FAQ

Study Sections

Results of FARE 2005

Results of FARE 2004

Results of FARE 2003

Results of FARE 2002

Results of FARE 2001

FARE 2004 Winning Abstracts

View (without abstracts) sorted by Name
View (without abstracts) sorted by Institute/Center
View (without abstracts) sorted by Study Section

Rachana Agarwal, Ph.D - NIA
Role of claudin-3 and claudin-4 in ovarian cancer
Ovarian cancer is one of the leading causes of cancer deaths in women due to difficulties in both diagnosis and therapy. A better understanding of the pathways involved in ovarian tumorigenesis will likely provide new targets for early and effective intervention.
Using serial analysis of gene expression, we have found that claudin-3 and claudin-4 are among the most highly overexpressed genes in ovarian cancer. Claudin proteins represent a large family of integral membrane proteins crucial for tight junction (TJ) formation and function. Although the roles of claudins in tumorigenesis remain unknown, structural and functional alterations in TJ have been suggested to be associated with epithelial carcinogenesis. In order to begin elucidating the function of claudin-3 and claudin-4 in ovarian tumorigenesis, we have 1) investigated their patterns of expression in ovarian tissues, and 2) created normal ovarian surface epithelial cell lines that exogenously express these proteins. While expressed at low levels in most normal human tissues, including the ovary, claudin-3 and claudin-4 are highly up-regulated in ovarian cancers of all subtypes. Immunohistochemical analyses using our ovarian tissue array confirmed the high level of expression of claudin-3 and claudin-4 in the majority of ovarian cancer tissues but many tumors exhibited cytoplasmic staining. Non-malignant ovarian cystadenomas did not frequently overexpress these proteins. In ovarian cancer cell lines, claudin-3 and claudin-4 expression was not associated with changes in transepithelial electrical resistance (TER), suggesting that claudin expression has no effect on the TJ function of these cells, but may have other functions such as signaling. In order to further investigate the effects of claudins in ovarian tumorigenesis, we have made mammalian expression vectors expressing full length, wild-type claudin-3 and claudin-4 and obtained human ovarian surface epithelial (HOSE-B) cells constitutively overexpressing these proteins. Although expression of these claudins in HOSE-B cells did not result in functional TJ formation, claudin-4 over-expressing HOSE-B cells demonstrated increased invasion through matrigel-coated filters. These results for the first time suggest a role of claudin proteins in ovarian cancer cell invasion and motility. Further work is in progress to understand the pathways involved in this process.

