Rapid Changes In Microbial Communities In A New Aquarium

[Dan_P]

@Hans-Werner recently shared the paper

The Role of Ammonia-Oxidizing Archaea During Cycling and Animal Introduction in a Newly Commissioned Saltwater Aquarium

It is an account of the start up of a 5000 gallon system at the John G. Shedd aquarium to house seahorses. Daily water samples were taken during the 54 day set up process and the microbial DNA analyzed. Archaea was found to be the dominant microorganism responsible for ammonia oxidation to nitrite, not bacteria. The other interesting information was the 54 day collection of replicate DNA analyses of the microbial community. I arranged this data in chronological order and share it in this post.

Briefly, the commissioning process started with only the 750 gallon sump half filled with seawater to which a slurry of squid and fish was added to start up ammonia oxidation. There were some other adjustments made over 36 days before the aquarium was filled and connected to the sump. I listed these adjustments below. Then live food was added followed by the seahorses and spiny urchins. The DNA plot of relative percent of DNA (sorry about eye test) is below the list of adjustments. There are two bar charts per day representing replicate analyses. The names of some of the abundant families are listed below the DNA plot.

While the big news is Archaea was found to be doing the lion’s share of ammonia oxidation (the big purple splotch on the DNA plot), what struck me was how quickly (24 hours) and how much (large color bars appear and disappear) the microbial community changes after a seemingly small change. For example, compare the bars in section A to B for a 5 C temperature increase, or I to J when there was a large reduction in Archaea after mysids were added. Also notice the rapid increase in Vibrionacaea when the seahorses were added. There is much to ponder.

I come away from this paper wondering about the stability of microbial communities in more established systems. Are one-off microbiome tests telling us anything useful if twenty four later the microbiome could change?


Adjustments Made To The Sump And System

A-400 gallons Instant Ocean into 750 gallon sump + squid and fish slurry to start ammonia oxidation

B-5 C temperature increase

C-Ammonia additions start

D-Topping off sump with saltwater

E-Backwashing sand filter

F-Topping off sump with saltwater

G-About a 2 C Drop in temperature

H-Topping of sump with freshwater

I-5000 gallon aquarium filled and connected to sump

J-Adult Mysids + nauplii

K-Seadragons (seahorse) and Thorny Urchins

 

[Randy Holmes-Farley Verified Community Expert]

Very interesting and thanks for posting it. :)

I agree the changes seem fast and does question the utility of testing in a tank where things are changing.

I would make a few comments about conclusions, and I think this sort of analysis can generate more questions than provide answers without a lot of related experiments.

I agree there's a big apparent change with some procedures (say, adding mysids), but that does not necessarily mean anything declined. It could just mean that some other things increased.

I've also often thought there was a disconnect between detected DNA and assuming those amounts/percentages are an indication of the amounts of organisms. How does the DNA get into the water, and is there any a priori reason to think that is the same for all organisms? Simple shed cells? Dead cells broken open? Does the nature of an organism that is suspended mean it shows up a lot while a benthic one shows up less for the same number of cells? I've never seen a discussion of that, though I admit I've also not looked for it.

Suppose that there are two organisms on the sand that are present in equal proportion, but one is 10x or 100x more likely to have its DNA end up in the water, this sort of analysis could be misinterpreted that there is more of the organism releasing more DNA.

FInally, even if there actually is 100x more ammonia oxidizing organism A than ammonia oxidizing organism B, that still cannot say which is oxidizing more total ammonia from the water.

 

[threebuoys]

I'm way out of my league here. I will pose what are probably ridiculous questions.
The chart seems to suggest that minor adjustments to the environment possibly results in significant changes/differences. Would it be logical to assume no two aquariums are identical and that point in time analysis of such information has little value to the average hobbyist? Likewise, trends overtime likely vary significantly from one environment to another. Unless the analysis points out either good or bad conditions that can be measured and/or managed/sustained, how can the data be used?

You chemistry nerds that look for and understand this complicated stuff are amazing!!

 

[Dan_P]

Very interesting and thanks for posting it. :)

I agree the changes seem fast and does question the utility of testing in a tank where things are changing.

I would make a few comments about conclusions, and I think this sort of analysis can generate more questions than provide answers without a lot of related experiments.

I agree there's a big apparent change with some procedures (say, adding mysids), but that does not necessarily mean anything declined. It could just mean that some other things increased.

I've also often thought there was a disconnect between detected DNA and assuming those amounts/percentages are an indication of the amounts of organisms. How does the DNA get into the water, and is there any a priori reason to think that is the same for all organisms? Simple shed cells? Dead cells broken open? Does the nature of an organism that is suspended mean it shows up a lot while a benthic one shows up less for the same number of cells? I've never seen a discussion of that, though I admit I've also not looked for it.

