This is a common symptom for our times. This is reported often in samples, and thus being examined if it reaches our threshold for inclusion as defined in A new specialized selection of suggestions links. It does.

Anxiety is likely a part of a feedback look — anxiety chemicals modify the microbiome resulting in more anxiety triggering chemicals being released.

Study Populations:

We have 3 symptom annotations that could be included

• Comorbid: High Anxiety [8.5 z-score x 45 samples]
• DePaul University Fatigue Questionnaire : Anxiety/tension [6.5 z-score x 63 samples]
• Condition: Generalized anxiety disorder [just 7 samples – so we exclude]

Taking the first two together we get 85 samples with an max z-score of 5.7

Symptom	Reference	Study
High Anxiety and Anxiety/tension	1115	85

• Bacteria Detected with z-score > 2.6: found 160 items, highest value was 5.7
• Enzymes Detected with z-score > 2.6: found 290 items, highest value was 5.7
• Compound Detected with z-score > 2.6: found No items

This hints that the enzymes being produced are likely more significant than the bacteria.

Interesting Significant Bacteria

All bacteria found significant (except 1) had too low levels. This is a common pattern found with these studies, it is not “bad bacteria bogie man bacteria” but an absence of “upstanding citizens bacteria”.

Bacteria	Reference Mean	Study	Z-Score
Actinobacillus pleuropneumoniae (species)	56	21	5.7
Haemophilus parainfluenzae (species)	1633	543	5.5
Haemophilus (genus)	1632	545	5.4
Actinobacillus porcinus (species)	183	68	5.2
Pasteurellaceae (family)	2001	760	5.1
Pasteurellales (order)	2001	760	5.1

This is echoed in some studies

• Early-Life Stress Modulates Gut Microbiota and Peripheral and Central Inflammation in a Sex-Dependent Manner [2021]

Interesting Enzymes

All enzymes found significant had too low levels.

Enzyme	Reference Mean	Study Mean	Z-Score
HslU—HslV peptidase; (3.4.25.2)	7177	3218	5.7
UDP-N-acetyl-alpha-D-glucosamine hydro-lyase (configuration-retaining; UDP-2-acetamido-2,6-dideoxy-alpha-D-xylo-hex-4-ulose-forming) (4.2.1.135)	1896	519	5.6
propanoate:CoA ligase (AMP-forming) (6.2.1.17)	1921	614	5.4
[RNA] 5′-hydroxy-ribonucleotide-3′-[RNA fragment]-lyase (cyclicizing; [RNA fragment]-3′- nucleoside-2′,3′-cyclophosphate-forming and hydrolysing) (4.6.1.19)	1541	534	5.3
(2S,3R)-3-hydroxybutane-1,2,3-tricarboxylate pyruvate-lyase (succinate-forming) (4.1.3.30)	1474	464	5.3
(S)-2-hydroxyglutarate:quinone oxidoreductase (1.1.5.13)	2241	710	5.3
propanoyl-CoA:oxaloacetate C-propanoyltransferase (thioester-hydrolysing, 1-carboxyethyl-forming) (2.3.3.5)	1519	500	5.3
CMP-N-acetyl-beta-neuraminate:beta-D-galactosyl-(1->4)-N-acetyl-beta-D-glucosaminyl-R (2->3)-N-acetyl-alpha-neuraminyltransferase (configuration-inverting) (2.4.99.6)	1630	549	5.3
4-hydroxybutanoate:NAD+ oxidoreductase (1.1.1.61)	2370	680	5.3
ATP:(Kdo)-lipid IVA 3-deoxy-alpha-D-manno-oct-2-ulopyranose 4-phosphotransferase (2.7.1.166)	1535	555	5.2
n/a (3.1.25.1)	1573	556	5.1
(S)-3-amino-2-methylpropanoate:2-oxoglutarate aminotransferase (2.6.1.22)	2424	876	5.1
5-aminopentanoate:2-oxoglutarate aminotransferase (2.6.1.48)	2431	883	5.1

Usually I leave the enzymes to nerds, but in this I decided to do a little exploration. I recall that Lactobacillus Casei has been shown to reduce anxiety in studies. So I looked up what enzymes are in it. In the above list of too low, I found the first one above, 3.4.25.2 in it. It opens up a possible model for picking probiotics a priori to address anxiety.

