Mechanisms of Ageing and Development 125 (2004) 107–110

The immunoregulatory effects of homocysteine and itsintermediates on T-lymphocyte function

Harry Dawson, Gary Collins, Robert Pyle, Vishwa Deep-Dixit, Dennis D. Taub∗

Laboratory of Immunology, Clinical Immunology Section, Gerontology Research Center, National Institute on Aging,NIH, 5600 Nathan Shock Drive, Baltimore, MD 21224, USA

Abstract

Elevated levels of homocysteine (Hcy) have been identified as independent risk identifiers for cardiovascular disease, cerebrovasculardisease as well as for all-cause mortality. Despite the potential importance of these observations, a definitive pathological role for Hcy or itsvarious metabolites in any of these conditions has not been established. Particularly deficient is a description of the effects of elevated levelsof homocysteine on immune function. Folic acid and vitamin B12 deficiency have been independently associated with decreased immunefunction, the apoptosis of bone marrow hematopoietic progenitor cells and the appearance of leukocytes with hypomethylated DNA in theperipheral circulation. A specific role for Hcy or its metabolites in these processes has not been described. We have examined the effectsof Hcy and its various derivatives on T cell activation, differentiation and cell viability. Our results have demonstrated that Hcy is a potentconcentration-dependent T cell activator promoting cellular activation and differentiation as well as potentiating activation-induced cell death(AICD) and cellular apoptosis. Overall, Hcy appears to exert diverse effects on immune function in the circulation and within the tissuemicroenvironment possibly contributing to age-related immune dysfunction and disease pathology.© 2003 Elsevier Ireland Ltd. All rights reserved.

Keywords: Aging; Lymphocytes; Immunodeficiency; Homocysteine; Folate; Apoptosis; Th1; Cardiovascular disease

1. Introduction

Homocysteine (Hcy) is the immediate precursor of theamino acid, methionine (Herrmann et al., 1999; Selhub,1999; Mattson et al., 2002). S-Adenosylhomocysteine(S-Ady Hcy) is formed duringS-adenosylmethionine-depen-dent methylation reactions (Fig. 1). The subsequent hydrol-ysis of S-Ady Hcy results inl-Hcy formation.l-Hcy maybe remethylated to form methionine by a folate-dependentreaction that is catalyzed by the enzyme, methionine syn-thase (vitamin B12-dependent). Alternately, homocysteinemay be metabolized to cysteine in reactions catalyzed bytwo vitamin B6-dependent enzymes. Thus, hyperhomocys-teinemia can be acquired as the result of dietary deficienciesof folate, vitamin B12 and/or vitamin B6. These nutrientsare necessary cofactors for the optimal function of methy-lene tetrahydrofolate reductase (MTHFR) and cystathionine�-synthase (C�S). Deficiencies in the absorption or trans-port of these vitamins can also cause hyperhomocysteine-

∗ Corresponding author. Tel.:+1-410-558-8159;fax: +1-410-558-8284.

E-mail address: taubd@grc.nia.nih.gov (D.D. Taub).

mia. Elevated level of methylmalonic acid, another toxicdownstream derivative of this pathway, typically suggests avitamin B12 deficiency. The most common inherited formof hyperhomocysteinemia results from a mutation in thegene encoding MTHFR, which leads to mild to moderatehyperhomocysteinemia in approximately 15% of patientswith premature cerebrovascular disease. A less commonform of inherited hyperhomocysteinemia is a heterozygousC�S deficiency, where individuals with this disorder areknown to have premature coronary artery disease.

Hcy has recently drawn a great deal of attention as in-creased plasma Hcy levels have been identified as a puta-tive risk factor for a number of age-associated disease statesincluding arteriosclerosis, myocardial infarction, peripheralarterial occlusive disease, subcortical vascular encephelopa-thy, Alzheimer’s disease and neural tube defects (Herrmannet al., 1999; Selhub, 1999; Mattson et al., 2002). PlasmaHcy levels have also been shown to be affected by a numberof biological, environmental or pathological factors includ-ing aging, inflammation, renal insufficiency, and smoking(Mattson et al., 2002). A specific role for Hcy or any of itsmetabolites, such asS-adenosyl Hcy or Hcy thiolactone, inthese conditions has not yet been firmly established.