Suppose that there are two organisms on the sand that are present in equal proportion, but one is 10x or 100x more likely to have its DNA end up in the water, this sort of analysis could be misinterpreted that there is more of the organism releasing more DNA.

FInally, even if there actually is 100x more ammonia oxidizing organism A than ammonia oxidizing organism B, that still cannot say which is oxidizing more total ammonia from the water.

I think you have created an excellent list of questions and reservations for an expert in this field to address and help us interpret this type of data.

I played around with a made up DNA composition of six organisms. Just increasing the amount of one organism’s data shrinks all the other organism’s percent although the hint that this was the situation is that the ratios of percents of the unchanged organisms stay the same. Also, if a percent DNA doesn’t change while that of another organism balloons means it actually increased to maintain its share. I was going to do this ratio analysis on the data when it dawned on me that the classifications at the family level could include hundreds of species. From a functional standpoint, i.e., what it means to the aquarium or microbial community, I couldn’t conclude anything. Who knows, this post might draw some experts into giving us some insight.

 

[Dan_P]

I'm way out of my league here. I will pose what are probably ridiculous questions.
The chart seems to suggest that minor adjustments to the environment possibly results in significant changes/differences. Would it be logical to assume no two aquariums are identical and that point in time analysis of such information has little value to the average hobbyist? Likewise, trends overtime likely vary significantly from one environment to another. Unless the analysis points out either good or bad conditions that can be measured and/or managed/sustained, how can the data be used?

You chemistry nerds that look for and understand this complicated stuff are amazing!!

These are very good questions!

And brings to mind a study referenced in the Shedd Aquarium paper that studied two “identical” systems. I will dig that up.

 

[BCSreef]

Thank you very much, Dan! Some fascinating information. I’m traveling the next three weeks, but look forward to having some time to study and more detail went back in the US. Many years ago has part of my job. I’ve done a few studies on mixed bacterial cultures and concluded that it doesn’t take much to shift the population balance.

 

[Tripod1404]

Very interesting and thanks for posting it. :)

I agree the changes seem fast and does question the utility of testing in a tank where things are changing.

I would make a few comments about conclusions, and I think this sort of analysis can generate more questions than provide answers without a lot of related experiments.

I agree there's a big apparent change with some procedures (say, adding mysids), but that does not necessarily mean anything declined. It could just mean that some other things increased.

I've also often thought there was a disconnect between detected DNA and assuming those amounts/percentages are an indication of the amounts of organisms. How does the DNA get into the water, and is there any a priori reason to think that is the same for all organisms? Simple shed cells? Dead cells broken open? Does the nature of an organism that is suspended mean it shows up a lot while a benthic one shows up less for the same number of cells? I've never seen a discussion of that, though I admit I've also not looked for it.

Suppose that there are two organisms on the sand that are present in equal proportion, but one is 10x or 100x more likely to have its DNA end up in the water, this sort of analysis could be misinterpreted that there is more of the organism releasing more DNA.

FInally, even if there actually is 100x more ammonia oxidizing organism A than ammonia oxidizing organism B, that still cannot say which is oxidizing more total ammonia from the water.

Yeah this is also my main question about these types of metagenomics studies as well. In fact, I am more inclined to think the results can represent the opposite of what the raw data suggests.

Most bacteria and archeae we have on our systems are benthic and/or biofilm forming species. So a large increase is DNA from a particular bacteria or archeae species can very well indicate die off. Like I I were to dump antibiotics into a system, I am certain that there will be a massive spike in DNA from certain species. But that wouldn’t indicate that they are thriving, it indicates that cells are dying and or detaching from the surfaces.

In my opinion, these experiments are similar to trying to catalogue insects in an area by driving around and analyzing the dead bugs on a cars windshield. Results would indicate that there are a lot of bees and flys and almost no ants and spiders.

 

[Dan_P]

I think you have created an excellent list of questions and reservations for an expert in this field to address and help us interpret this type of data.

I played around with a made up DNA composition of six organisms. Just increasing the amount of one organism’s data shrinks all the other organism’s percent although the hint that this was the situation is that the ratios of percents of the unchanged organisms stay the same. Also, if a percent DNA doesn’t change while that of another organism balloons means it actually increased to maintain its share. I was going to do this ratio analysis on the data when it dawned on me that the classifications at the family level could include hundreds of species. From a functional standpoint, i.e., what it means to the aquarium or microbial community, I couldn’t conclude anything. Who knows, this post might draw some experts into giving us some insight.

