Do we need to match Rabaey et al exactly, or is it possible that there is a multi-mineral on the market that is “good enough”?
@“Gerrit”#p178 Brilliant question, and the answer of course is no, we don’t need to match exactly. But how do we know which elements are optional?
At a 20ml volume I'd say that chemicals are cheap and time is expensive, so my instinct would be to start with something we know works, then remove elements we suspect (or hope) may not be needed.
>
remove elements we suspect (or hope) may not be needed.
That sounds time consuming :) But yeah, if our intent is to reproduce the experiments from Pous et al exactly, then we should start there. I was thinking we could look to see what trace element solutions are used for similar HOB experiments, and where they overlap.
What I mean is if we skip three elements and see no growth we have no useful data. We have no idea what prevented growth. If we replicate someone else’s work with the only difference being our bioreactor we at least know it wasn’t the lack of a certain trace mineral that killed it. That seems like the fastest route to success.
We absolutely could attempt to copy any other HOB experiment. My preference for Pous et al, is that if we see no growth, Narcís will have a better chance of seeing where we went wrong & is more likely to be willing to take a look than someone who doesn't know us.
The closest of the three you have sourced may be the [multi mineral veggie capsule](https://bigvits.co.uk/hog-multi-mineral.html#additional) as we have sodium and sulphate in our 'meso-nutrient' solution and we'd really rather avoid chlorine (as it will electrolyse to form bleach and kill the bugs). So if you add iron nickel and cobalt you at least should have all the elements they want. You then just need to hope that things like iodine and chromium aren't inhibitory and all the organics (which you wouldn't normally have in trace nutrients) don't mean that a non-HOB proliferates and makes one of the other nutrients rate limiting.
I'm wondering if a cheaper mineral pill, less bioavailable to humans (avoiding the organics), may be better for the HOBs. I tried enlisting AI:
Gemini hallucinated a [few products](https://gemini.google.com/app/679e42d2dc30d25b) that I couldn't find for sale anywhere, the ones it suggested that I could find seemed no better than those above.
Copilot gave me Sanatogen A-Z multivitamin & a [fertiliser](https://www.agroliquid.com/products/micronutrients/micro-1000/) that it warned is not intended for human consumption.
As you have two pioreactors, a really useful experiment would be to run one as per Pous et al and run a second with your mineral capsules.
@Gerrit & I have been doing a little background work on this. We haven’t yet managed to find a food grade source of the Rabaey et al micronutrients, but @Gerrit found Onyxmet who can sell enough of each of the 9 components to make 42L of the micronutrient solution for £53.50.
I'm reaching out to some industry contacts to see if anyone can source food grade equivalents. If you know anyone who may be able to help out, let us know.
If we draw a blank I'd suggest that our approach is for one of us to purchase the chemicals from Onyxmet, then package up enough dry chemical for others to make 2L (allowing for wastage). We can then:
-
try enrichment cultures with this non-food-grade media,
-
develop an easily available food-grade solution,
-
then start the enrichment culture with it from scratch.
Multus just came up in a funding conversation. Perhaps we should reach out to them about jointly developing a micro nutrient solution?
I was looking through the Waag BioHack Academy archives and they have a trace mineral solution recipe. They mention using water conditioners from aquarium stores as an alternative source of trace minerals, which I thought was interesting. Turns out you can buy it in tablet form:
1 Like
Also, I went looking through the DSMZ list of organisms, and most HOB strains they have use “Trace element solution SL-6”:
1 Like
I’m just wondering if one of the reasons that external electrolysis results in higher growth rates is that in-situ electrolysis results in a little HOCl formation from the trace minerals. The HOCl would most likely kill as many HOB as it came in contact with until it was consumed. I wonder how much impact there would be using a non-standard trace mineral solution that had fully beneficial anions…?
@NarcisP have you contemplated avoiding Cl2 in your trace nutrients?
HOCl generation can occur when high chlorides concentrations are present in the medium. To avoid this reaction to occur: 1) Use Platinized titanium as anode, 2) Avoid chloride addition to the medium. Trace minerals are not specially relevant due to its low content. Macronutrients (N, P) are the key. Specially, avoid ammonium chloride, you can use other ammonium salts (e.g., ammonium sulphate) or urea (synthetically produced and widely used as fertilizer).
1 Like
I was looking for food-grade sources of boron and noticed that these boron capsules use sodium borate (borax). I’m having a really hard time finding a food-grade source of boric acid (H3BO3), so maybe borax (Na2B4O7.10H2O) is the way to go?
According to this borax supplier in the US there is only technical grade borax, which is not approved for food or beverage use. But, since it’s non-synthetic and mined as a mineral salt, it’s possible to find nutritional suppliers using food grade containers and packaging methods for borax.
Cai from Multus offered us a pretty good price to make up 50x 100ml bottles of trace nutrient solution. All that we needed to do was provide them with the powders.
Sourcing chemicals to produce Rabaey, Yu or SL6 or even a combination of these that minimised chlorides (to avoid Cl2 then hypochlorous acid [bleach] production at the anode) but included all critical ions wasn’t so difficult.
But I had a bee in my bonnet about getting these food grade, so that we could say that our single cell protein strains have only ever been fed with food grade nutrients. Unfortunately we haven’t managed to find suppliers of food grade:
- NiCl2.6H2O: Nickel(II) Chloride Hexahydrate
- CoCl2.6H2O: Cobalt(II) Chloride Hexahydrate - could we use vitamin B12?