There was only one lactobacillus with more than one of these enzymes, Lactobacillus yamanashiensis – not available as a retail probiotic. There were three bifidobacterium which had more than one of these enzymes:

• Bifidobacterium dentium (3)
• Bifidobacterium thermophilum (3)
• Bifidobacterium subtile (2)

Unfortunately, none are available as a retail probiotic. HOWEVER, there is one commercial probiotic that appears to have EIGHT (8) of these enzymes: Escherichia coli, available as Mutaflor (E.Coli Nisse 1917) or Symbioflor-2. So if you have anxiety, the theoretically best probiotic for you to take is an E.Coli one!!

A surprise!

I went to look at the a priori suggestions and was shocked/delighted that the only probiotic on the suggestion list was Symbioflor-2. Even with the complexities of 160 bacteria to adjust, somehow our enzymes calculations and this totally study based approach came to agreement! I repeat, I was shocked —

This selecting probiotics based on enzyme deficiency has an interesting scent and I will likely follow up on future posts.

A small number of possible ones are reported in the literature, but the quality of the results are suspect.

Tax Rank	Tax Name	Shift	Citation Link
genus	Enterococcus (NCBI:1350 )	High	📚 PubMed
genus	Klebsiella (NCBI:570 )	High	📚 PubMed
genus	Prevotella (NCBI:838 )	High	📚 PubMed
genus	Salmonella (NCBI:590 )	High	📚 PubMed
genus	Staphylococcus (NCBI:1279 )	High	📚 PubMed
genus	Streptococcus (NCBI:1301 )	High	📚 PubMed
phylum	Firmicutes (NCBI:1239 )	Low	📚 PubMed
species	Acinetobacter baumannii (NCBI:470 )	High	📚 PubMed
species	Bifidobacterium longum (NCBI:216816 )	Low	📚 PubMed
species	Enterococcus faecalis (NCBI:1351 )	High	📚 PubMed
species	Enterococcus faecalis (NCBI:1351 )	Low
species	Enterococcus faecium (NCBI:1352 )	High	📚 PubMed
species	Escherichia coli (NCBI:562 )	High	📚 PubMed
species	Escherichia coli (NCBI:562 )	High
species	Klebsiella pneumoniae (NCBI:573 )	High	📚 PubMed
species	Methanobrevibacter smithii (NCBI:2173 )	High	📚 PubMed
species	Methanobrevibacter smithii (NCBI:2173 )	High
species	Pseudomonas aeruginosa (NCBI:287 )	High	📚 PubMed

The usual method of testing is from breath tests. From KEGG.JP we can get a list of bacteria that produces the compounds detected in the breath — the number is huge.

This page/video was suggested reading/viewing by a reader.

Question: How do you search on Microbiome Prescription for H2 etc?

On My Own Sample

To me, the more information about the microbiome that you have, the better it is to identify issues and build a treatment course. Below, the numbers will speak for themselves.

Second Class Tests

I deem these as 2nd class for several reasons: the number of bacteria reported is low (compare to others), the lab method requires the bacteria to be culturable (hence many bacteria will never be reported), they do no provide a suitable CSV file for upload to Microbiome Prescription (allowing 2nd opinions for treatment), they mechanism of measurement is not compatible to Microbiome Prescription (and most recent microbiome studies on the US National Library of Medicine)

Lab Name	BacteriaReported
Bioscreen (cfu/gm)	17
Biovis Microbiome Plus (cfu/g)	40
DayTwo	76
Diagnostic Solution GI-Map (cfu/gm)	38
GanzImmun Diagnostic A6 (cfu/gm)	76
GanzImmun Diagnostics AG Befundbericht	25
Genova Gi Effects (cfu/g)	28
Genova Parasitology (cfu/g)	7
GI EcologiX (Invivo)	55
GI360 Stool (UK)	67
Gut Zoomer (vibrant-wellness)	152
InVitaLab (cfu/gm)	23
Kyber Kompakt (cfu/g)	11
Medivere: Darm Mikrobiom Stuhltest (16s limited)	16
Medivere: Darn Magen Diagnostik (16s Limited)	16
Medivere: Gesundsheitscheck Darm (16s Limited)	17
Metagenomics Stool (De Meirleir) (16s Limited)	53
Smart Gut (ubiome 16s – Limited Taxonomy)	23
Verisana (cfu/ml) aka (kbe/ml)	11
Viome (No objective measures)	29

First Class Tests

These are first class because: they allow an easy upload to Microbiome Prescription, use an appropriate measurement process, report based on bacteria DNA/RNA and not cultured count. The key words are 16s and shotgun analysis and not cultured.

There can be considerable cost differences between these labs (links to my first choices are below).

