← Reviews
Aug 23, 2026

Heat-killed Mycobacterium aurum reduces hyperglycemia in streptozotocin-induced diabetic mice

Both prophylactic oral and therapeutic intradermal administration of heat-killed M. aurum attenuated hyperglycemia in an STZ mouse model of diabetes, with no effect in non-diabetic animals.

Evidence levelDNarrative / animal / in vitro / mechanistic
Study typein_vitro
Sample
Effect directionMechanistic only
CertaintyVery low
Clinical applicabilityVery low
Overinterpretation risk1/5 · Low
PICO
Population
Intervention
Comparator
Outcome

What the study showed

Oral prophylactic treatment reduced blood glucose by 39–60% at weeks 5–8 post-STZ, with a trend toward improved serum C-peptide but no significant changes in glucose transport or antioxidant proteins. Intradermal therapeutic dosing lowered blood glucose ~30–40% and modestly enhanced insulin secretion, normalizing hepatic UCP2, hepatic α-LDH, and skeletal muscle UCP3 protein levels.

How it was done

Male BALB/c mice received STZ (150 mg/kg) to induce diabetes; heat-killed M. aurum (1 mg) was given orally (3 prophylactic doses pre-STZ) or intradermally (6 weekly therapeutic doses post-STZ); glycemic parameters, C-peptide/insulin by ELISA, and tissue protein expression by Western blot were assessed.

Risk of bias

This is entirely an animal study; the abstract does not report per-group sample sizes or full variance data. The STZ model incompletely recapitulates human diabetes pathophysiology, and mechanistic pathways remain only partially described.

Interpretation limit

What this study does NOT prove

The study does not demonstrate efficacy, safety, or mechanism in humans, nor does it establish that M. aurum preserves beta-cell mass or reverses established diabetes.

In clinical practice

No clinical recommendations can be derived from this study. Preclinical murine findings require replication in controlled human trials before any therapeutic consideration.

Limitations

This is entirely an animal study; the abstract does not report per-group sample sizes or full variance data. The STZ model incompletely recapitulates human diabetes pathophysiology, and mechanistic pathways remain only partially described.

Technical appendix

Version history

  • 1.0 · 2026-08-23 — Auto-generated under Evidence Standard v1.0
Source: DOI 10.3390/nu18111652 · 2026

Paid access: structured summary from public metadata; consult the original study at the source.

Microbiota Weekly

The week in microbiota evidence, in your language. Structured summaries, traceable to the source.