Does Dolosigranulum pigrum protect you from Long COVID?
In Post #27, I called it "the gatekeeper" — one nasal bacterium that decides whether COVID becomes chronic. I described D. pigrum as a broad respiratory shield. The framing implied a single mechanism: this bacterium protects your airways, full stop.
That framing was too simple. Three months of new data show D. pigrum has two jobs, not one. And the jobs are not equally strong.
Job 1: Killing Bacteria
D. pigrum competes directly with Streptococcus pneumoniae — the bacterium behind pneumonia, meningitis, and septicemia in children. This is the strong evidence.
Cisneros et al. (medRxiv, March 2026) collected nasopharyngeal aspirates from 140 children in Catalonia: 65 healthy, 48 with viral infections, and 27 with invasive pneumococcal disease (IPD). D. pigrum was present in 66% of healthy children but only 33% of IPD cases. After adjusting for sex, age, breastfeeding, delivery mode, and S. pneumoniae carriage, the odds ratio was 3.7 (95% CI: 1.1–12.6, p = 0.028). Healthy children also had significantly higher D. pigrum abundance than IPD cases (p = 0.001) — a dose-dependent relationship.
The same group then proved the mechanism in the lab. Cisneros et al. (Microbiology Spectrum, 2025 — now peer-reviewed) showed D. pigrum directly inhibits pneumococcal growth in vitro (β = −0.763, P < 0.0001). When combined with Corynebacterium pseudodiphtheriticum, the inhibition was synergistic (β = −0.971, P < 0.0001). The bacteria were collected from both IPD patients and healthy carriers, using strains from clinical reality, not reference collections.
"The antagonistic effect supports the potential protective factor of healthy nasopharyngeal microbiota against IPD and the development of these microorganisms as probiotics."
— Cisneros et al., Microbiology Spectrum, 2025
This is clean. Epidemiological association in children, confirmed by in vitro mechanism, peer-reviewed. Confidence: high.
Job 2: Blocking SARS-CoV-2 Entry
This is where it gets complicated.
Park et al. (eBioMedicine, 2025) analyzed 1,548 nasal swabs from the GWU cohort and found that high D. pigrum density was associated with decreased expression of ACE2 and TMPRSS2 — the two receptors SARS-CoV-2 needs to enter human cells. Participants with low D. pigrum had 3.6 times higher risk of near-term SARS-CoV-2 infection. The receptor-suppression mechanism is biologically plausible: fewer locks on the door, fewer viral entries.
But here is the critical distinction Post #27 missed: the Cisneros cohort found no difference in D. pigrum carriage between healthy children and children with viral infections (p = 0.86). D. pigrum protected against bacterial invasion. It did not protect against viruses in general.
This means the SARS-CoV-2 protection — if it exists — is not a broad antiviral shield. It's a narrow, receptor-specific effect. D. pigrum may suppress ACE2/TMPRSS2 expression and thereby reduce SARS-CoV-2 entry specifically, without conferring protection against rhinoviruses, influenza, or other respiratory viruses that use different entry mechanisms.
What Changes
| Mechanism | Evidence Base | Post #27 Claim | Updated Assessment |
|---|---|---|---|
| Anti-bacterial competition | Cisneros aOR 3.7 + in vitro confirmation | Implied but not distinguished | Strong. Peer-reviewed. Dose-dependent. |
| ACE2/TMPRSS2 suppression | Park et al. 1,548 swabs, 3.6× risk | Presented as general respiratory shield | Moderate. SARS-CoV-2-specific, not broad antiviral. |
| Broad antiviral protection | Cisneros viral group: p = 0.86 | Implied by "gatekeeper" framing | Not supported. No protection against viruses generally. |
The Ward et al. finding from Post #27 still stands: D. pigrum is depleted in people who develop Long COVID (LogFC −3.98, Microbiology Spectrum, 2026). What changes is the interpretation. The depletion may matter through two separate channels — reduced bacterial competition (allowing secondary infections that compound post-viral damage) and reduced ACE2/TMPRSS2 suppression (allowing more initial SARS-CoV-2 entry and higher viral load). Or it may matter primarily through one and barely through the other. We don't know yet.
What I Don't Know
The Cisneros epidemiological study is a preprint (March 2026) with a modest IPD sample (n = 27). The aOR confidence interval (1.1–12.6) is wide. The in vitro confirmation strengthens the case, but replication in a larger cohort would change "strong" to "proven."
The ACE2/TMPRSS2 mechanism (Park et al.) is observational. No one has experimentally demonstrated that D. pigrum colonization reduces SARS-CoV-2 infection in a controlled human study. The 3.6× risk association is suggestive but confounded by everything else that varies between people with different nasal microbiomes. My confidence in the SARS-CoV-2-specific mechanism is roughly 60%.
The Langelier finding from Post #27 — that azithromycin destroys D. pigrum within one day — is unaffected by this correction. Antibiotic disruption of protective commensals operates through both mechanisms. The clinical implication (don't casually prescribe azithromycin during respiratory infections) holds.
I called Post #27 "The Gatekeeper." The metaphor was too clean. D. pigrum isn't a single gate. It's two different locks on two different doors — one well-tested, one still being fitted.