Lab/Processor	Low	Typical	High	Uploads
AmericanGut	73	156	213	18
BiomeSight	11	570	1305	1283
BiomeSightRdp	279	656	862	11
Nirvana / CosmosId	6	430	705	33
es-xenogene	1461	2880	5225	4
Medivere	530	721	934	7
Microba	53	122	153	16
SequentiaBiotech	166	313	460	36
Ombre Labs	185	667	2238	1066
uBiome (Out of Business)	6	249	589	813

Most Useful Lab?

Microbiome Prescription does some Fuzzy Logic Artificial Intelligence (FLAI) a.k.a. Dr. Artificial Intelligence. The accuracy is a function of the data available to him, that is the number of uploads from a specific lab. The newer lab BiomeSight is the winner here. This preference is made stronger because they will ship worldwide while #2, OmbreLabs (according to feedback) will only ship to the US. Microbiome Prescription mission is to support people worldwide .. so the handwriting on the statistics board is clear.

We have a growing number of special studies using BiomeSight data.

1. 🗺️ Allergic Rhinitis (Hay Fever) ( 165 candidate bacteria) with highest z-score of 9.5
2. 🗺️ Autism ( 192 candidate bacteria) with highest z-score of 8.3
3. 🗺️ Bloating ( 126 candidate bacteria) with highest z-score of 5.4
4. 🗺️ Brain Fog ( 93 candidate bacteria) with highest z-score of 5.2
5. 🗺️ Chronic Fatigue Syndrome (CFS/ME) ( 170 candidate bacteria) with highest z-score of 6.6
6. 🗺️ Cold Extremities ( 145 candidate bacteria) with highest z-score of 12.7
7. 🗺️ COVID19 (Long Hauler) ( 218 candidate bacteria) with highest z-score of 12.6
8. 🗺️ Easily irritated ( 248 candidate bacteria) with highest z-score of 9.9
9. 🗺️ High Anxiety ( 170 candidate bacteria) with highest z-score of 5.9
10. 🗺️ Histamine or Mast Cell issues ( 138 candidate bacteria) with highest z-score of 8.4
11. 🗺️ Inflammatory bowel disease ( 245 candidate bacteria) with highest z-score of 12.5
12. 🗺️ irritable bowel syndrome ( 150 candidate bacteria) with highest z-score of 6.7
13. 🗺️ ME/CFS with IBS ( 193 candidate bacteria) with highest z-score of 8.4
14. 🗺️ ME/CFS without IBS ( 251 candidate bacteria) with highest z-score of 7.3
15. 🗺️ Post-exertional malaise ( 175 candidate bacteria) with highest z-score of 6
16. 🗺️ Tinnitus (ringing in ear) ( 133 candidate bacteria) with highest z-score of 6.7
17. 🗺️ Unrefreshed sleep ( 165 candidate bacteria) with highest z-score of 6.9

Social Media Questions

For other tests not covered above, see 16s Providers

On Sun Genomics

On Gut Zoomer

“Show me the beef!” This appears to be pure marketing hype — or more specifically, a substance for microbiome analysis that is high in pathogens!

Gut Zoomer [Est. 2016] – 170 -200 species (per their advertising) must be ordered thru a physician

While Xerogene typically reports TEN TIMES as many!!! It is DEFINITELY NOT THE MOST COMPLETE, it is infact at the bottom of their first class competitors but the top of second class competitors.

What about Biohm? 

Based on their official View a sample report. they report on only 44 bacteria items.

We have IBS annotated in three different ways:

• Autonomic Manifestations: irritable bowel syndrome
• Official Diagnosis: Irritable Bowel Syndrome
• Condition: ME/CFS with IBS

I ran each possible combination of the above and Official Diagnosis: Irritable Bowel Syndrome gave the strongest results and this will be reported here. When other things are combined, it is common for associations to become diffused/weaker. This is part of  A new specialized selection of suggestions links.

Study Populations:

Symptom	Reference	Study
Official Diagnosis: Irritable Bowel Syndrome	1122	56

• Bacteria Detected with z-score > 2.6: found 148 items, highest value was 6.7
• Enzymes Detected with z-score > 2.6: found 218 items, highest value was 6.1
• Compound Detected with z-score > 2.6: found No items

Interesting Significant Bacteria

The results are very striking — low bifidobacterium across the board at the top! Almost all of the bacteria associated are low.

