CD28 Induces Mitochondrial Respiration through Irf4 for Long Lived Plasma Cells Survival

Successful vaccination strategies against emerging pathogens like that of the Zika virus depend upon the sustained production of antigen-specific antibodies by terminally differentiated B cells known as plasma cells. Because the half-life of circulating antibodies is on the order of days to weeks, durable humoral immunity requires the induction of long-lived plasma cells (LLPCs) that home to specialized survival niches in the bone marrow and live indefinitely. Therefore, understanding the cellular and molecular interactions that govern LLPC survival are critical for the design of effective vaccines.

We have published that CD28, the canonical T cell co-stimulatory molecule, is required for LLPC survival in the bone marrow. In order to maintain this extraordinary degree of longevity, LLPCs must be able to compete successfully in an environment wherein normal hematopoiesis puts constraints on the availability of nutrients. Furthermore, LLPCs are highly biosynthetic as antibody producing factories and necessitate a high level of metabolic activity. In T cells CD28 is known to induce glycolysis at the expense of mitochondrial respiration. To our great surprise, CD28 signaling in LLPCs increases mitochondrial respiration whilst not greatly affecting glycolysis. However, CD28 enhanced the glycolytic capacity and reserve, suggesting that CD28 signaling regulates their ability to better compete for nutrient availability and metabolic fitness. CD28 induces the expression of the glucose transporter Glut1 and subsequent glucose uptake as assessed by staining with the autofluorescent glucose analogue NBDG. In contrast to the T cell literature, CD28 increases the mitochondrial mass and membrane potential in LLPCs, suggesting that rather than using glucose primarily for glycolysis it is converted into pyruvate which then is used in mitochondrial respiration.

Mechanistically, CD28 induces NFkB dependent IRf4 upregulation, a transcription factor known to regulate metabolism in T cells and survival in LLPCs. Furthermore, inhibition of NFkB abrogates CD28-induced increases in glucose uptake and mitochondrial mass. We have recently published that the Grb2/Vav binding domain on the CD28 cytoplasmic tail is required for CD28-mediated survival. In mice wherein this domain is mutated (AYAA mutants), LLPCs have decreased glucose uptake and mitochondrial mass. Plasma cells from AYAA mutant mice also have decreased levels of Irf4, facilitating a model wherein CD28 through its Grb2/Vav binding domain increased Irf4 for metabolic fitness. We also describe for the first time an NFkB superenhancer element upstream of the Irf4 transcriptional start site, suggesting that CD28 may govern both direct Irf4 promoter activity as well as DNA folding for increased Irf4 production. By using siRNA to target Irf4 at the genetic level, we see diminished mitochondrial mass, suggesting that Irf4 is the direct regulator of CD28-regulated metabolism in LLPCs.

A major byproduct of mitochondrial respiration is the production of reactive oxygen species (ROS). CD28 increased the amount of ROS production in LLPCs as seen by staining with a mitochondrial-ROS specific dye. This is counterintuitive, as ROS are well characterized cell damaging agents. Paradoxically, when we inhibited ROS with MnTBAP, a ROS scavenger, CD28 no longer conferred a survival signal in LLPCs. Taken together these results suggest a model wherein CD28 through its Grb2/Vav binding domain induces NFkB dependent upregulation of Irf4 directly on the promoter and augments further Irf4 production via a previously undescribed superenhancer element upstream of the Irf4 gene. Irf4 then goes on to increase mitochondrial respiration for CD28-mediated LLPC survival and metabolic fitness. This understanding will not only allow for the augmentation of vaccine design and efficacy, but may also lead to novel therapeutic strategies to combat antibody-mediated autoimmunity.