Our group investigates the mechanism of action of Vitamin A in immune and other cells. One will ask: why? After hundred years of vitamin A research, 500,000+ publications, do we not already know how vitamin A works? Is the retinoic acid paradigm of gene trans-activation not adequate to explain vitamin A action?
Our issues are not with retinoic acid. We do not challenge the magnificent edifice built around this hormone. We believe, as do many colleagues, that vitamin A is indispensable for the physiology of most cell types, and itself plays an essential role in common signaling processes. Here is the unresolved paradox: Most cells in the body are awash in vitamin A, transported to tissues via the blood stream in the form of the retinol binding protein/transthyretin/vitamin A complex, and taken up by virtually all cells by the ubiquitous (intra)cellular retinol binding protein. Most cells store vitamin A as retinyl-esters, and metabolize vitamin A to a variety of products, including 14-hydroxy-retro-retinol (14-HRR), dihydroxyretinol (DHR),. The occasional cells will also oxidize retinol to retinoic acid, or dehydrate retinol to anhydroretinol (AR). Why would cells expend considerable capital in the enzyme machinery for uptake, esterification, storage, retrieval through ester hydrolysis, metabolic conversion to hydroxylated forms, if not for a purpose? Why lay in disproportionately high supplies of retinol (measuring in micromoles) if the sole purpose were conversion to retinoic acid needed at sub-nanomolar concentration, even ignoring that many cells do not produce retinoic acid at all? In answer we claim that a fourth arm of vitamin A physiology exists (besides RA-mediated transcriptional regulation, vision, and storage) that works through retinol itself, or utilizes the hydroxylated forms, 14-HRR and DHR). Cell signaling emerged as an area of major import.
Another fact at odds with sole usage of vitamin A as precursor for biosynthesis of retinoic acid (and of retinaldehyde in the eye) is that retinoids arose in evolution long before retinoic acid receptors. Insects possess most enzymes for uptake, storage and metabolism to the products described above, but lack retinoic acid receptors and transcriptional regulation by these receptors. These did not make an appearance until the vertebrate/invertebrate interface.
A decade ago we raised convincing evidence for a distinct role of vitamin A in signal transduction by identifying high affinity binding sites on serine/threonine kinases. All members of the PKC and cRaf families bound retinol at discrete sites localized within the conserved zinc-finger structures of the N-terminal regulatory domains. By coincidence, these same structures harbor also the binding sites of respective second messengers, diacyl-glycerol for PKC and GTPras for cRaf families. Because the extracellular retinol concentration exceeds the Ka of the binding sites the kinases are likely constitutively complexed with vitamin A. Yet no impact on function was initially noted, as both the phorbol pathway or the Ras pathway proved independent of retinoids. However, non-classical activation of PKC and cRaf mediated by a redox mechanism was significantly enhanced by retinol, leading to the hypothesis that retinol may play a role as facilitator of oxidation of serine/threonine kinases.
Redox activation of PKC was common to all isoforms, raising the question why retinol, the putative redox catalyst, was bound at the zinc-finger domains, that is: at the very activation centers of these kinases. PKC zinc-fingers are composite structures formed by six cysteines and two histidines and coordinated by two zinc ions. Crucial to the integrity of these domains are the thiolate anions of cysteines. We postulate that during redox activation thiolate anions are oxidized to disulfide, leading by necessity to the disruption of zinc-coordination centers, and setting the stage for conformational unfolding of the protein. Such hinge-like action of zinc-finger structures has been invoked for the redox-regulated bacterial chaperone, Hsp33.
The agonistic action of retinol as putative redox catalyst was mimicked by 14-hydroxy-retro-retinol. In contrast, anhydro-retinol was not only ineffective, but behaved as an antagonist. Mutual reversible inhibition experiments implied competition for a common receptor, and this concept was verified at the molecular level with recombinant PKC zinc-finger proteins. Meanwhile cultures of several cell types, notably lymphocytes, were found to require vitamin A for growth and survival, whereas depletion of vitamin A from the medium led to cell death, albeit not before intracellular retinyl-ester stores were depleted as well. Using the antagonist, anhydroretinol, induction of cell death was accelerated, but in keeping with the idea of competition for a shared receptor, retinol, as well as 14-hydroxy-retro-retinol, rescued AR-compromised cells.
Why would vitamin A depletion, either nutritionally induced, or imposed by AR, lead to cell death? What are the intracellular targets? We found that mitochondria stood out as the likely target where AR inflicted prominent and progressive damage. Within minutes after exposure to AR the mitochondrial membrane potential dissipated, leading to opening of permeability transition pores, release of pro-apoptotic molecules such as cytochrome C and activation of the caspase 8 cascade. Worse, in acutely sensitive cells where the proton-motive force was rapidly lost and ATP production severely compromised, death occurred by necrosis. In keeping with our paradigm that retinol and AR compete for the same binding sites, addition of retinol in excess over AR afforded protection. These results are best explained by the existence of regulatory signal networks that control mitochondrial function, with one or more components dependent on retinol as co-factor. Indeed, such signal networks are not only known, but include PKC at a crucial juncture and, moreover, are known to be redox regulated. Of added interest is the finding that mitochondria release oxygen radicals that are converted by SOD to hydrogen peroxide and as such serve as second messengers for activating a variety of physiological responses. Included in these is a feed-back loop that regulates mitochondria themselves, and within this redox pathway PKC
has been reported to play a prominent role. We hypothesize that when AR disrupts PKC signaling and the attendant feed-back loop, the integrity of mitochondria is compromised.