0047-6374/$ – see front matter © 2003 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.mad.2003.11.013

108 H. Dawson et al. / Mechanisms of Ageing and Development 125 (2004) 107–110

Fig. 1. The methionine-Hcy cycle.

Particularly absent is a description of the effects of ele-vated Hcy levels on immune function. Several studies haveexamined the effects of Hcy on monocyte, neutrophil, andB cell function, inflammation and chemokine production;however, little is known about the Hcy effects on T lym-phocytes (Mattson et al., 2002). Several lines of evidencehave suggested that Hcy may exert a suppressive effect onT cell functions. First, pharmacological agents that inhibitS-Ady-Hcy hydrolase (resulting in significantly increasedintracellular concentrations ofS-Ady Hcy) have been shownto be potent T cell immunosuppressants both in vitro andin vivo (Wolos et al., 1993; Saso et al., 2001). Second,subjects with folic acid or vitamin B12 deficiency, whereHcy levels can become quite elevated, demonstrate an in-crease in frequency of circulating lymphocyte demonstrat-ing nuclear fragmentation and micronuclei (Fenech, 1999).Third, elevated plasma levels of Hcy have been observedin several diseases where there is altered immune func-tion including HIV, common-variable immunodeficiency,systemic lupus erythematosus, inflammatory bowel diseaseand rheumatoid arthritis. Finally, Hcy has been shown tobe quite cytotoxic to cells in culture and may be one ofthe underlying causes of its postulated role in atherogenesisand neuronal degeneration (Kruman et al., 2000; Mattsonet al., 2002). In the current work, we describe the abilityof Hcy and various derivatives to mediate T cell deathas well as T cell activation and differentiation. The im-

plications of these findings to disease pathology shall bediscussed.

2. Results

Given the increased levels of Hcy in elderly subjectsand various inflammatory disease states, we initially soughtto determine whether Hcy exhibits any direct cytotoxic orpro-apoptotic effects on human T lymphocytes. Our stud-ies revealed that treatment of resting human T cells withHcy (10–1000�M) resulted in a dose-dependent increase inapoptotic cell death.d,l-Hcy was more potent than Hcy thi-olactone in this respect whileS-Ady-Hcy was found to besignificantly less active or inactive in many cases. We be-lieve the inactivity ofS-Ady Hcy may be due to its inabil-ity to be optimally transported across the T cell membrane.Apoptosis was assessed through the examination of the per-centage of annexin-PI positive cells, lactate dehydrogenase(LDH) and nuclear matrix protein (NMP) release, and nu-clear fragmentation (Fig. 2A). The concentrations requiredto mediate these effects were in the physiological plasmaand tissue ranges observed in a number of disease states.To examine the mechanism(s) and pathways by which Hcymediates T cell death, we utilized a number of inhibitorsof established apoptotic mediators and pathways to deter-mine their role in our observed effects. We found that the

H. Dawson et al. / Mechanisms of Ageing and Development 125 (2004) 107–110 109

Fig. 2. The effects of Hcy on T cell activation, differentiation and apoptosis. (A) Diagram of molecular mechanisms of Hcy-induced T-cell apoptosisand cell death. (B) Summary of effects of Hcy on T cell function.