Bacteria	Reference Mean	Study	Z-Score
Bifidobacterium gallicum (species)	3741	587	6.7
Bifidobacterium kashiwanohense PV20-2 (strain)	325	62	6.6
Bifidobacterium catenulatum subsp. kashiwanohense (subspecies)	315	62	6.6
Thermosediminibacterales (order)	49	16	6.3
Lactiplantibacillus pentosus (species)	123	28	5
Escherichia (genus)	5983	1397	4.9
Bifidobacterium cuniculi (species)	81	30	4.7
Bifidobacterium angulatum (species)	184	32	4.7

Interesting Enzymes

As is often seen with various symptoms/conditions, the associations are due to insufficiency and not surplus.

Enzyme	Reference Mean	Study Mean	Z-Score
substrate,NADPH—hemoprotein reductase:oxygen oxidoreductase (RH-hydroxylating or -epoxidizing) (1.14.14.1)	51	19	6.1
NADPH:hemoprotein oxidoreductase (1.6.2.4)	51	19	6.1
propanoyl-CoA:oxaloacetate C-propanoyltransferase (thioester-hydrolysing, 1-carboxyethyl-forming) (2.3.3.5)	1480	381	6.1
(2S)-3-(4-hydroxyphenyl)-2-isocyanopropanoate,2-oxoglutarate:oxygen oxidoreductase (decarboxylating) (1.14.20.10)	131	26	6
(2S)-3-(4-hydroxyphenyl)-2-isocyanopropanoate,2-oxoglutarate:oxygen oxidoreductase (1.14.20.9)	131	26	6
(2S,3R)-3-hydroxybutane-1,2,3-tricarboxylate pyruvate-lyase (succinate-forming) (4.1.3.30)	1432	375	5.9
ATP:amicoumacin A 2-phosphotransferase (2.7.1.230)	41	18	5.7
(E)-4-(trimethylammonio)but-2-enoyl-CoA:L-carnitine CoA-transferase (2.8.3.21)	1354	388	5.7
gamma-butyrobetainyl-CoA:electron-transfer flavoprotein 2,3-oxidoreductase (1.3.8.13)	1365	390	5.7
L-carnitine:CoA ligase (AMP-forming) (6.2.1.48)	1339	443	5.4
hydrogen-sulfide:ferredoxin oxidoreductase (1.8.7.1)	9912	4683	5.3
NADPH:acceptor oxidoreductase (1.6.99.1)	5684	2284	5.3
(2S,3S)-2-hydroxybutane-1,2,3-tricarboxylate hydro-lyase [(Z)-but-2-ene-1,2,3-tricarboxylate-forming] (4.2.1.79)	1284	407	5.2
S-methyl-5′-thioadenosine:phosphate S-methyl-5-thio-alpha-D-ribosyl-transferase (2.4.2.28)	3638	1668	5.1
acyl-CoA,ferrocytochrome b5:oxygen oxidoreductase (6,7 cis-dehydrogenating) (1.14.19.3)	1062	284	5
butanoyl-CoA:acetoacetate CoA-transferase (2.8.3.9)	3892	2229	5
L-carnitinyl-CoA hydro-lyase [(E)-4-(trimethylammonio)but-2-enoyl-CoA-forming] (4.2.1.149)	1422	362	5

Bottom Line

Looking at the bacteria, the probiotic solution would be:

• Bifidobacterium (a wide variety)
• E. Coli Probiotic (Mutaflor or Symbioflor-2)

Lactobacillus are HOSTILE to E.Coli, so Lactobacillus probiotics should likely be avoided.

Looking at some of the recommendations, we see Human milk oligosaccharides (prebiotic, Holigos, Stachyose) which are known to greatly encourage bifidobacterium. We also see saccharomyces boulardii which is also known to increase Bifidobacterium [2020].

As always, using your own 16s samples would produce the best suggestions.

A reader messaged me the following concern

hi ken, what would you do if there’s a massive contradiction between most of the suggestions generated by the general consensus and the suggestions generated by your new “from special studies” biomesight algorithm

The first thing that I want to point out is the warning on that page.

The main issue you may be seeing is in the selection of bacteria. With the regular selection, you focus on extreme values, i.e. top 10%, outside of standard lab ranges, outside of reference ranges from Jason Hawrelak and others. The amount outside of the reference range is used to give a weight to each bacteria for the importance of shifting. Different algorithms are used with different approaches (we do not know what the ideal one is).

With the special studies, we up-ended the algorithm. We picked the bacteria based on a simple “if the amount is above or below the reference norm+/- twice the standard deviation of the mean for the reference population and then use the z-score as the weight (the statistical significance for this bacteria)”.