Eiman A Aleem, Ph.D. - NCI-CCR
LOSS OF CDK2 LEADS TO SEVERE GROWTH DEFECTS IN MOUSE EMBRYONIC FIBROBLASTS (MEFs)
Cyclin-dependent kinases (Cdks) control cell cycle progression in mammalian cells. Unlike Cdk4 and Cdk6, whose function appear to be restricted to the early G1 phase, Cdk2 plays a central role in most cell cycle phases: It is activated by cyclin E in late G1 to promote S phase entry, and by cyclin A during S and G2/M phases. Moreover, Cdk2 has been implicated in the initiation and maintenance of DNA replication, centrosome duplication and apoptosis.
To analyze the specific role of Cdk2 in vivo we generated Cdk2 knockout mice, which are unexpectedly viable, and studied the growth properties of early (passage 2) and late passages (> 30) of MEFs derived from Cdk2-/- and Cdk2+/+ mice.
Cdk2-/- MEFs show severe growth defects. Although continuous proliferation is not impaired in Cdk2 -/- MEFs, the number of daily counted cells decreased by 55 % in comparison to Cdk2+/+ MEFs from the same litter. In order to gain more insight into the underlying mechanism of this growth defect, which persisted in late passages, we studied the effect of serum starvation on cell cycle kinetics and apoptosis. The slower growth rate of Cdk2 -/- MEFs could be attributed to several factors acting in concert, such as: (1) some delay in S phase entry as revealed by 1.6 fold decrease in the percentage of BrdU-labeled cells compared to Cdk2+/+ MEFs after 15 hr of release, (2) enhanced apoptosis through 2-3 fold increase in caspase-3 activity in serum starved Cdk2 -/- compared to Cdk2+/+ MEFs until the 18 hr release, as well as by upregulation of p21 in Cdk2 -/- versus Cdk2+/+ MEFs, (3) Induction of senescence in late passages of Cdk2 -/- MEFs. While the expression of Cdc2, Cdk4 and Cdk6 proteins were constant in both Cdk2 -/- and Cdk2+/+ MEFs, p27 was less expressed in Cdk2 -/- MEFs. The slight increase in Cdc2 activity starting from 12 hr of release in Cdk2 -/- compared to Cdk2+/+ MEFs indicate that Cdc2 may compensate for the loss of Cdk2. Cyclin E is present in complexes in Cdk2 -/- MEFs, an unexpected finding, since Cdk2 is considered the major binding partner of E-type cyclins.
Further studies include DNA damage and oncogene-mediated cell transformation in Cdk2 -/- MEFs. Our results address important questions, namely: (1) whether there is functional overlap between Cdk2 and other Cdks, which renders Cdk2 dispensable, (2) whether there is a distinct kinase other than Cdk2 that is able to bind with cyclin E in Cdk2 -/- MEFs, (3) the potential of Cdk2 as a target for cancer therapy.

Johnny O Alexander, M.D. - CC
Systemic prostate specific antigen (PSA) is a measure of vascular permeability: A study of Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI).
Purpose: To evaluate the correlation between vascular permeability and prostate specific antigen in prostate disease using DCE-MRI. Methods and Materials: Prostate cancer is a major cause of morbidity and mortality in American men and nearly 29,000 men will die from the disease this year. Given its importance much attention has been focused on early diagnosis and treatment. It is well accepted that progressively higher PSA levels correlate with greater likelihood of malignancy. The major focus has been on correlating PSA with malignancy and not on the relationship between PSA and vascular changes within the prostate. It is our hypothesis that systemic PSA levels correlate with vascular permeability changes in the prostate and not necessarily with malignancy. A group of patients were selected for this study after satisfying the following criteria: 1) evaluated with DCE-MRI; 2) serum PSA available prior to imaging study; 3) biopsy or surgical validation available. The images from DCE-MRI study were evaluated in an image processing program written in IDL (Interactive Data Language). The uptake slope of the contrast enhancement curve was measured on a voxel by voxel basis. We defined the voxels with a higher slope 1.5 X enhancement/minute as high permeability voxels (HPV) and computed the total volume of HPV. Results: 32 patients were categorized by their PSA levels and grouped in low(n=6), moderate(n= 18), or high(n=8) PSA ranges. The low PSA range was set at <4ng/ml, moderate 4-10ng/ml, and high >10ng/ml. The DCE-MRI exam consists of the following 3DFSPGR (T2*) sequence, injection rate 0.1mmol/kg at 3cc/sec, total time, 4 minutes. The patients average PSA were 1.88(low), 6.56(mod.), and 20.6 (high). And volume of HPV were 3.5 (+/-4.4), 9.2 (+/-6.4), and 15.3 (+/-10.5) respectively. These results were significant (p<0.05) and suggest patients with high vascular permeability have high serum PSA, and patients with low HPV have low PSA levels. Two patients with low PSA had cancer, while 2 patients in the high PSA group had prostatitis. Conclusion: There was a strong correlation between the PSA value and the total volume of high permeability voxels in both cancerous and non-malignant prostate tissue. Other serum markers for disease such as CEA may have similar correlation with vascular changes and angiogenesis.

