level. I am not saying its impossible, but has appeared too difficult to do properly given the time and resources available right now. Its on the list of future upgrades I'd like to add.

 

[AquaBiomics]

Fyi - please separate my samples... As it will be instructive if you can

Will do. I always sequence them separately but have not chosen to complicate the report further by splitting them in the report. But I can run the same analysis on yours, and consider including it in the report if its generally of interest.

 

[Huskymaniac]

@AquaBiomics Eli, quick question for you. It would be interesting to know how much of a non pathogenic/pathogenic reference would it take to get a positive result. In other words, what’s the limit of detection of the system for a single species of bacteria?

 

[Nano sapiens]

Hi Eli - Wondering if you might have test results from a healthy reef in nature. Would be interesting to compare with our captive reef systems.

 

[AquaBiomics]

@AquaBiomics Eli, quick question for you. It would be interesting to know how much of a non pathogenic/pathogenic reference would it take to get a positive result. In other words, what’s the limit of detection of the system for a single species of bacteria?

Important but also a challenging question. Its not easy to put an exact number on the answer, because its affected by a few unknowns. I estimate one per several thousand cells.

To increase sensitivity further we could make small improvements by spending a lot more money on sequencing, or larger improvements by filtering a larger volume of water.

It would be interesting to answer your question experimentally by adding a known concentration of cells that are not already present in the aquarium, then immediately sampling to measure their relative abundance. Of course it will be affected by the existing concentration of cells in the aquarium, I will have to think some more about how to measure the limit effectively this way.

 

[Huskymaniac]

Important but also a challenging question. Its not easy to put an exact number on the answer, because its affected by a few unknowns. I estimate one per several thousand cells.

To increase sensitivity further we could make small improvements by spending a lot more money on sequencing, or larger improvements by filtering a larger volume of water.

It would be interesting to answer your question experimentally by adding a known concentration of cells that are not already present in the aquarium, then immediately sampling to measure their relative abundance. Of course it will be affected by the existing concentration of cells in the aquarium, I will have to think some more about how to measure the limit effectively this way.

Thanks Eli for responding. What I am trying to figure out is where I picked up my myco I fection from. Unfortunately I had 3 QT tanks running plus 3 display tanks. From a sensitivity standpoint is that 1 per several thousand pretty close to 100% that a specific bacteria is not present in the aquarium?

 

[MnFish1]

Thanks Eli for responding. What I am trying to figure out is where I picked up my myco I fection from. Unfortunately I had 3 QT tanks running plus 3 display tanks. From a sensitivity standpoint is that 1 per several thousand pretty close to 100% that a specific bacteria is not present in the aquarium?

It only takes 1 - and bad luck. Im not sure there is an answer to your question.

 

[Huskymaniac]

It only takes 1 - and bad luck. Im not sure there is an answer to your question.

I would think that if there was enough bacteria in the tank to infect me that the test would be sensitive enough to pick it up. That's kind of what my question was? I wish I still had my old QT tanks up and running to see what was in there also.

 

[AquaBiomics]

Thanks Eli for responding. What I am trying to figure out is where I picked up my myco I fection from. Unfortunately I had 3 QT tanks running plus 3 display tanks. From a sensitivity standpoint is that 1 per several thousand pretty close to 100% that a specific bacteria is not present in the aquarium?

Legal disclaimers first, I cannot offer medical advice and would have to say the same if the bug was detected.

But speaking generally about detecting either organisms or genes with DNA sequencing, its generally hard to be certain about a negative, right? Not saying that either to dodge the question or to be snarky. Its genuinely something we often would like to know (is this thing truly absent) but its never easy to be sure.

One thing in our favor, M. marinum is free living and infections occur from exposure to the water itself. So our sampling method should be able to capture it. The sample of 60 ml is expected to contain over 6 million cells based on the typical range of microbial densities in aquarium water. So we're sampling this population pretty deeply, millions of cells. Overall our sampling should be a reasonably effective way of capturing the cells if they are in the aquarium.

So the DNA I extracted and prepared for sequencing should reflect contributions from millions of microbes. Our sequencing of 10,000 reads didn't turn it up, but perhaps its there at very low levels (1 in 1 million?) I cannot rule that out. Sequencing that deeply would be prohibitively costly. And even then we'd wonder what if its here at 1 per 10 million? etc.

It looks like there are PCR based methods for specific detection of M. marinum. These will be more sensitive than the general 16S primers I used to amplify (in principle) all Bacteria & Archaea. I have your DNA samples in hand, so this will be an interesting test case. If its absent from the standard 16S tests, is it also absent from the more sensitive species specific test? It won't be instant, but I can order those primers with my next batch and keep you posted. I'll check this at the level of PCR, so we won't have to wait for sequencing results to come back.

 

[AquaBiomics]

I would think that if there was enough bacteria in the tank to infect me that the test would be sensitive enough to pick it up. That's kind of what my question was? I wish I still had my old QT tanks up and running to see what was in there also.

