Anders Corbett: On “Crafting” Your Gut Microbiome, Probiotics, and Hormone Optimization | TUH #233

Corbett’s service, Craft Microbiome, is built on the idea that personalized probiotics outperform generic ones. He mails a collection kit; after receiving a sample, he plates the bacteria under conditions matching their original environment (aerobic, anaerobic, specific nutrients). Genomic sequencing confirms strain identity. The client receives a menu of bacteria associated with published functions (bone density, serotonin/dopamine, androgens, oxygen utilization, etc.) and chooses targets. Corbett then grows those strains from the client’s own sample, freeze‑dries them, and returns them in capsules. The whole process turns gut‑microbiome modulation into a bespoke service, akin to ordering a tailored supplement stack from your own biological material. He emphasizes starting small with one or two benign strains, especially for people with IBD, and building from there based on subjective feedback—‘I’m sleeping better, I’m feeling better.’

Corbett discovered Veillonella’s potential by sampling World’s Strongest Man competitors and UFC fighters before and after extreme exertion. He saw that Veillonella levels remained high for weeks after competition, indicating the bacteria were feeding on residual lactate. He then isolates that same Veillonella from a client’s sample and returns it in high numbers. The rationale is straightforward: the more of these lactate‑eating bacteria you have, the faster you clear the byproduct that causes muscle burn and performance decline. He has applied this for NBA players, where travel and back‑to‑back games impair recovery; by keeping Veillonella populations high during the season, they can theoretically bounce back faster.

Corbett’s friend suffered years of Lyme disease, a PICC line for antibiotics, and eventual leg amputation from misdiagnosis. His gut was virtually sterile. Corbett gave him ‘craft strength’ — a targeted probiotic with L. plantarum and L. reuteri. Within a few months, the man’s testosterone climbed from low to 800‑900 ng/dL and remained there for years, accompanied by the ability to run a flat mile and live a normal life. Corbett sees this as unblocking growth‑hormone pathways that had been shut down by dysbiosis. He notes that similar effects appear in bloodwork for other clients — B vitamins rise, testosterone jumps 300 points — simply by repopulating the gut with the right strains.

Corbett draws on studies showing that bacteria acquired during birth seem to hold hierarchical authority — they can ‘tell other bacteria what to do.’ B. infantis, by most evidence, is that first colonizer. He uses it as his go‑to for restoring order after antibiotics, before layering on other functional bacteria. In patients with severe gut inflammation (diverticulitis, ulcerative colitis, Crohn’s), he starts with the most benign Bifidobacterium from their own stool, aiming solely to cut inflammation before attempting any performance‑oriented strains. The feedback is typically rapid — less bloating, better sleep, and a calmer gut within days. He credits the supercentenarian data (118‑year‑old with Bifidobacterium levels like a 35‑year‑old) as indirect support for its central role.

After trying an oral rinse with the bacteria (signaling approach) that failed, Corbett searched for a bacterium that directly consumes glucose. He wore a CGM and ate the identical Wendy’s meal every day. Initially, his glucose rose to 120–140 mg/dL. When he took the bacterium 90 minutes before the meal, the spike dropped to 115–120 mg/dL — a meaningful reduction for a small intervention. After seeing the effect on himself, he provided the same formulation to a client who had morning glucose readings around 200 mg/dL, and the client’s spikes also diminished. He sees this as a proof‑of‑concept that you can use your own bacteria to manage acute glycemic responses, and it fits into his broader philosophy of ‘overloading’ beneficial bugs to achieve a desired metabolic outcome.

Corbett cites large‑scale studies showing that specific bacteria decline with age and that the gut becomes less oxygen‑tolerant, a change correlated with the number of pharmaceuticals a person takes. He also highlights the 118‑year‑old supercentenarian who had Bifidobacterium levels typical of a 35‑year‑old. Storing a young sample creates a ‘biological backup’ — as the technology improves, you could go back to that original sample and isolate more strains. This aligns with his view that early gut dysbiosis is the first domino in chronic disease. By banking your youth microbiome, you preserve the option to re‑introduce those bacteria years later, potentially restoring some of the resilience you had when younger.

Corbett doesn’t claim a cure but describes his approach: he dives into research to find which metabolites are over‑ or under‑produced in the condition, then aims to either consume the harmful ones or overproduce the beneficial ones using the client’s own bacteria. He acknowledges it’s harder to eliminate ‘bad’ bacteria than to ‘outcompete’ them by flooding the system with the beneficial strains. For eczema/psoriasis, he would try to find the specific missing anti‑inflammatory bacteria from the client’s sample and return them in high numbers. This is a personalized, literature‑guided version of the probiotic approach, and it’s why he emphasises storing the original sample — as research advances, more strains can be isolated from the banked sample.