This change means that a bacteria that is at the 70%ile may be included in the selection (which is very unlikely with the the first methods), and this bacteria could have a very high weight (which is based on statistical significance and NOT the difference from a mile post). A Bacteria at the 99%ile will be totally ignored if it is not statistically significant for the condition.

Statistically, I prefer the special studies approach because we are using the statistical significance of the bacteria for the significance/weight for suggestions instead of the naïve assuming that being high or low is the cause.

Bottom Line

• We pick bacteria based on statistical significance for a specific condition and not whether they are high or low in general
• We give the bacteria a weight based on statistical significance for a specific condition and not the difference from a bound.

In theory, with the identical same bacteria and counts selected for two different conditions, you will get different suggestions because the weight assigned will be different since the weight is based on the statistical significance for the condition.

I well understand the confusion of some, the model being used is getting more advanced and handling more complexities.

This is a common symptom for both ME/CFS and Long COVID. This is reported often in samples, and thus being examined if it reaches our threshold for inclusion as defined in A new specialized selection of suggestions links. It does.

My default view is that this symptom is likely due to vesicular constriction/inflammation or “stick blood” (which contains many coagulation possibilities) resulting in low warming blood flow.

Study Populations:

Symptom	Reference	Study
Sleep: Unrefreshed Sleep	1087	78

• Bacteria Detected with z-score > 2.6: found 145 items, highest value was 12.6
• Enzymes Detected with z-score > 2.6: found 170 items, highest value was 5.3
• Compound Detected with z-score > 2.6: found No items

The highest z-scores above are more than most other symptoms. This indicates that the causes are more homogeneous bacteria shifts.

Interesting Significant Bacteria

All bacteria found significant had too low levels. The dominant genus is Prevotella with the P.Copri being usually 90% of the genus count in our reference, but only 14% of the genus count with Cold Extremities indicating that the shift is very species specific. Looking at Prevotella copri in isolation, we see the best documented ways of increasing it are: mediterranean diet, resveratrol (grape seed/polyphenols/red wine), berberine, navy bean,  Conjugated Linoleic Acid and linseed(flaxseed).

It is interesting to note that some of these are known to improve coagulation or reduces inflammation

• Therapeutic Potential of Resveratrol in COVID-19-Associated Hemostatic Disorders [2021]
• One-year consumption of a grape nutraceutical containing resveratrol improves the inflammatory and fibrinolytic status of patients in primary prevention of cardiovascular disease. [2012]
• Interaction between resveratrol and thrombin and its biological implication. [2011]
• Berberine alleviates myocardial ischemia-reperfusion injury by inhibiting inflammatory response and oxidative stress: the key function of miR-26b-5p-mediated PTGS2/MAPK signal transduction. [2022]
• Berberine as a natural modulator of inflammatory signaling pathways in the immune system: Focus on NF-κB, JAK/STAT, and MAPK signaling pathways. [2022]
• Dietary conjugated linoleic acid links reduced intestinal inflammation to amelioration of CNS autoimmunity. [2021]

Bacteria	Reference Mean	Study	Z-Score
Prevotella copri (species)	66900	3090	12.6
Prevotella (genus)	74698	21293	7.6
Prevotellaceae (family)	82172	36608	6.1
Bifidobacterium kashiwanohense PV20-2 (strain)	326	78	5.9
Bifidobacterium catenulatum subsp. kashiwanohense (subspecies)	317	78	5.8
Sporolactobacillaceae (family)	173	58	5.7
Thermosediminibacterales (order)	48	18	5.7
Sporolactobacillus (genus)	174	60	5.5
Sporolactobacillus putidus (species)	174	60	5.5
Bifidobacterium gallicum (species)	3784	783	5.5
Lactiplantibacillus pentosus (species)	123	22	5.3
Phocaeicola coprocola (species)	7584	663	5

Interesting Enzymes

Most (99%) enzymes found significant had too low levels.

Enzyme	Reference Mean	Study Mean	Z-Score
n/a (3.4.24.20)	40	16	5.3
2-acetylphloroglucinol C-acetyltransferase (2.3.1.272)	182	58	5.1
ATP:L-threonine O3-phosphotransferase (2.7.1.177)	2416	529	5

Bottom Line

In this study, one species shouts out as the cause by not being there Prevotella copri. I would really emphasis the items listed above as encouraging its growth.

It may be available “soon” as a probiotic, “The gut bacterium Prevotella copri (P. copri) has been shown to lower blood glucose levels in mice as well as in healthy humans, and is a promising candidate for a next generation probiotic aiming at prevention or treatment of obesity and type 2 diabetes.” [2021]

This article is an interesting read: The Strange Case of Prevotella copri: Dr. Jekyll or Mr. Hyde? Especially for those who think all bacteria can be classified as either good or bad.

