Reference
Axelsson AS, Tubbs E, Mecham B, et al. Sulforaphane reduces hepatic glucose production and improves glucose control in patients with type 2 diabetes. Sci Transl Med. 2017;9(394):eaah4477.
Objective
To find new medicines that may help address an important pathologic mechanism of type 2 diabetes mellitus—the liver’s ability to produce glucose by way of gluconeogenesis.
Study Design
Randomized, double-blind, placebo-controlled study
Participants
Investigators recruited 103 Scandinavian patients with type 2 diabetes mellitus (T2D), diagnosed within 10 years before study start. Participants had either well-controlled or poorly controlled T2D; poorly controlled T2D was defined as a glycated hemoglobin (HbA1c) level above 50 mmol/mol. For reference, 48 mmol/mol or greater is equivalent to an HbA1c of 6.5%; 42 to 47 mmol/mol is equivalent to an HbA1c of 6.0% to 6.4% (prediabetes); and HbA1c less than 42 mmol/mol represents normal blood sugar. Ninety-seven patients completed the study; 60 had well-controlled and 37 had poorly controlled T2D. Seventeen of the patients with poorly controlled disease were obese. All but 3 participants (who were well-controlled) were taking metformin.
Participants with poorly controlled disease were divided into 2 groups—nonobese and obese (BMI>30 kg/m2)—because hepatic glucose production is more severely affected in obese patients.
Study Parameters Assessed
Blood levels of fasting glucose (a representation of hepatic glucose production) and hemoglobin A1c were assessed at the beginning and end of the study. After initial blood testing (fasting glucose, HbA1c, and an oral glucose tolerance test) participants received 1 dose per day of oral broccoli sprout extract (BSE) or placebo. The BSE contained 150 mmol sulforaphane (SFN) per dose. Blood tests were repeated at the end of the 12-week period.
Primary Outcome Measures
Change in fasting glucose and hemoglobin A1c levels from baseline at 12 weeks.
Key Findings
Sulforaphane given as concentrated BSE did improve fasting blood glucose levels and decrease HbA1c levels in obese patients with T2D. The magnitude of reduction of HbA1c was greater for participants with higher HbA1c levels at the beginning of the study (−0.2 mmol/mol per 1 mmol/mol higher HbA1c at start; P=0.004). The association between baseline HbA1c levels and the magnitude of change was not significant in the placebo group (P=0.5). There was also an association between BMI and change in HbA1c in the BSE-treatment group, with the magnitude of reduction greater for obese participants (−0.4 mmol/mol per 1 kg/m2 or higher BMI; P=0.015). The association between BMI and change in HbA1c was not significant among obese participants in the placebo group.
These findings are noteworthy given that more than 400 million people worldwide are afflicted with diabetes, and an even larger number have prediabetes.
There were no safety concerns using SFN and it was well-tolerated.
Practice Implications
In this study the authors report on the benefits of SFN as BSE in regulating blood glucose levels in diabetics.
Preclinical experiments
The clinical study described here was preceded by extensive research to identify a novel drug to treat diabetes. The investigators generated a disease signature based on diabetes-associated gene networks in liver tissue, the site of overproduction of glucose in T2D, then matched them with drug signatures from a large database. After searching through the extensive database, they found that SFN had the most overlap with the gene signatures relevant to diabetes associated with hepatic glucose production.
They first tested the effect of SFN on glucose production using a rat hepatoma cell line. Incubation of these cells with SFN showed a dose-dependent decrease in blood glucose production. This mechanism may be in part explained by nuclear translocation of nuclear factor erythroid 2–related factor 2(NRF2) and the associated downregulation of key enzymes for gluconeogenesis.
They then went on to test SFN on different animal models in vivo. They were looking at glucose intolerance in rats fed high-fat and high-fructose diets. There was benefit in both diets, and in fact the magnitude of benefit was quite similar to the use of metformin. Additionally, the rats given SFN had decreased hepatic glucose production, again similar in benefit to metformin. Further, there was benefit in glucose tolerance for mice that had diet-induced diabetes.
Clinical trial
After both in vitro and in vivo studies supported SFN’s potential for treating diabetes, the investigators proceeded to test its effects on glucose control in humans with T2D, the clinical trial described in this review. The results showed that SFN in the form of concentrated BSE did improve fasting blood glucose levels and decrease HbA1c levels in obese patients with T2D.
These findings are noteworthy given that more than 400 million people worldwide have diabetes, and an even larger number have prediabetes.1 Poorly controlled blood sugar also increases risk of cancer, especially breast cancer.2,3
The SFN in this study was given as dried powder of an aqueous extract of broccoli sprouts. The choice of BSE was informed by other clinical studies that used BSE as a source of SFN, including studies on cancer,4 chronic obstructive pulmonary disease,5 inflammatory disorders, and autism. In this study SFN reduced HbA1c levels in diabetics at a daily dose of BSE that contained 150 mmol SFN. A number of human studies have shown that the dose of SFN should be 40 to 60 mg for its many health benefits.6
Clinically, we would likely have improved outcomes with a whole plant, such as sprouted broccoli seeds, because the chewing process and microbial enzymes in our mouths (myrosinase) help activate SFN. This can be achieved by consuming roughly 100 g of broccoli sprouts.