Yeah, a positive control would be great! Thats why I'm always interested to find the rare tank with a fish or coral pathogen, so I can use it as a positive control for comparisons.

 

[Huskymaniac]

Legal disclaimers first, I cannot offer medical advice and would have to say the same if the bug was detected.

But speaking generally about detecting either organisms or genes with DNA sequencing, its generally hard to be certain about a negative, right? Not saying that either to dodge the question or to be snarky. Its genuinely something we often would like to know (is this thing truly absent) but its never easy to be sure.

One thing in our favor, M. marinum is free living and infections occur from exposure to the water itself. So our sampling method should be able to capture it. The sample of 60 ml is expected to contain over 6 million cells based on the typical range of microbial densities in aquarium water. So we're sampling this population pretty deeply, millions of cells. Overall our sampling should be a reasonably effective way of capturing the cells if they are in the aquarium.

So the DNA I extracted and prepared for sequencing should reflect contributions from millions of microbes. Our sequencing of 10,000 reads didn't turn it up, but perhaps its there at very low levels (1 in 1 million?) I cannot rule that out. Sequencing that deeply would be prohibitively costly. And even then we'd wonder what if its here at 1 per 10 million? etc.

It looks like there are PCR based methods for specific detection of M. marinum. These will be more sensitive than the general 16S primers I used to amplify (in principle) all Bacteria & Archaea. I have your DNA samples in hand, so this will be an interesting test case. If its absent from the standard 16S tests, is it also absent from the more sensitive species specific test? It won't be instant, but I can order those primers with my next batch and keep you posted. I'll check this at the level of PCR, so we won't have to wait for sequencing results to come back.

If it's not to much trouble and I can certainly pay for the test. Just trying to pinpoint this as close to 100% as possible.

 

[Huskymaniac]

Legal disclaimers first, I cannot offer medical advice and would have to say the same if the bug was detected.

But speaking generally about detecting either organisms or genes with DNA sequencing, its generally hard to be certain about a negative, right? Not saying that either to dodge the question or to be snarky. Its genuinely something we often would like to know (is this thing truly absent) but its never easy to be sure.

One thing in our favor, M. marinum is free living and infections occur from exposure to the water itself. So our sampling method should be able to capture it. The sample of 60 ml is expected to contain over 6 million cells based on the typical range of microbial densities in aquarium water. So we're sampling this population pretty deeply, millions of cells. Overall our sampling should be a reasonably effective way of capturing the cells if they are in the aquarium.

So the DNA I extracted and prepared for sequencing should reflect contributions from millions of microbes. Our sequencing of 10,000 reads didn't turn it up, but perhaps its there at very low levels (1 in 1 million?) I cannot rule that out. Sequencing that deeply would be prohibitively costly. And even then we'd wonder what if its here at 1 per 10 million? etc.

It looks like there are PCR based methods for specific detection of M. marinum. These will be more sensitive than the general 16S primers I used to amplify (in principle) all Bacteria & Archaea. I have your DNA samples in hand, so this will be an interesting test case. If its absent from the standard 16S tests, is it also absent from the more sensitive species specific test? It won't be instant, but I can order those primers with my next batch and keep you posted. I'll check this at the level of PCR, so we won't have to wait for sequencing results to come back.

Do you have an ETA on when you think you will have the PCR results back? Not that I am inpatient just dont want too bother you if it's going to take a month, etc.

 

[AquaBiomics]

Do you have an ETA on when you think you will have the PCR results back? Not that I am inpatient just dont want too bother you if it's going to take a month, etc.

After reviewing the options I chose a two step process that can be done with equipment on hand. It will take a month to order reagents and test each step, in principle its a rapid assay when the materials are on hand and tested. So give me a month this time. A week in the future if we ever wanted to retest a tank.

Its an important enough bug that this was a good one to focus on. And the general approach here (confirming results from DNA sequencing with a rapid test I can complete in-house) could be useful for a few cases like this.

 

[Huskymaniac]

After reviewing the options I chose a two step process that can be done with equipment on hand. It will take a month to order reagents and test each step, in principle its a rapid assay when the materials are on hand and tested. So give me a month this time. A week in the future if we ever wanted to retest a tank.

Its an important enough bug that this was a good one to focus on. And the general approach here (confirming results from DNA sequencing with a rapid test I can complete in-house) could be useful for a few cases like this.

Thanks and definitely agree. Am super appreciative of running the test to confirm the negative tests. Once that is done it would be left down to LFS Counter (water in open wound), 3 QT tanks, or my Anemone tank.

 

[Huskymaniac]

After reviewing the options I chose a two step process that can be done with equipment on hand. It will take a month to order reagents and test each step, in principle its a rapid assay when the materials are on hand and tested. So give me a month this time. A week in the future if we ever wanted to retest a tank.

Its an important enough bug that this was a good one to focus on. And the general approach here (confirming results from DNA sequencing with a rapid test I can complete in-house) could be useful for a few cases like this.

I also just put in order in for another kit. Might as well test my anemone tank also.