Multiple myeloma (MM) is a hematological malignancy of bone marrow resident plasma cells. Although patient outcomes have improved with new classes of proteasome inhibitors, immunomodulatory drugs, and targeted monoclonal antibodies, the disease remains largely incurable. Understanding the mechanisms by which resistant myeloma cells are able to survive is critical for the development of therapeutic strategies to prevent patient relapse.

We have published that expression of CD28, the canonical T cell co-stimulatory molecule, increases during myeloma disease progression, is associated with poor prognosis, and provides a pro-survival signal to myeloma cells in vitro and in vivo; however, the mechanistic basis for this is unclear. During T cell activation, CD28 induces glycolysis at the expense of mitochondrial respiration. In four myeloma cell lines tested (MM1S, U266, 8226 and KMS11,) CD28 is able to increase the expression of glucose transporter glut1. Upregulation of glut1 is PI3k dependent, as treatment with the PI3k inhibitor LY294002 prevents CD28-mediated upregulation of glut1. Interestingly, CD28 has minimal effect on the glycolytic rate, but does increase the glycolytic capacity and reserve, suggesting that CD28 enhances metabolic fitness in myeloma cells when nutrient availability is low.

Recently published data demonstrate a critical role for the metabolic phenotype in driving myeloma resistance to treatment, and that disruption of mitochondrial biogenesis promotes myeloma disease progression. CD28 activation induces mitochondrial biogenesis in multiple myeloma, increasing both the basal oxygen consumption as well as the maximal respiratory capacity. Mechanistically, CD28 activation increases protein levels of Irf4 and its direct target Myc in myeloma cells, which has been demonstrated in the literature to regulate mitochondrial biogenesis.

Both glycolysis-derived pyruvate and lipid based fatty acids (FA) can be used in mitochondrial respiration. To our great surprise, CD28-mediated mitochondrial respiration is dependent upon fatty acid uptake through the mitochondrial membrane resident FA transporter CPT1a rather than glucose-derived pyruvate. In the bone marrow microenvironment, myeloma cells may not be able to depend upon the extracellular availability of glucose or fatty acids for use in energy production. Interestingly, CD28 activation increases autophagy in myeloma cells, a process of cellular degradation that has the capacity to provide lipid precursors for use in fatty acid oxidation. Inhibition of autophagy depresses myeloma mitochondrial respiration, suggesting a model wherein CD28-mediated upregulation of Irf4-Myc drives mitochondrial biogenesis, and by inducing autophagy, CD28 provides a supply of fatty acid fuel for increased mitochondrial respiration.

A major byproduct of mitochondrial respiration is the production of reactive oxygen species (ROS), known to regulate both cell signaling as well as DNA damage. Inhibition of ROS prevents CD28-mediated survival in multiple myeloma. ROS augment PI3k signaling in myeloma, as inhibition of ROS leads to decreased phosphorylation of Akt, critical for CD28-mediated survival signaling. Furthermore, ROS inhibition prevents CD28-mediated increases in mitochondrial biogenesis, mitochondrial respiration, and the maximal respiratory capacity. In this manner, byproducts of CD28-mediated metabolic fitness reinforce signaling for myeloma cell survival.

Treatment with a sublethal dose of Bortezomib decreased glycolytic and mitochondrial metabolism in myeloma. This leads to an attractive clinical application of our findings, wherein blockade of CD28-mediated metabolic fitness with Abatacept, an FDA approved drug for rheumatoid arthritis, may augment the efficacy of Bortezomib and other therapeutic strategies that exploit the metabolic demands of multiple myeloma throughout disease progression and relapse.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.