Anil B. Mukherjee, MD, PhD, Head, Section on Developmental Genetics
Zhongjian Zhang, MD, PhD, Staff Scientist
Rabindranath Ray, PhD, Postdoctoral Fellow
Pei Chen Tsai, MS, Technical Training Fellow
Sung-Jo Kim, PhD, Visiting Fellow
Yi-Ching Lee, PhD, Visiting Fellow
Moonsuk Choi, PhD, Adjunct Scientist
Sondra W. Levin, MD, Adjunct Scientist
Aprile Pilon, PhD, Adjunct Scientist
Jingya Wang, BA, Summer Intern
Alison Heffer, BS, Summer Intern
Emiko Hitomi, MD, Guest Researcher

We conduct laboratory and clinical investigations into the molecular mechanisms of complex genetic disorders of inflammation, autoimmunity, and neurodegeneration in order to develop novel and rational therapeutic approaches. Our research focuses primarily on the regulation and physiological functions of two genes: uteroglobin and palmitoyl-protein thioesterase-1 (PPT1). Mutations of PPT1 are the genetic basis of infantile neuronal ceroid lipofuscinosis (INCL), a devastating neurodegenerative storage disease of childhood also known as Batten disease. Investigations into both genes have led to ongoing clinical trials. Recently, by using PPT1-knockout mice, which recapitulate virtually all clinical and pathological features of INCL, we discovered that PPT1 deficiency leads to endoplasmic reticulum (ER) stress, which mediates the activation of unfolded protein response (UPR). In addition, we discovered that overwhelming ER stress activates cysteine proteinase in the ER as well as caspase-12 in mice and caspase-4 in humans, leading to caspase-3 activation and apoptosis. Moreover, ER stress causes disturbance in Ca2+ homeostasis and generates reactive oxygen species (ROS), which activates the mitochondrial caspase-9, further contributing to rapid neuronal death. Our results are significant not only for providing insight into a complex mechanism of neurodegeneration in INCL but also for identifying several potential targets for the development of rational therapeutic approaches for this uniformly fatal disease.

Uteroglobin prevents pulmonary fibrosis

Lee, Zhang, Mukherjee

Pulmonary fibrosis (PF) is a complex disease with high mortality and morbidity. Although the molecular mechanism(s) of PF remains unclear, it appears that a breakdown of homeostatic mechanisms that normally prevent inflammation plays critical pathogenic roles. Uteroglobin (UG), also known as Clara cell 10 kDa protein (CC10), the founding member of the newly recognized secretoglobin superfamily, is a steroid-inducible secreted protein constitutively expressed by the airway epithelia in all mammals; the protein manifests potent anti-inflammatory and anti-chemotactic properties. Mice lacking UG sporadically develop focal PF, although the underlying molecular mechanism(s) are not yet understood. We find that UG-KO mice are extraordinarily sensitive to bleomycin, an anti-cancer agent known to induce PF, and that they readily develop PF when treated with an extremely low dose of the agent, which has virtually no effect on wild-type littermates. We further demonstrated that UG prevents PF by suppressing bleomycin-induced production of pro-inflammatory T-helper 2 cytokines and FGF-β, which are also pro-fibrotic. Our results define a critical role of UG in preventing the development of PF and provide the proof of principle that recombinant UG may have therapeutic potential.

Interaction of uteroglobin with lipocalin-1 receptor suppresses cancer cell motility and invasion