In terms of retail probiotics suggested, the E.Coli probiotic, Symbioflor-2 is suggested.

The a priori suggestions, shown below, contains the items for P. Copri cited above.

Social Feedback

This is a common symptom for both ME/CFS and Long COVID. This is reported often in samples, and thus being examined if it reaches our threshold for inclusion as defined in A new specialized selection of suggestions links. It does, but the degree of association (z-scores) are lower than prior special studies despite having a larger study population.

Study Populations:

Symptom	Reference	Study
Sleep: Unrefreshed Sleep	1041	107

• Bacteria Detected with z-score > 2.6: found 139 items, highest value was 5.1
• Enzymes Detected with z-score > 2.6: found 208 items, highest value was 5.2
• Compound Detected with z-score > 2.6: found 1 items, highest value was 2.8 – effectively zero when false detection rate is considered.

The highest z-scores above are less than other symptoms despite larger sample size. This indicates that the causes are more diverse and thus less homogeneous bacteria shifts.

Interesting Significant Bacteria

All bacteria found significant had too low levels.

Low Veillonella is reported in some studies associated with sleep issues with it’s consumption of lactic acid(lactate) being cited as a possible factor:

• Gut microbiota in obstructive sleep apnea–hypopnea syndrome: disease-related dysbiosis and metabolic comorbidities [2019]
• “patients with sleep disturbances were characterized by lower bacterial diversities and distinct microbial composition in comparison to those without sleep disturbances. The relative abundances of Salivarius, Veillonella, Klebsiella, and Eubacterium were independent predictors of sleep disturbances in patients.” [2021] for the specific condition studied.

Bacteria	Reference Mean	Study	Z-Score
Veillonella (genus)	4063	2282	5.1
Actinobacillus (genus)	353	133	4.9
Bifidobacterium catenulatum subsp. kashiwanohense (subspecies)	321	100	4.8
Clostridium cellulovorans (species)	40	17	4.8
Bifidobacterium kashiwanohense PV20-2 (strain)	318	100	4.8
Actinobacillus porcinus (species)	184	73	4.6
Thiobacillus thiophilus (species)	85	26	4.6

Interesting Enzymes

All enzymes found significant had too low levels.

Enzyme	Reference Mean	Long COVID Mean	Z-Score
(1->4)-alpha-D-galacturonan lyase (4.2.2.2)	2453	1074	5.2
2-acetylphloroglucinol C-acetyltransferase (2.3.1.272)	185	61	4.6
phylloquinone:disulfide oxidoreductase (1.17.4.4)	28	15	4.6
[RNA] 5′-hydroxy-ribonucleotide-3′-[RNA fragment]-lyase (cyclicizing; [RNA fragment]-3′- nucleoside-2′,3′-cyclophosphate-forming and hydrolysing) (4.6.1.19)	1596	634	4.5
CMP-N-acetyl-beta-neuraminate:beta-D-galactosyl-(1->4)-N-acetyl-beta-D-glucosaminyl-R (2->3)-N-acetyl-alpha-neuraminyltransferase (configuration-inverting) (2.4.99.6)	1682	659	4.5
propane-1,3-diol:NAD+ 1-oxidoreductase (1.1.1.202)	811	106	4.5
ATP:(Kdo)-lipid IVA 3-deoxy-alpha-D-manno-oct-2-ulopyranose 4-phosphotransferase (2.7.1.166)	1591	645	4.5

Bottom Line

The key take away is that lactate/lactic acid levels appears to be a significant contributor. The reason that it is high is the lack of lactate consumers. Removal of lactate producing probiotics from supplements appears to be a logical first step (i.e. Lactobacillus), followed by taking Vitamin B1 shortly before bed time.

• Lactate as a Biomarker for Sleep [2012]

For suggestions on lowering it, see this old post of mine in the ME/CFS context.

• Reducing Lactic Acid Producing Bacteria – LABO [2019]

The issue may not be lactate by itself, but by the form of lactate (d-lactate –> bad)

Results and conclusion: Administration of L-lactate does not influence sleep-wake cycle of experimental animals. At the same time, its artificial optical analog D-lactate induces the significant (as compared to the control) decrease in wake (34.8% to 26.5%) and increase in slow wave sleep (57.4% to 69.2%). It has been suggested that D-lactate may be the antagonist of one or several L-lactate receptors.