pro-apoptotic effects of Hcy were abrogated with the ad-dition of serum, caspase inhibitors, and poly-ADP-ribosepolymerase (PARP) inhibitors to the cell cultures. These re-sults suggest that Hcy, like other apoptotic stressors, leadsto the activation of the caspase cascade and eventually tothe cleavage of the key cellular proteins, like PARP, eventu-ally leading to the typical morphological changes observedin cells undergoing apoptosis. PARP is involved in repair ofDNA damage and functions by catalyzing the synthesis ofpoly (ADP-ribose) and by modifying nuclear proteins andbinding to DNA strand breaks. The ability of PARP to re-pair DNA damage is prevented following cleavage of PARPby caspase-3. In our current model, we have also observedPARP cleavage in resting human T cells post treatment withHcy. Similar inhibition of PARP activity has been reportedto abrogate methyl mercury-induced apoptosis of resting Tcells (Close et al., 1999; Guo et al., 1998) and Hcy-inducedneuronal cell death (Kruman et al., 2000). We have alsofound that Hcy-mediated apoptosis is inhibited by the phos-phatase inhibitor, sodium orthovanadate, the intracellularcalcium chelator, BAPTA-AM, and the protein synthesis in-hibitor, cycloheximide. Moreover, cellular activation via thecell surface ligation of CD3 and/or CD28 also partially ab-rogated Hcy-mediated resting T cell death suggesting thatHcy may mediate its pro-apoptotic effects only on resting Tcells or via ongoing pro-apoptotic signals. In support of this,we have found that Hcy appears to potentiate cellular deathinduced by a number of other established apoptotic signalsincluding activation-induced cell death (AICD), heat shock,and Fas ligand- and HIV-mediated T cell death. It shouldbe noted that this potentiation of ongoing pro-apoptotic sig-naling is not generalizable for all apoptotic stressors as Hcyexhibited little to no effect on�-irradiation-induced death.

Thus, Hcy appears to be a potent T cell apoptotic agent re-sulting in significant cell death with minutes to hours of ex-posure. While the significance of Hcy-induced T cell deathremains to be elucidated, we believe this effect may actu-ally play a role in AICD. We have recently found that Tcells can express Hcy upon activation and that Th1 popula-tions express greater levels of Hcy when compared to Th2and Th0 populations. Given that Th1 cells are more suscep-tible to AICD, it seems possible that Hcy may contribute tothis process during the activation process. A role for Hcy inAICD is currently under investigation.

In addition to the pro-apoptotic effects of Hcy, we alsoexamined the ability of Hcy to influence T cell activationand cytokine secretion. Stimulation of mononuclear cells orisolated T cells with immobilized anti-CD3 mAb in the pres-ence of Hcy or thiolactone but notS-Ady Hcy (10-100�M)resulted in a significant increase in several type 1 cytokinesincluding IL-2, IFN-�, TNF-�, and IL-10 but not type 2 cy-tokines IL-4 or IL-5 (Fig. 2B). More detailed examinationof the Hcy effects on T cell activation revealed that this type1 cytokine production profile is mediated, in part, throughthe production of IL-18 and possibly IL-12. The precisemechanism involved in the generation of these cytokinesis currently under investigation but we believe the Hcy ef-fect is being mediated, in part, by specific stress-associatedsignals post Hcy treatment. Hcy has been shown to inducethe activation of a number of kinase and phosphatase sig-naling pathways within a variety of cell types and has alsobeen shown to have a profound effect on gene expression(Mattson et al., 2002). We also believe that Hcy is a T cellactivator facilitating T cell proliferation and expansion atcertain concentrations and inducing significant apoptosis atother concentrations (Fig. 2B). Given that T cells appear to

110 H. Dawson et al. / Mechanisms of Ageing and Development 125 (2004) 107–110

be able to express Hcy, it seems possible that localized ex-pression of certain forms of Hcy during an immune responsemay potentiate T cell activation and differentiation. Moredetailed analysis of the endogenous role of Hcy, its variousderivatives and the various enzymes within methionine-Hcypathway in the various processes of T cell activation are cur-rently under investigation. Moreover, preliminary data fromour laboratory using microarray technology has revealedsignificant increase in the gene expression of a variety ofmolecules involved in cell cycle, apoptosis, inflammation,metabolism, and cell growth in Hcy-treated lymphocytes.The significance of these gene expression changes and theirrelationship to cellular and oxidative stress and cytokineproduction is currently being explored.