Zhang, Kim, Wang, Chowdhury, ¹ Lee, Choi, Mukherjee

Cellular migration and invasion are precisely regulated processes that play critical roles in chemotaxis, embryonic development, and implantation. Dysregulated migration and invasion of cancer cells lead to metastasis, which accounts for more than 90 percent of cancer deaths. Thus, understanding the molecular mechanism(s) that regulate cancer cell motility and invasion may facilitate the development of novel therapeutic approaches. We previously reported that UG 8.75 binds to several cell types, including some cancer cells, and inhibits their migration and invasion. More recently, we reported that HTB-81 adenocarcinoma cells, which do not bind to UG, are refractory to UG-mediated inhibition of migration and invasion. Given that UG shares some biological properties with lipocalin-1, which binds to several cell types via its receptor (Lip-1R), we sought to determine whether UG might also bind to this receptor and mediate inhibitory effects on migration and invasion of HTB-81 adenocarcinoma cells. Using COS-1 cells transfected with a green fluorescent protein-Lip-1R cDNA construct, we demonstrated that induced expression of Lip-1R facilitates specific binding of ¹²⁵I-humanUG (hUG) to the transfected cells, which do not normally bind to hUG. We further demonstrated that forced expression of Lip-1R in HTB-81 cells promotes binding of ¹²⁵I-hUG with high affinity and specificity. As a result of Lip-1R expression, HTB-81 cells become fully responsive to hUG-mediated inhibition of migration and invasion. Our results suggest that Lip-1R is a UG-binding protein and that its forced expression in UG-resistant cancer cells renders such cells sensitive to UG-mediated inhibition of migration and invasion.

A balance between pro- and anti-inflammatory prostaglandins maintains homeostasis

Mandal, ² Zhang, Mukherjee

Cyclooxygenase-2 (COX-2) is a critical enzyme for the production of lipid mediators of inflammation such as prostaglandins (PGs). Thus, for many years, COX-2 has been the favorite target for anti-inflammatory drug development. However, recent revelations of the adverse effects of selective COX-2 inhibitors have been the subject of intense debate. While some PGs are pro-inflammatory, others have anti-inflammatory effects, although the molecular mechanism(s) by which they achieve such effects remains unclear. We found that, via receptor-mediated pathways, inflammatory PGs (e.g., PGD2 and PGF₂α) mediate the activation of nuclear factor kappa B (NF-kB). NF-kB stimulates the expression of COX-2, which is essential for inflammatory PG production. Most interestingly, the anti-inflammatory PG (PGA1) counteracts the activation of NF-kB and consequently suppresses COX-2 gene expression. Our results suggest that pro- and anti-inflammatory PGs play a "yin-yang" role to counteract each other's effects and maintain homeostasis. We propose that selective COX-2 inhibitors disrupt such balance and thus manifest the reported adverse effects.

Endoplasmic reticulum stress-mediated apoptosis leads to neurodegeneration in infantile Batten disease

Zhang, Kim, Lee, Tsai, Choi, Heffer, Mukherjee

Numerous proteins undergo modification by palmitic acid (S-acylation) in order to perform their biological functions, including signal transduction, vesicular transport, and maintenance of cellular architecture. Although palmitoylation is an essential modification, such proteins must also undergo depalmitoylation for their degradation by lysosomal proteases. The lysosomal enzyme palmitoyl-protein thioesterase-1 (PPT1) cleaves thioester linkages in S-acylated proteins and removes palmitate residues, facilitating the proteins' degradation. Thus, inactivating mutations in the PPT1 gene cause infantile neuronal ceroid lipofuscinosis (INCL). Although rapidly progressing brain atrophy is the most dramatic pathological manifestation of INCL, the underlying molecular mechanism(s) remains unclear. Using PPT1-knockout (PPT1-KO) mice that mimic human INCL, we found that the endoplasmic reticulum (ER) in the mouse brain cells is structurally abnormal. Further, we demonstrated that the level of growth-associated protein-43 (GAP-43), a palmitoylated neuronal protein, is elevated in the mouse brains. Moreover, forced expression of GAP-43 in PPT1-deficient cells results in the abnormal accumulation of the protein in the ER. Consistent with these results, we found evidence for activation of unfolded protein response (UPR) marked by elevated levels of the phosphorylated translation initiation factor eIF2alpha, increased expression of chaperone proteins such as glucose-regulated protein-78, and activation of caspase-12, a cysteine proteinase in the ER that mediates caspase-3 activation and apoptosis. Our results link, for the first time, PPT1 deficiency with activation of UPR, apoptosis, and neurodegeneration in INCL and identify potential targets for therapeutic intervention in this uniformly fatal disease.