Juan Alguacil, M.D. - NCI-DCEG
Urinary pH, cigarette smoking and bladder cancer risk in Spain
Aromatic amines play an important role in bladder carcinogenesis from tobacco smoking. Urine pH has been found to strongly influence the presence of free aromatic amine compounds in the urine. Liver-synthesized N-glucuronides (aromatic amines metabolites) are excreted into the urinary bladder, where they are relatively stable under neutral pH conditions. However, in the presence of acidic conditions, they are rapidly hydrolyzed leading to the formation of their corresponding arylamines, which can then undergo further metabolism to form DNA adducts. We conducted a hospital-based case-control study based on interviews with 1212 incident bladder cancer cases and 1276 hospital controls between May 1998 and June 2001 in 18 hospitals in Spain. Of these, 737 cases and 625 controls measured their urinary pH twice a day (early in the morning and before dinner) during 4 consecutive days two weeks after hospital discharge. Subjects with all of their pH readings less than or equal to 6 (46% of cases, 39% of controls) were categorized as having a consistently acidic urine pH. The bladder cancer risk associated with acidic urine pH was OR=1.4 (95% CI 1.04-1.8), adjusted for age, gender, geographic area, current tobacco use, frequency of urination, and educational level. We observed a significant dose-response relationship for increased urine acidity and bladder cancer risk. Compared to subjects with urine pH > 6.0, subjects with urine pH > 5.0 to 6.0 had an OR of 1.3 (1.01-1.7) and those with urine pH <= 5 had an OR of 1.7 (0.9-2.9). Gender specific analyses showed higher risk for urinary pH among men. Urinary pH modified the association between cigarette smoking and bladder cancer. Risk estimates for current cigarette smokers were OR=5.8 (3.5-9.5) and OR=11.0 (6.2-20.0) among subjects with non-acidic pH and acidic pH, respectively. Voiding at night appeared to modify the risk of cigarette smoking among subjects with acidic pH, but had no impact on those with non-acidic pH. Subjects with acidic urine who voided at night had an OR=6.7 (3.0-15), while those with acidic urine who did not void at night had an OR=15.0 (7.2-31). This is the first epidemiologic study reporting an increased risk of bladder cancer among subjects with acidic urine. Urine pH appears to modify the carcinogenic effect of tobacco in the bladder. In summary, results from this study support the hypothesis that urinary pH plays an important role in human bladder carcinogenesis.

Oral Alpan, MD - NIAID
Overcoming the immunological barriers to curing hemophilia
Severe bleeding episodes in hemophilia patients are currently treated by infusion of the missing factor (factor VIII or IX), a method of treatment that is expensive, occasionally requires hospitalization, carries the risk of transmission of hepatitis and other viruses, and often results, after long term treatment, in the production of inhibitory antibodies that render further treatments difficult. Although gene therapy studies have shown that the missing factors can be replaced, the host's immune system nearly always produces a response that causes the rejection of the transfected cells and/or generates the production of inhibitory antibodies, and thus limits the useful lifespan of the therapy. We initially decided to tackle this problem in Hemophilia B mice, by feeding the missing clotting factor to induce oral 'tolerance', a form of immunity in which the host's immune response to a particular protein becomes specifically hypo-reactive after oral administration of that protein. As a safe and inexpensive source of F.IX we used milk from a genetically engineered pig that secretes 3g/liter of human factor IX into its milk. While studying the effect on the immune system of feeding this F.IX-containing pig milk, we discovered the truly astonishing finding that feeding the milk to hemophilia B mice not only prevented their immune response to F.IX (upon subsequent immunization with F.IX in CFA), but also treated their bleeding disorder. Hemophiliac mice that had been fed with the F.IX milk no longer bled profusely, but had perfectly normal clotting times, and this lasted for up to two days after a single feeding. We also demonstrated that repeated feeding of F.IX containing milk every other day for 2 months did not generate antibodies in the blood in hemophilia B mice, and protected them from the formation of inhibitory antibodies after deliberate immunization. Based on these results, we tested whether F.VIII, which is almost 8 times larger than F.IX, would behave similarly, and found that oral administration also corrects bleeding time and induces oral tolerance to F.VIII. Thus oral administration of clotting factors is a promising to induce tolerance to F.IX for successful gene therapy and at the same time provide hemostasis to prevent complications secondary to bleeding. We are now planning a pre-clinical trial with hemophilic puppies, to determine if this might be a safe and effective treatment for hemophilic human babies.