3. Discussion

d,l-Hcy, thiolactone and homocystine to exert a numberof differential effects on immune cells, which may alter im-mune function in the circulation and tissue microenviron-ment with age and disease pathology. The activation state ofa given immune cell and the levels and types of Hcy within atissue microenvironment may significantly influence the lo-cal immune response observed resulting in either enhancedinflammation via immune cell activation and alterations inthe cytokine profiles or diminished responsiveness throughcellular apoptosis and death. In various disease states whereplasma Hcy levels are significantly increased, we believe theHcy levels within a diseased tissue or lesion are significantlyhigher than the plasma levels resulting in even greater acti-vation and cellular damage. Moreover, chronic exposure oftissue and circulating cells to Hcy may even yield more dra-matic alterations in immune cell function. Perhaps Hcy pro-duction within athlerosclerotic lesions may induce T cell andmonocyte activation resulting in T cell differentiation into aTh1 phenotype. These proinflammatory Th1 cells may fur-ther activate the local macrophage, smooth muscle and foamcell populations potentiating plaque development. Overall, agreater understanding of the potential modulatory effects ofHcy and its metabolites on immune function may result in

the development of potential therapeutic strategies to con-trol and optimize immune responses with age and in variousage-associated disease states.

Acknowledgements

By acceptance of this article, the publisher or recipientacknowledges the right of the US Government to retain anonexclusive, royalty-free license in and to any copyrightcovering the article.

References

Close, A.H., Guo, T.L., Shenker, B.J., 1999. Activated human T lympho-cytes exhibit reduced susceptibility to methylmercury chloride-inducedapoptosis. Toxicol. Sci. 49, 68–77.

Fenech, M., 1999. Micronucleus frequency in human lymphocytes isrelated to plasma vitamin B12 and homocysteine. Mutat. Res.16:428 (1/2), 299–304.

Guo, T.L., Miller, M.A., Datar, S., Shapiro, I.M., Shenker, B.J., 1998.Inhibition of poly(ADP-ribose) polymerase rescues human T lympho-cytes from methylmercury-induced apoptosis. Toxicol. Appl. Pharma-col. 152, 397–405.

Herrmann, W., Quast, S., Ullrich, M., Schultze, H., Bodis, M., Geisel,J., 1999. Hyperhomocysteinemia in high-aged subjects: relation ofB-vitamins, folic acid, renal function and the methylenetetrahydrofo-late reductase mutation. Atherosclerosis 144, 91–101.

Kruman, I.I., Culmsee, C., Chan, S.L., Kruman, Y., Guo, Z., Penix, L.,Mattson, M.P., 2000. Homocysteine elicits a DNA damage response inneurons that promotes apoptosis and hypersensitivity to excitotoxicity.J. Neurosci. 20, 6920–6926.

Mattson, M.P., Kruman, I.I., Duan, W., 2002. Folic acid and homocysteinein age-related disease. Ageing Res. Rev. 1 (1), 95–111.

Saso, Y., Conner, E.M., Teegarden, B.R., Yuan, C.S., 2001.S-Adenosyl-l-homocysteine hydrolase inhibitor mediates immunosuppressive ef-fects in vivo: suppression of delayed type hypersensitivity ear swellingand peptidoglycan polysaccharide-induced arthritis. J. Pharmacol. Exp.Ther. 296, 106–112.

Selhub, J., 1999. Homocysteine metabolism. Annu. Rev. Nutr. 19, 217–246.

Wolos, J.A., Frondorf, K.A., Davis, G.F., Jarvi, E.T., McCarthy, J.R.,Bowlin, T.L., 1993. Selective inhibition of T cell activation by aninhibitor of S-adenosyl-l-homocysteine hydrolase. J. Immunol. 150,3264–3273.