[D-lactate as a novel somnogenic factor?] [2020]

“Thiamine (Vitamin B1) replenishment at intravenous doses of 100 mg every 12 h resolved lactic acidosis and improved the clinical condition in 3 patients.” [1997]

“Lactomin[300 mg Lactobacillus acidophilus, 300 mg Bifidobacterium longum] was discontinued, and she was treated with sodium bicarbonate and oral antibiotics. The probiotics the patient had taken were likely the cause of D-lactic acidosis ” [2010]

For almost a decade I have suspected that there was an interaction between the microbiome and Antiphospholipid syndrome (APS) also known as Hughes Syndrome (after the MD, see below). This is also called  “sticky blood syndrome” [HealthLine]. For some researchers, it is deemed to be a significant contributor to fatigue in Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) [1999 D. Berg] and likely also applies to Long COVID. My own singleton experience seems to confirm it for myself.

A reader asked about phospholipids on Facebook today, so I revisited available literature

This article by Graham R.V. Hughes, MD, FRCP (the discoverer) in 2016 is well worth reading.

For me, APS/Hughes syndrome is very much a neurological condition. Brain function does seem to be especially targeted—the more APS patients one sees, the wider and wider the neuropsychiatric ripples spread.

APS: What Rheumatologists Should Know about Hughes Syndrome • By Graham R.V. Hughes, MD, FRCP

Of course, running off the experience of just one, or even a few people, is not the best practice. Testimonials suck because of rose color glasses, fake testimonials, mainly positive responders report, and placebo effects. So what does the literature state. First there is some literature that are general discussions without the type of detail that I would love to see:

• Crosstalk Between the Gut Microbiome and Bioactive Lipids: Therapeutic Targets in Cognitive Frailty [2020]
• Phospholipid catabolism by gut microbiota and the risk of cardiovascular disease [2013]

Then we come to this article: Phosphatidylglycerols are induced by gut dysbiosis and inflammation, and favorably modulate adipose tissue remodeling in obesity [2019] which uses one of my favorite information source, the Kyoto Encyclopedia of Genes and Genomes. “We found that PGs were positively associated with microbiomes enriched with endotoxin-synthesis genes and associated with markers of inflammation.”

• ” In antiphospholipid syndrome, a thrombotic auto-immune disorder, autoreactive T cells and antibodies cross-react with auto-antigen mimicking peptides from gut commensals which appears to contribute to the pathophysiology. ” Gut microbiota and their metabolites in cardiovascular disease, 2021

Digging further we find:

 Bacteroides thetaiotaomicron, Actinomyces massiliensis, Pseudopropionibacterium propionicum, Corynebacterium amycolatum, Ruminococcus gnavus and Roseburia intestinalis[2021] lead to the formation of pathogenic T‑cell and autoantibody responses via the cross-reactivity with autoantigens (Ro60, dsDNA and ß₂ glycoprotein I). 

The role of the microbiome in lupus and antiphospholipid syndrome [2020]

M. pneumoniae and Streptococcus spp. infections, which are among the most prevalent bacterial infections in children and young adults, were linked to the occurrence of aPL. …. an anaerobic bacterium Fusobacterium necrophorum, although a variety of other bacteria such as streptococci, staphylococci, and enterococci may be also responsible…. a specific change in the gut microbial composition in APS patients. Particularly, a decrease of bacteria belonging to the genus Bilophila and overgrowth of bacteria of the Slackia genus were shown…  enrichment by Slackia spp. and by the lower abundance of butyrate-producing Butyricimonas 

Environmental Triggers of Autoreactive Responses: Induction of Antiphospholipid Antibody Formation [2019]

More discussion of mechanism is in The Role of the Gut Microbiota in the Pathogenesis of Antiphospholipid Syndrome [2015]

Bottom Line

APS only requires one of the bacteria above to trigger it. In terms of using Microbiome Prescription, I would look at Bilophila and Butyricimonas – if below 50%ile, hand pick it, then look at Slackia, if above 50%ile then hand pick it. Check the other bacteria cited above, and if any are over 75%ile, hand pick those. “It only takes one rotten apple to spoil the barrel” seems to apply here.

I have added APS to my PubMed reference list:

Personal Observations

I checked my samples from my last ME/CFS flare and found that Bacteroides thetaiotaomicron went from 73%ile on first sample after onset, to 96%ile on second sample, down to 79%ile, then 70%ile then 20%ile a few months later with recovery and returning to work. The key triggering bacteria will likely be different for each person but you at least have a candidate list to work from.