Zandrea Ambrose, Ph.D. - NCI-CCR
A Novel Microbicide that Prevents Intravaginal Transmission of SIV
Sexual transmission accounts for greater than 90% of worldwide HIV infection. Moreover, the incidence and prevalence of HIV infection in women has been increasing. Vaginal microbicides provide a female-controlled strategy to prevent HIV transmission. We have chosen to evaluate an HIV inactivating agent, 2-hydroxypropyl-betacyclodextrin (BCD) as a potential microbicide. Significantly, BCD recently has been proven as an effective microbicide in a mouse model for intravaginal HIV-1 transmission and is used extensively for other purposes in individuals. First, we evaluated the efficacy of virus neutralization in vitro, using a single cycle replication assay with HIV or SIV in the presence or absence of BCD. BCD was successful in neutralizing both HIV and SIV in vitro. In fact, we could not see outgrowth of virus after 30 days in culture. Based on this data, we administered BCD intravaginally in rhesus macaques, followed by inoculation with highly pathogenic SIV. Control animals were treated with gel alone or nothing before inoculation with the same dose of SIV. The animals have been evaluated for infection by RT-PCR of gag sequences in their plasma and nested PCR of gag sequences in their PBMC. These animals continue to be monitored for antiviral humoral and cellular immune responses. Our in vivo data indicate that intravaginal pretreatment with BCD significantly reduces SIV mucosal transmission relative to untreated control animals. Currently we are performing repeated challenges with BCD and SIV in the uninfected animals to assess whether they continue to be protected from infection. In addition, we are determining whether or not the uninfected BCD-treated animals have antiviral immune responses as compared to the controls. Should BCD continue to prevent SIV transmission and not perturb mucosal tissues in this model, its current approved use in humans suggests it would be an important candidate for clinical consideration for use as an anti-HIV microbicide.

Anthony Antonellis, BS - NHGRI
Glycyl-tRNA Synthetase Mutations in Charcot-Marie-Tooth Disease Type 2D and Distal Spinal Muscular Atrophy Type V
Charcot-Marie-Tooth (CMT) disease and distal spinal muscular atrophy (dSMA) are heterogeneous neuromuscular disorders characterized by distal muscle weakness and wasting. CMT2D and dSMA-V are subtypes of these diseases that show an autosomal dominant mode of inheritance and a more severe phenotype in the upper extremities, specifically in the thenar eminence and first dorsal interosseous muscle groups. The main characteristic that distinguishes these disorders is a distal sensory loss in CMT2D patients. To date three pedigrees for these subtypes have been reported: a North American family with CMT2D, a Bulgarian family with dSMA-V and a Mongolian family with phenotypic features of both CMT2D and dSMA-V. In addition, we have identified a Sephardic Jewish pedigree with dSMA-V. Previous genetic mapping efforts indicated that the responsible gene(s) reside on a ~1.25-Mb segment of human chromosome 7p14. This region contains 11 known genes, none of which have an obvious role in neuronal development or survival. We therefore screened all genes in this region by PCR and direct sequencing in individuals from each family.
Our efforts have led to the identification of four disease-associated missense mutations in the glycyl tRNA synthetase (GARS) gene in CMT2D and dSMA-V families. Specifically, we detected a G240R amino-acid change in the North American pedigree; a L129P amino-acid change in the Bulgarian pedigree; a E71G amino-acid change in the Mongolian pedigree; and a G526R amino-acid change in the Sephardic Jewish pedigree. These mutations segregate with disease status, are not present in appropriate control populations, and change amino acids conserved in species ranging from primates to yeast. GARS is a member of the family of aminoacyl tRNA synthetases responsible for charging tRNAs with their cognate amino acids. Specifically, GARS is involved in attaching glycine to its cognate tRNAs.
These findings represent the first example of a defect in an aminoacyl tRNA synthetase being directly associated with a human genetic disease. While further study is needed to delineate the molecular pathology of these mutations, it is interesting that they lead to a specific peripheral nerve phenotype. Furthermore, the molecular pathology must explain for the severity of the phenotype in the upper extremities. Based on these results, the genes for all aminoacyl tRNA synthetases should be considered relevant candidates for inherited neuropathies and motor neuron diseases.