OK, here’s the abstract. The study was on a new set up, not an established aquariu

Microbial Community Succession and Nutrient Cycling
Responses following Perturbations of Experimental Saltwater
Aquaria

Holly M. Bik,*a Alexandra Alexiev,a Sabreen K. Aulakh,a Lakshmi Bharadwaj,a Jennifer Flanagan,a John M. Haggerty,b
Sarah M. Hird,a,c Guillaume Jospin,a Jenna M. Lang,a Laura A. Sauder,d Josh D. Neufeld,d Andrew Shaver,a
Akshay Sethi,a Jonathan A. Eisen,e,f David A. Coila
aUC Davis Genome Center, University of California—Davis, Davis, California, USA
bDepartment of Biology, San Diego State University, San Diego, California, USA
cDepartment of Molecular and Cell Biology, University of Connecticut, Storrs, Connecticut, USA
dDepartment of Biology, University of Waterloo, Waterloo, Ontario, Canada
eDepartment of Evolution and Ecology, University of California—Davis, Davis, California, USA
fDepartment of Medical Microbiology and Immunology, University of California—Davis, Davis, California, USA


ABSTRACT Although aquaria are common features of homes and other buildings,little is known about how environmental perturbations (i.e., tank cleaning, waterchanges, addition of habitat features) impact the diversity and succession of aquarium microbial communities. In this study, we sought to evaluate the hypotheses thatnewly established aquaria show clear microbial successional patterns over time and
that common marine aquarium-conditioning practices, such as the addition of ocean-derived “live rocks” (defined as any “dead coral skeleton covered with crustose coralline algae” transferred into an aquarium from open ocean habitats) impactthe diversity of microbial populations as well as nitrogen cycling in aquaria. We col-
lected water chemistry data alongside water and sediment samples from two independent and newly established saltwater aquaria over a 3-month period. Microbial communities in samples were assessed by DNA extraction, amplification of the 16SrRNA gene, and Illumina MiSeq sequencing. Our results showed clear and replicablepatterns of community succession in both aquaria, with the existence of multiple
stable states for aquarium microbial assemblages. Notably, our results show that changes in aquarium microbial communities do not always correlate with water chemistry measurements and that operational taxonomic unit (OTU)-level patterns relevant to nitrogen cycling were not reported as statistically significant. Overall, our results demonstrate that aquarium perturbations have a substantial impact on microbial community profiles of aquarium water and sediment and that the addition oflive rocks improves nutrient cycling by shifting aquarium communities toward amore typical saltwater assemblage of microbial taxa.

 

[Dan_P]

Thank you very much, Dan! Some fascinating information. I’m traveling the next three weeks, but look forward to having some time to study and more detail went back in the US. Many years ago has part of my job. I’ve done a few studies on mixed bacterial cultures and concluded that it doesn’t take much to shift the population balance.

Hurry home. We will be waiting for your insights!

 

[Hans-Werner]

I agree there's a big apparent change with some procedures (say, adding mysids), but that does not necessarily mean anything declined. It could just mean that some other things increased.

This is the reason why I have set exclamation marks in "relative(!)" in my response to Lasse. It just means, other DNA has increased. It does not necessarily mean the number or activity of AOA has decreased.

I've also often thought there was a disconnect between detected DNA and assuming those amounts/percentages are an indication of the amounts of organisms. How does the DNA get into the water, and is there any a priori reason to think that is the same for all organisms? Simple shed cells? Dead cells broken open? Does the nature of an organism that is suspended mean it shows up a lot while a benthic one shows up less for the same number of cells? I've never seen a discussion of that, though I admit I've also not looked for it.

There is quite a number of publications on the role of AOA in lower ammonia waters. It is worth to read a few. I also have linked a few above. Of course the authors must have read most of the previous publications and their article is in line with these publications.

 

[jeremie]

OK, here’s the abstract. The study was on a new set up, not an established aquariu

Microbial Community Succession and Nutrient Cycling
Responses following Perturbations of Experimental Saltwater
Aquaria

Holly M. Bik,*a Alexandra Alexiev,a Sabreen K. Aulakh,a Lakshmi Bharadwaj,a Jennifer Flanagan,a John M. Haggerty,b
Sarah M. Hird,a,c Guillaume Jospin,a Jenna M. Lang,a Laura A. Sauder,d Josh D. Neufeld,d Andrew Shaver,a
Akshay Sethi,a Jonathan A. Eisen,e,f David A. Coila
aUC Davis Genome Center, University of California—Davis, Davis, California, USA
bDepartment of Biology, San Diego State University, San Diego, California, USA
cDepartment of Molecular and Cell Biology, University of Connecticut, Storrs, Connecticut, USA
dDepartment of Biology, University of Waterloo, Waterloo, Ontario, Canada
eDepartment of Evolution and Ecology, University of California—Davis, Davis, California, USA
fDepartment of Medical Microbiology and Immunology, University of California—Davis, Davis, California, USA