Endoplasmic reticulum stress-mediated unfolded protein response activates caspase-4 causing neuronal apoptosis and degeneration in INCL

Kim, Zhang, Hitomi, Lee, Mukherjee

As stated above, INCL is caused by mutations in the PPT1 gene. PPT1 cleaves thioester linkages in S-acylated (palmitoylated) proteins, and mutation of the gene causes abnormal intracellular accumulation of fatty-acylated proteins and peptides, thus leading to INCL pathogenesis. Although apoptosis is the suggested cause of neurodegeneration in INCL, the molecular mechanism(s) of apoptosis remains unclear. Using the PPT1-KO mice, which mimic INCL, we previously reported that one mechanism of apoptosis involves ER stress-induced caspase-12 activation. However, the human caspase-12 gene contains several mutations that make it functionally inactive. Thus, it has been suggested that human caspase-4 is the counterpart of murine caspase-12. Here we report that, in the human INCL brain, ER stress-induced activation of unfolded protein response (UPR) mediates caspase-4 and caspase-3 activation and apoptosis. Moreover, we showed that the INCL brain contains a high level of GAP-43, which is known to undergo palmitoylation. We also demonstrated that transfection of cultured INCL cells with a green fluorescent protein-GAP-43 cDNA construct shows abnormal localization of the protein in the ER. Further, INCL cells manifest evidence of ER stress and UPR (elevated levels of Grp-78/Bip and GADD153), caspase-4 as well as caspase-3 activation, and cleavage of poly(ADP)-ribose polymerase, which is a compelling sign of apoptosis. Most importantly, we show that inhibition of caspase-4 activity protects INCL cells from undergoing apoptosis.

Caspase-9 activation contributes to rapid neurodegeneration in INCL

Kim, Zhang, Lee, Mukherjee

As mentioned above, we recently reported that one of the major pathways of neuronal apoptosis that mimics INCL in PPT1-KO mice is mediated by ER stress-induced caspase-12 activation. ER stress also increases the production of reactive oxygen species (ROS), disrupts Ca2+ homeostasis, and increases the potential for destabilizing mitochondrial membrane. Mitochondrial membrane destabilization activates the caspase-9 present in this organelle and can mediate apoptosis. We found that the levels of superoxide dismutase (SOD) in human INCL and PPT1-KO mouse brain tissues—most likely induced by ROS—are markedly elevated. Moreover, we demonstrated an increase in these tissues of activated caspase-3 and cleaved-PARP, which are indicative of apoptosis. Using cultured neurospheres from PPT1-KO and wild-type mouse fetuses, we further demonstrated elevated levels of ROS, SOD-2, cleaved caspase-9, activated caspase-3, and cleaved-PARP. We propose that ER stress attributable to PPT1 deficiency increases ROS and disrupts calcium homeostasis and thus activates caspase-9 and that caspase-9 activation mediates caspase-3 activation and apoptosis, contributing to rapid neurodegeneration in INCL.

A combination therapy with Cystagon™ and N-acetylcysteine for INCL patients

Levin, Zhang, Caruso, Gropman, Mukherjee

NCL is the most common (1 in 12,500) heritable neurodegenerative lysosomal storage disorder of childhood. Mutations of at least seven genes are responsible for various forms of NCL. The infantile form of NCL (INCL) is the most severe disease and, as stated above, is caused by inactivating mutations in the gene encoding lysosomal PPT1 located on chromosome 1p32. PPT deficiency leads to abnormal accumulation of acylated-proteins, called ceroids, in lysosomes. Thus, INCL is a newly recognized lysosomal storage disease. Given that thioester linkages are susceptible to nucleophilic attack, drugs with nucleophilic property have therapeutic potential for INCL. As described above, we recently demonstrated that neuronal death in INCL results from ER stress-mediated activation of unfolded protein response (UPR) and caspase-12 activation, which leads to apoptosis. Moreover, ER stress leads to increased production of ROS, which mediates the activation of caspase-9 and apoptosis. Thus, agents that scavenge ROS may be of therapeutic benefit. Accordingly, we tested several compounds with nucleophilic and ROS-scavenging properties (i.e., cysteamine, phosphocysteamine, cysteamine bitartrate, and N-acetylcysteine) and found that the compounds disrupt thioester linkages in the model PPT1-substrate [14C] palmitoyl~CoA, releasing [14C] palmitic acid (Zhang et al. Nat Med 2002;7:478).