Ana C Anzulovich, PhD - NIMH
CIRCADIAN, TISSUE AND SEX-SPECIFIC EXPRESSION OF THE CIG30 GENE
The master clock in the vertebrate brain controls slave oscillators throughout the body by delivering both neural and humoral cues about the phase of the cycle. These inputs however, must interact with a host of environmental signals to produce the integrated multi-system rhythmicity necessary for homeostasis. To better understand how these processes are integrated at the molecular level, we searched for links between metabolic and clock regulation by interrogating the circadian liver transcriptome of the transgenic 'fatless' mouse, a model of severe liver decompensation. The circadian regulation of the Cold Inducible Gene (CIG) 30, a very long fatty acid elongase with a role in thermogenesis, emerged out of this analysis as a candidate worthy of further study. We discovered that while circadian expression of CIG30 was abrogated in 'clockless' mice, rhythmic CIG30 expression remained intact in the liver of 'fatless' mice. Thus, the CIG30 promoter could provide a unique window to probe the cross-talk between the circadian and metabolic pathways of gene control. We first performed northern blot analysis to precisely establish the temporal profile of CIG30 gene expression. Unexpectedly, we found that circadian expression of CIG30 was sexually dimorphic since it was detected only in males. To dissect the complex pattern of CIG30 gene regulation, we isolated the CIG30 proximal promoter region and tested its response to various transcription factors. The CIG30 promoter was not stimulated by the diurnal BMAL/CLOCK activator but it was repressed after forced expression of RevErba, a component of the clock machinery which blocks transcription of target genes during the day. Furthermore, SREBP1, a key metabolic regulator of lipogenic genes, was able to activate the CIG30 promoter. Preliminary studies revealed that cleavage-dependent activation of SREBP1 is circadian in the liver, peaking during the dark phase. Taken together, these results can account for the late night peak in CIG30 mRNA expression. Current studies attempt to probe the basis for the sexual dimorphism in CIG30 expression, which could provide a molecular explanation for the known sex differences in thermogenesis. In summary, control of CIG30 gene expression, which is exerted by the clock, sex and feeding related signaling pathways, provides an excellent case study for the integration of multiple regulatory cascades on a specific promoter.