- H3BO3: Boric Acid - we found one US supplier, but they won’t ship overseas, and are making enquiries with a European supplier we have just found.
It seems very unlikely that we’ll find a food grade source for nickel. In this paper from 1970 it does appear that nickel is a required trace element for Xanthobacter autotrophicus, so we can’t just leave it out.
Something caught my eye while looking at a new amendment to EU regulations regarding maximum nickel levels in food that comes into force next year. Maybe a wild idea, but given the comparatively large amount of nickel allowed in seaweed (30-40 mg/kg), could we use seaweed as a source of trace minerals?
1 Like
I love the idea of using seaweed!
I was thinking that the question would be how we know if a particular seaweed has all the nutrients required. One relatively easy way would be running comparisons with full-range trace minerals and seaweed. I was thinking that the seaweed may not let the optimum culture grow, but conversely any given nutrient mix may conceivably optimise conditions for a sub-optimal strain. It could be that we have a better strain (in terms of taste, nutrition, etc.) in an enrichment culture but it is outcompeted by an inferior strain for us. Given our limited analytical capabilities we’d never know. So perhaps I’ve been overthinking it…
Perhaps the best route is to take the best ingredients (in terms of sustainability, availability, cost & consumer optics) and just see what grows with them. If it’s good, great. If nothing grows, try different ingredients. If that’s correct then the main thing is probably getting as many different cultures running as we can.
Earlier in this thread I mentioned SeaChem Flourish tabs:
I went through the whole list of Seachem products today and almost fell off my chair when comparing the ingredients of Seachem - Reef Trace to our list of trace elements:
- Boron
- Cobalt
- Copper
- Manganese
- Molybdenum
- Zinc
- Rubidium
- Nickel
- Vanadium
It contains everything from Yu et al and SL-6 and is only missing iron and calcium from Rabaey et al (which may not be required). It has some rubidium that may not be necessary, but everything else is useful for HOB! I’ve ordered a 100mL bottle so that I can finally start my first HOB experiment.
1 Like
Brilliant, and an excellent price, I’ve just bought a bottle myself - I see under Allergen Information on Amazon that they are Yam Free!, but more usefully, here’s the contents per 1 g:
| Element | Amount (mg) |
|---|---|
| Boron (B) | 0.028 |
| Cobalt (Co) | 0.0003 |
| Copper (Cu) | 0.032 |
| Manganese (Mn) | 0.085 |
| Molybdenum (Mo) | 0.003 |
| Zinc (Zn) | 0.169 |
| Rubidium (Rb) | 8E-05 |
| Nickel (Ni) | 3E-05 |
| Vanadium (V) | 2E-05 |
So I’ve been trying to calculate how Reef Trace compares to Yu et al and Rabaey et al. I used ChatGPT to calculate the proportions of each element (based on the molar mass of the compounds) and their amounts per gram (to be able to compare against the information provided by Seachem):
| Element | Reef Trace (mg / g) | Rabaey | Yu |
|---|---|---|---|
| Boron (B) | 0.028 | 0.00105 | 0.1048 |
| Cobalt (Co) | 0.0003 | 0.1086 | 0.1816 |
| Copper (Cu) | 0.032 | 0.000945 | 0.0008 |
| Manganese (Mn) | 0.085 | 0.0437 | 0.0262 |
| Molybdenum (Mo) | 0.003 | 0.0168 | 0.028 |
| Zinc (Zn) | 0.169 | 0.003358 | 0.081 |
| Rubidium (Rb) | 8.00E-05 | ||
| Nickel (Ni) | 3.00E-05 | 0.0109 | 0.0181 |
| Vanadium (V) | 2.00E-05 |
Plotting it on a graph:
So one element that Reef Trace could be deficient in compared to the others is cobalt, but the other elements are at least in the same ball park. That is if the calculations are correct - it may be a good idea for someone to double-check this.
Well I tried to do this with in my first chat with ChatGPT’s new Advanced Voice. I also did it in a spreadsheet and got the following when I exclude Rabaey’s iron and calcium (which we can easily get elsewhere):
| Reef Trace | Rabaey (ex Fe & Ca) | Yu | SL6 | |
|---|---|---|---|---|
| Zn | 53.26% | 23.60% | 14.92% | 14.80% |
| Mn | 26.79% | 19.51% | 5.46% | 5.42% |
| B | 8.82% | 0.74% | 34.41% | 34.14% |
| Cu | 10.08% | 0.52% | 1.67% | 2.43% |
| Ni | 0.01% | 4.16% | 3.24% | 3.21% |
| Mo | 0.95% | 10.03% | 7.80% | 7.74% |
| Co | 0.09% | 41.43% | 32.50% | 32.25% |
So yes, by my calculations cobalt is most lacking, but nickel is only fractionally better (the two hardest elements to source food grade, not that Reef Trace is food grade anyway). Molybdenum is also only 12% of the lowest of the three published trace minerals.
Glad to hear your results match mine. So I think we should be able to use Reef Trace as a trace mineral solution, adding 1mL per litre. It will be interesting to see if the lower levels of nickel and cobalt turn out to be rate-limiting.
Do you think we’ll then be selecting for HOB that can cope with low CoNiMo?
Definitely worth cracking on anyway to see what we get. But if the Reef Trace cultures underperform the literature, that would be the first thing to check.