Kynurenine pathway metabolites: relevant to vitamin B-6 deficiency and beyond

See corresponding article on page 934.

Pyridoxal 5′ phosphate (PLP)⁴, the active form of vitamin B-6, is involved in a wide variety of physiologic processes including gluconeogenesis and the synthesis of sphingolipids and neurotransmitters. It also functions as a cofactor for many enzymes required for amino acid metabolism. Vitamin B-6 can be acquired from both animal and plant foods. It is also available through fortified cereals and supplements. Clinically, signs of isolated B-6 deficiency are rare. However, subclinical deficiency can precipitate biochemical changes that become more obvious as deficiency progresses.

An important biochemical change associated with vitamin B-6 deficiency is that of hyperhomocysteinemia (1). It was the recognition of hyperhomocysteinemia as a cardiovascular event risk factor that led to several trials aimed at assessing the clinical impact of supplementation with vitamin B-6, as well as folate and vitamin B-12. One such trial, conducted in Norway, was designed to examine these vitamins for secondary prevention of cardiovascular events and reduction in mortality (2). Although the trial ultimately showed no benefit from the supplements, the rich data set collected from >3000 enrolled adults has been useful in addressing additional questions about vitamin status.

In this issue of the Journal, Ulvik et al (3) use these data to identify a new serum marker for examining functional vitamin B-6 status. To achieve this, they measured several metabolites in the tryptophan-to-kynurenine pathway. As shown in their Figure 1, tryptophan is metabolized to kynurenine via either indoleamine 2,3 dioxygenase (IDO1 or IDO2) or tryptophan 2,3 dioxygenase (TDO). Other than 3-hydroxykynurenine (HK), all subsequent metabolites require a vitamin B-6 (PLP)–dependent enzyme for generation. They show that ratios of HK to metabolites downstream of PLP-dependent enzymes [xanthuranic acid (XA), 3-hydroxylanthranilic acid (HAA), and kynurenic acid] correlate with PLP concentrations better than HK alone. The relation was strongest at the lowest PLP concentrations, and importantly, the ratios normalized after vitamin B-6 supplementation was instituted.

It is intriguing to think that plasma HK:XA and HK:HAA ratios reflect vitamin B-6 function in real-time physiology. Validation of this assay in a more diverse population would surely be needed before it could be adopted; even then, it would likely be most useful as a complementary test to the direct measurement of PLP concentrations. Still, there is much to be learned from this highly detailed and data-rich study. Along with the authors’ previous work (4), this study provides insight as to the normal ranges of the entire panel of the kynurenine pathway metabolites and their ratios in a very large, albeit homogeneous, population. Moreover, important baseline population characteristics are provided, including inflammatory markers and smoking status.

These data are particularly relevant now as we begin to appreciate and explore the variety of important physiologic and pathophysiologic functions mediated by kynurenine metabolites. For example, the immunomodulatory capacities of IDO are in part driven by kynurenines directly suppressing inflammatory T cell responses (5). This same mechanism likely supports tumoral immune escape in IDO- and TDO-expressing cancers. Both kynurenine and quinolinic acid (another kynurenine metabolite downstream of PLP-dependent enzymes) directly support tumor cell proliferation by activating β-catenin signaling in colon tumors (6). This pathway also promotes tumor growth through activation of the human aryl hydrocarbon receptor in other tumor types (7). Kynurenine itself also has vasodilatory properties and perhaps acts as a compensatory mechanism in atherosclerotic coronary disease (8). Finally, several of the kynurenine metabolites have neuromodulator effects that are predicted to play important roles in both psychiatric and neurodegenerative disorders (9). It is worth noting that the signaling events for most of these processes are likely attributable to high local kynurenine metabolite concentrations, rather than to circulating concentrations as discussed in the current article.

The plasma kynurenine-to-tryptophan ratio (KTR) has also been widely examined as a biomarker of immune activation. This ratio reflects activity of IDO1, which is highly upregulated in response to interferon-γ and expressed in several tissue types as well as in macrophages and dendritic cells. Measurement of the IDO1 substrate and its downstream product limits potential bias from variations in dietary tryptophan. For example, the KTR corresponds well with Crohn disease activity and positively correlates with both the erythrocyte sedimentation rate and C-reactive protein (10). Unbiased metabolomic studies have confirmed that intestinal inflammation leads to changes in several metabolites of the kynurenine pathway as detected in urine and serum (11). In the current study, the authors accounted for the KTR, C-reactive protein, and neopterin in their analysis. Patients with higher inflammatory markers tended to have higher HK concentrations. In turn, the HK-based ratios more strongly correlated to PLP concentrations in these patients.

Future investigations will clarify whether these kynurenine pathway metabolite ratios will be useful for measuring functional vitamin B-6 status in patients with subclinical deficiency. At least one recent study provides support for their conclusions with the use of a mathematical modeling approach (12). Future clinical studies seeking to address this same question will also hopefully include more heterogeneous populations and include patients with autoimmune/inflammatory conditions. Until that time, however, Ulvik et al have certainly given us new insight into the complex world of using metabolite ratios to understand the physiologic underpinnings of the kynurenine pathway.