ABSTRACT Although aquaria are common features of homes and other buildings,little is known about how environmental perturbations (i.e., tank cleaning, waterchanges, addition of habitat features) impact the diversity and succession of aquarium microbial communities. In this study, we sought to evaluate the hypotheses thatnewly established aquaria show clear microbial successional patterns over time and
that common marine aquarium-conditioning practices, such as the addition of ocean-derived “live rocks” (defined as any “dead coral skeleton covered with crustose coralline algae” transferred into an aquarium from open ocean habitats) impactthe diversity of microbial populations as well as nitrogen cycling in aquaria. We col-
lected water chemistry data alongside water and sediment samples from two independent and newly established saltwater aquaria over a 3-month period. Microbial communities in samples were assessed by DNA extraction, amplification of the 16SrRNA gene, and Illumina MiSeq sequencing. Our results showed clear and replicablepatterns of community succession in both aquaria, with the existence of multiple
stable states for aquarium microbial assemblages. Notably, our results show that changes in aquarium microbial communities do not always correlate with water chemistry measurements and that operational taxonomic unit (OTU)-level patterns relevant to nitrogen cycling were not reported as statistically significant. Overall, our results demonstrate that aquarium perturbations have a substantial impact on microbial community profiles of aquarium water and sediment and that the addition oflive rocks improves nutrient cycling by shifting aquarium communities toward amore typical saltwater assemblage of microbial taxa.

I reanalyzed some of their nitrifier sequences using a newer database when I first read the paper a few years ago. It seems that the majority of AOA from both aquariums and live rocks from the ocean also belong to the genus Nitrosopumilus, just like in the study by Oliaro et al.

Application of the consortia of nitrifying archaea and bacteria for fish transportation may be beneficial for fish trading and aquaculture​

This study also found that Nitrosopumilus is the main nitrifier in their saltwater aquarium biofilter. Interestingly, the dominant genus in Dr. Tim's bottled product is actually Nitrosocosmicus, which has been shown to have a lower affinity for ammonia compared to other AOA ( Jung et al., 2021). This made me wonder whether other cycling products also contain AOA, and whether the difference in performance is actually the result of differences in taxa.

 

[Randy Holmes-Farley Verified Community Expert]

Going back to the various uncertainties in going from dna concentrations to microbe numbers to microbe activities, this paper suggests that at the same microbe mass, AOA can be much more active than AOB, at least under some conditions:

Hence, we predict that, in direct nutrient-limited competition, the ammonium turnover per unit bio-
mass of Nitrosopumilus-like AOA would be at least 5 times higher than of oligotrophic heterotrophs, and more than 30 times higher than of the most oligotrophic diatoms known so far.

https://www.researchgate.net/public...fying_Archaea_and_Bacteria_Nature_461_976-981

 

[rtparty]

This is all well above my pay grade but I did watch the BRS Biome Series again the other day and my opinion is they also saw pretty rapid changes in that series. They were testing every 2 weeks

This is the part where we try to line up multiple tests and datasets to find the replication between them all

 

[Dan_P]

Going back to the various uncertainties in going from dna concentrations to microbe numbers to microbe activities, this paper suggests that at the same microbe mass, AOA can be much more active than AOB, at least under some conditions:

Hence, we predict that, in direct nutrient-limited competition, the ammonium turnover per unit bio-
mass of Nitrosopumilus-like AOA would be at least 5 times higher than of oligotrophic heterotrophs, and more than 30 times higher than of the most oligotrophic diatoms known so far.

https://www.researchgate.net/public...fying_Archaea_and_Bacteria_Nature_461_976-981

I am going to make a gigantic leap after reading the quote you posted (paper was interesting too).

It is the higher ammonia affinity of AOA that is behind the observation that aquaria started with live rock have less or no nuisance organism growth compared to those stared with dry rocks and bottled nitrifying bacteria.

This conjecture is a consequence of my other conjecture that ammonia availability drives much of the dinoflagellate and cyanobacteria issues. This ammonia is not detected in the water column because it never makes it off the surface. It is strictly cycled in and around biofilms. A biofilm dominated by AOA rather than AOB would would not have enough ammonia to allow nuisance organisms to proliferate.