For items like antibiotics and probiotics, I have for a long time been a strong advocated for continuous rotation. The original source for this attitude was Cecil Jadin’s treatment protocol for occult rickettsia (which originated with the Pasteur Institute for Tropical Medicine). This was followed by reading studies finding that rotating or even just pulsing (2 weeks on/ 2 weeks off) was more effective in reducing bacteria than continuous. Probiotics often function via the natural antibiotics they produce (a lot of prescription antibiotics originated with bacteria); hence probiotic rotation became part of my preaching.

If you have microbiome related issues, my soapbox has been “your goal is make the stable dysfunctional microbiome, unstable“. Today I read a study on Nature that further clarifies what may be needed.

Together, these findings suggest that the human gut microbiome’s metabolic potential reflects dietary exposures over preceding days and changes within hours of exposure to a novel nutrient. The dynamics of this ecological memory also highlight the potential for intra-individual microbiome variation to affect the design and interpretation of interventions involving the gut microbiome.

Ecological memory of prior nutrient exposure in the human gut microbiome [2022]

If the goal is to make the microbiome unstable, then this gives some clear indication of strategy.

• Every two weeks change the dominant starch – for example, if pasta is a regular meal item then
  • Made from glucomannan—a starch found in the konjac yam/ Konjac Flour (Source)
  • Made from red lentils and quinoa (Source)
  • Made from white rice flour, organic amaranth flour (Source)
  • Made from chickpea flour, organic yellow lentil flour, organic red lentil flour, organic kale powder, organic spinach powder (source)
• Every two weeks change dominant proteins source
  • Fish
  • Pork
  • Lamb
  • Duck
  • Chicken
  • Turkey
• Change vegetables and fruit too…
• Change main spices used….

The key aspect is that every new addition results in a change of the microbiome. If you have microbiome issues, that is what you want to do. You do NOT want to take the same supplements, herbs, spices, vitamin or comfort food – continuously. You want to shake things up!

While the location of SIBO and the results of a stool test are a good physical distance apart, as part of an ongoing series of posts on special studies, I ran SIBO thru this morning — really expecting to find nothing, or perhaps a few overgrowths. To my surprise, the results was significant undergrowth with high statistical significance!

A formal post on the results is in the queue, but I thought that I should make the results available for those who wish to commit SIBO heresy!

SIBO reaches our threshold for inclusion as defined in A new specialized selection of suggestions links. A summary table of various studies has been added there which shows the statistical association is actually pretty strong.

Of course, you are saying “but this is down stream how can it impact upstream?!!?” To me, it is like saying, “I was waiting in a queue and everyone is healthy except the first person in the queue who had COVID, how could I get it?” Bacteria spread, including upstream.

 I should point out that these bacteria may not be the cause, rather they may be ‘the canaries in the coal mine’ of the microbiome.  A few items of special interest are:

Bacteria	Reference Mean	Study
Shuttleworthia (genus)	273	33
Prevotella stercorea (species)	6378	31
Coprococcus eutactus (species)	7987	1463
Streptococcaceae (family)	3567	1473
Veillonellaceae (family)	17295	10944
Tissierellales (order)	4162	1457
Peptoniphilaceae (family)	4158	1456

In terms of enzymes, the four items that were most significant were all low levels of:

• propanoate:CoA ligase (AMP-forming) (6.2.1.17)
• catechol:oxygen 2,3-oxidoreductase (ring-opening) (1.13.11.2)
• (2S,3R)-3-hydroxybutane-1,2,3-tricarboxylate pyruvate-lyase (succinate-forming) (4.1.3.30)
• propanoyl-CoA:oxaloacetate C-propanoyltransferase (thioester-hydrolysing, 1-carboxyethyl-forming) (2.3.3.5)

Bottom Line

SIBO by definition is OVERGROWTH, the stool samples and statistics says it is an UNDERGROWTH condition. Undergrowth means that substances do not get processed fully….

The sample size is low, more people with SIBO should transfer their data to Biomesight, then back to Microbiome Prescription, and then annotate it with SIBO. See Video on Transferring Data from Ombre/Thryve to Biomesight and then to Microbiome Prescription. I will wait until the sample size is larger before doing a formal post.

In the mean while, people with SIBO can try the new algorithm on the changing microbiome page. Feel free to add comments here on experience after trying suggestions for a month.

REMEMBER: The usual test is finding various compounds on the breath. Are the compounds from producing too much(overgrowth) or the compounds are not being consumed enough (undergrowth). Both scenarios produce a positive result. It feels like someone jumped to a conclusion and this arbitrary decision stuck in the name.