For the last three and a half years, we have been conducting a clinical trial to determine whether Cystagon™ is beneficial for INCL patients. So far, we have treated six patients (three females and three males). Our preliminary results indicate that Cystagon™ slows the rapid neurodegeneration that is characteristic of INCL. Moreover, our patients have not developed new epileptic seizures, a common complication of the disease. In fact, before the initiation of Cystagon™ treatment, one patient's EEG revealed epileptic foci, which were not detected in follow-up EEG tests after treatment with the drug. The most dramatic effect of Cystagon™ is the complete clearance of lysosomal ceroids in white blood cells within six months of therapy. In parallel with the clinical trial, we used a mouse model of INCL to test the effects of Cystagon™ alone or in combination with N-acetylcysteine (Mucomyst®), which manifests anti-apoptotic and neuroprotective properties. Our preliminary results show that the combination therapy delays the development of neurological symptoms and reduces apoptosis in the mice for a longer period of time than in mice treated with Cystagon™ alone. These results prompted us to implement a combination therapy with Cystagon™ and Mucomyst® for INCL patients. So far, we have initiated the combination therapy in three patients, two of whom initially received Cystagon™ alone. The last patient admitted to our protocol is a ten-month-old child with INCL who is the youngest patient recruited so far. The child is expected to provide valuable data regarding the efficacy of the combination therapy. Moreover, the participation of the child in the protocol may enable us to document a complete natural history of INCL, which is not currently available.

¹ Bhabadeb Chowdhury, PhD, former Research Fellow

² Asim K. Mandal, PhD, former Postdoctoral Fellow

COLLABORATORS

Eva Baker, MD, PhD, Diagnostic Radiology Department, Warren G. Magnuson Clinical Center, NIH, Bethesda, MD
Bruce J. Baum, MD, Gene Therapy and Therapeutics Branch, NICDR, Bethesda, MD
Rafael Caruso, MD, Ophthalmic and Visual Function Branch, NEI, Bethesda, MD
Eli Eisenstein, PhD, Hadassah Medical Center, Jerusalem, Israel
Jiliang Gao, PhD, Laboratory of Molecular Immunology, NIAID, Bethesda, MD
Andrea Gropman, MD, Neurology Department, Georgetown University Medical Center, Washington, DC
Sandra L. Hofmann, MD, PhD, University of Texas Southwestern Medical Center, Dallas, TX
Shau-Ku Huang, PhD, The Johns Hopkins University Medical School, Baltimore, MD
Alan Koretsky, PhD, NMR Center, NINDS, Bethesda, MD
Jeeva Munasinghe, PhD, NMR Center, NINDS, Bethesda, MD
Nicholas Patronas, MD, Diagnostic Radiology Department, Warren G. Magnuson Clinical Center, NIH, Bethesda, MD
Nagarajan Pattabiraman, PhD, Lombardi Cancer Center, Washington, DC
Scott Paul, MD, Department of Rehabilitation Medicine, Warren G. Magnuson Clinical Center, NIH, Bethesda, MD
Zenaide Quezado, MD, Department of Anesthesiology, Warren G. Magnuson Clinical Center, NIH, Bethesda, MD
Gary Silverman, PhD, University of Pittsburgh School of Medicine, Pittsburgh, PA
Krystyna Wisniewski, MD, PhD, New York State Institute for Basic Research, _Staten Island, NY
Yi-Jin Xiao, PhD, Cleveland Clinic, Cleveland, OH
Yan Xu, PhD, Cleveland Clinic, Cleveland, OH
Junying Yuan, PhD, Harvard Medical School, Boston, MA

For further information, contact mukherja@mail.nih.gov.