So, if correct, when we start aquaria with bottled AOA (I have to write up the patent) instead of AOB, aquaria started with dry rock will more closely resemble aquaria with live rock (this might be stretching the idea too far).

 

[Dan_P]

I reanalyzed some of their nitrifier sequences using a newer database when I first read the paper a few years ago. It seems that the majority of AOA from both aquariums and live rocks from the ocean also belong to the genus Nitrosopumilus, just like in the study by Oliaro et al.

Application of the consortia of nitrifying archaea and bacteria for fish transportation may be beneficial for fish trading and aquaculture​

This study also found that Nitrosopumilus is the main nitrifier in their saltwater aquarium biofilter. Interestingly, the dominant genus in Dr. Tim's bottled product is actually Nitrosocosmicus, which has been shown to have a lower affinity for ammonia compared to other AOA ( Jung et al., 2021). This made me wonder whether other cycling products also contain AOA, and whether the difference in performance is actually the result of differences in taxa.

Interesting.

Question, are AOA considered strictly benthic or can they also be planktonic?

 

[Randy Holmes-Farley Verified Community Expert]

Interesting.

Question, are AOA considered strictly benthic or can they also be planktonic?

They can be planktonic.

 

[Marc G-L]

Many decades ago when I was a poor microbiology student supporting myself through campus work, I counted colliform bateria (fecal) from sampled state lakes and waterways. I actually plated water samples on slides etched with lines where I had to count and categorize the organisms (all kinds). What was distinctly noticeable from even the same lake was who did the sampling (some right from the shore---lazy samplers) others from deeper out. Variances also showed up based on the time of day (amount of sunlight).

My point being, I actually counted the bacteria and still the variances were often profound. Some organisms feed and grow off the waste of others and that alone can represent reason enough along with temp and salinity changes to skew every supposed trend.

 

[Dan_P]

Many decades ago when I was a poor microbiology student supporting myself through campus work, I counted colliform bateria (fecal) from sampled state lakes and waterways. I actually plated water samples on slides etched with lines where I had to count and categorize the organisms (all kinds). What was distinctly noticeable from even the same lake was who did the sampling (some right from the shore---lazy samplers) others from deeper out. Variances also showed up based on the time of day (amount of sunlight).

My point being, I actually counted the bacteria and still the variances were often profound. Some organisms feed and grow off the waste of others and that alone can represent reason enough along with temp and salinity changes to skew every supposed trend.

Great observation. Bacteria ecology in an aquarium might be equally heterogenous, especially on the surfaces.

Did you pursue a career in something close to microbiology?

 

[Dan_P]

They can be planktonic.

Then I suppose AOA could exhibit similar behavior as bottled AOB, initially oxidizing ammonia suspended in the water before settling on the surface. That might explain the observation of high abundance of AOA DNA in new systems followed by tiny amounts while the system maintains the ammonia oxidation function. My assumption is that DNA from microorganisms that live in biofilms might not be as abundant in a water sample as a planktonic microorganism.

 

[jeremie]

Interesting.

Question, are AOA considered strictly benthic or can they also be planktonic?

Different AOA also seem to have different habitat preferences, as shown in this study:

Phylogenetic diversity and distribution patterns of ammonia-oxidizing microorganisms in marine environments
"Among the classified AOA, Ca. Nitrosopelagicus, Nitrosoarchaeum, and Nitrosopumilus emerged as dominant clades with distinct habitat preferences. Ca. Nitrosopelagicus was most abundant in the water column, whereas Nitrosoarchaeum was prevalent in sediments, followed by water, and Nitrosopumilus exhibited a widespread distribution. The presence of distinct subclades within these groups further suggests functional differentiation, with potential adaptations to specific environmental conditions."

I am going to make a gigantic leap after reading the quote you posted (paper was interesting too).

It is the higher ammonia affinity of AOA that is behind the observation that aquaria started with live rock have less or no nuisance organism growth compared to those stared with dry rocks and bottled nitrifying bacteria.

This conjecture is a consequence of my other conjecture that ammonia availability drives much of the dinoflagellate and cyanobacteria issues. This ammonia is not detected in the water column because it never makes it off the surface. It is strictly cycled in and around biofilms. A biofilm dominated by AOA rather than AOB would would not have enough ammonia to allow nuisance organisms to proliferate.

So, if correct, when we start aquaria with bottled AOA (I have to write up the patent) instead of AOB, aquaria started with dry rock will more closely resemble aquaria with live rock (this might be stretching the idea too far).

This reminds me of your dino/diatom succession experiment. Did the dinos decline after you switched the nitrogen source from ammonia to nitrate?