Corbett’s process starts with a stool or saliva sample the size of a penny. He plates the bacteria under the specific oxygen and nutrient conditions each strain needs, then genomic‑sequences colonies to identify strain and substrain. Once verified, he grows large quantities in a bioreactor, freeze‑dries the ‘goo’, and ships it back in capsules—while banking a reserve. The service is presented like a menu: ‘Here are the bacteria and their documented effects—what goals do you have?’ Customers pick outcomes (focus, testosterone, recovery, lower inflammation), and he isolates the matching strains from their own sample. He emphasizes the low‑risk nature: ‘It’s your own bacteria.’ He contrasts this with generic probiotics that can cause problems for some (e.g., people with allergies often react poorly to Lactobacillus). The approach also allows ‘overloading’ a chosen bacterium to influence pathways.

Corbett discovered this by sampling before and after stressful events (UFC fights, World’s Strongest Man, Olympic events). In World’s Strongest Man athletes, he measured Veillonella levels four times before, during, and after competition. He could distinguish athletes who had competed two weeks earlier because they still had elevated Veillonella—the bacteria were still consuming leftover lactate. He correlated his findings with papers describing Veillonella’s lactate‑consuming ability. He then isolates Veillonella from a client’s own microbiome and returns it as a high‑dose targeted probiotic. For an elite NBA player like LeBron James, Corbett’s focus was recovery; the hypothesis is that maintaining a high Veillonella population helps clear lactate from intense basketball games and travel fatigue. He phrases it as a simple engineering principle: if you have a performance‑limiting byproduct and you introduce a biological ‘factory’ that consumes it, performance improves.

The patient had been on a PICC line with antibiotics for years, leaving virtually no gut bacteria. His blood tests showed low testosterone. Corbett gave him ‘craft strength,’ a probiotic containing Lactobacillus plantarum (plus a substrate from South Korea) and a substrain of Lactobacillus reuteri. The L. reuteri substrain is documented to lower interleukin‑17 alpha, which in turn allows Leydig cells and testes to enlarge and produce the body’s own testosterone. After repopulation, the man’s testosterone settled at 800‑900 ng/dL and remained so for four to five years. Beyond testosterone, the patient recovered athletic ability—running a flat mile with Corbett—and returned to a normal life. Corbett notes that such results show up on quarterly blood work: B vitamins rise, testosterone jumps 300 points. He frames it as removing a ‘block’ that prevents the body’s own growth‑hormone pathways from functioning.

Corbett references papers showing that bacteria acquired during birth (from the birth canal or breast milk) seem to have ‘seniority’—they can organize the other bacteria. Bifidobacterium infantis, by most indications, is that first colonizer. He says after a massive antibiotic course (e.g., for dental surgery), the first step is to give the patient their own isolated B. infantis, which ‘settles down a lot of people’s guts’ within days. He uses it as a baseline before building up other strains for specific goals. He also ties this to the supercentenarian study where a 118‑year‑old woman had high levels of Bifidobacterium relative to a 35‑year‑old, supporting its central role in longevity.

To solve a client’s high morning glucose (~200 mg/dL), Corbett first tried oral sequencing to see which bacteria were missing. He found peer‑reviewed papers linking specific bacteria absence to pre‑diabetes or diabetes. He then isolated those missing bacteria from himself (as a safe initial test), then made a mouthwash to signal the brain before meals—it didn’t work. He pivoted to finding a bacterium that directly consumes glucose. He wore a CGM and ate the same Wendy’s meal each day. Pre‑supplement, his glucose spiked to 120–140 mg/dL; after taking the bacterium 90 minutes before the meal, spikes lowered to 115–120. He handed the formulation to the client, who then saw similarly reduced spikes. This is an example of his iterative ‘try, measure, refine’ approach using his own body first.

In discussing where the field is heading, Corbett mentions a ‘crazy paper’ where bacteria infected by a phage made mice smarter and flies have longer memory. The phage targets only a specific bacterium (e.g., Lactobacillus), but the interaction alters the host’s cognition. No one knows why yet. He sees this as an unexplored lever—potentially using engineered phages to tweak bacterial function to boost memory or processing speed—a futuristic extension of the ‘crafting’ concept.