Andre Bafica, MD - NIAID
In vivo Induction of Integrated HIV-1 Expression by Mycobacteria Is Critically Dependent on Toll-Like Receptor 2
Mycobacterial infection has been implicated as a possible factor in AIDS progression in populations where HIV-1 and Mycobacterium tuberculosis are co-endemic. In support of this concept, we have previously demonstrated that HIV-1 transgenic mice (Tg) infected with mycobacteria display enhanced viral gene and protein expression. In vitro and in vivo studies by other investigators have shown that the pattern recognition receptor toll-like receptor 2 (TLR2) is involved in triggering the pro-inflammatory response to mycobacteria. To investigate the role of TLR2 in HIV-1 induction by mycobacteria, we generated HIV-1 transgenic TLR2 deficient (Tg/TLR2-/-) mice by crossing HIV-1 Tg and TLR2 KO animals. We observed that the enhanced HIV-1 gene and protein expression triggered by mycobacteria in this model is crucially dependent on TLR2. Thus, spleen cells from Tg/TLR2-/- mice were found to be completely defective in their HIV-1 p24 response to live M. tuberculosis or Mycobacterium avium as well as mycobacterial products such as monosylated phosphatidylinositol (PIM) and culture filtrate protein (CFP). In contrast, TNF-alpha production was only partially impaired in the same Tg/TLR2-/-cultures. To assess the role of TLR2 in HIV-1 activation in vivo, we infected Tg/TLR2-/- and Tg control animals with M. avium or Toxoplasma gondii (a pathogen known not to trigger TLR2 signaling). Importantly, in vivo mycobacterial infection failed to induce plasma p24 as well as tissue gag and env mRNA expression in Tg/TLR2-/- mice whereas the stimulation of HIV-1 expression by T. gondii was unaltered in the same animals. These observations support the concept that TLR2 is a crucial receptor in the immune activation of HIV-1 by mycobacterial co-infections both in vitro and in vivo and suggest that this pathway may provide a potential target for inhibiting AIDS progression in mycobacteria-infected individuals.

Earle E Bain, M.D. - NIMH
Performance of Unmedicated Depressed Patients on Tasks of Affective and Decision-Making Processes
Abstract Body: Functional neuroimaging studies in depression have shown dysfunction in the amygdala, ventral striatum, anterior cingulate cortex and medial/orbital prefrontal cortex, which suggests that reward and decision-making abilities may be compromised in depression. Most prior neuropsychological studies of depression included medicated subjects. This study investigated performance of unmedicated depressed patients on a battery of neuropsychological tests that included experimental tasks of decision-making and affective processes. Methods: Twelve unipolar and fifteen bipolar unmedicated, currently depressed subjects and twenty healthy control subjects performed a battery of neuropsychological tasks that included tests of memory, attention and executive function from the CANTAB computerized battery and two experimental tasks. The Decision Making task required subjects to place bets based on probabilities that changed from trial to trial. During the Affective Shifting task, subjects were required to press a response bar to positively valenced words during happy, or to negatively valenced words during sad, conditions. These conditions were alternated in a HHSSHHSS pattern to create shift and nonshift response blocks. Results: Depressed and healthy subjects performed equivalently on the intelligence, memory, and executive function tests. In the Decision Making task, unipolar subjects failed more often than healthy subjects to make optimal decisions when odds were equivocal (p=.03 for the 6:4 odds condition) and took longer to make decisions when odds more clearly favored a particular choice (p=.019 in the 9:2 odds condition and p=.015 in 8:2 odds condition). In the Affective Shift task, depressed subjects made more target (omission) errors (p=.02), and more distractor (commission) errors (p=.03) to affectively valenced words than did controls. Additionally, unipolar (p=.01) and bipolar (p=.03) depressed subjects made more distractor errors for sad words than did healthy subjects, but only during the nonshift blocks. Conclusions: The pattern of decision-making displayed by the depressed patients suggests less flexibility in adjusting strategies as risks change. Results from the Affective Shifting task suggest an attentional bias limiting the ability to inhibit responses to affectively valenced stimuli, particularly mood-congruent ones, in major depression.

Bjoern Bauer, PhD - NIEHS
THE NUCLEAR XENOBIOTIC RECEPTOR, PXR, UPREGULATES P-GLYCOPROTEIN AT THE BLOOD-BRAIN BARRIER
The blood-brain barrier not only protects the central nervous system (CNS) from potentially toxic xenobiotics but also limits entry of therapeutics, e.g. anti-cancer drugs, antibiotics and antivirals. An important element of the functional barrier is the multidrug resistance gene product p-glycoprotein (p-gp), an ATP-driven drug efflux pump located at the luminal membrane of the brain capillary endothelium. Recent reports focusing primarily on liver demonstrate transcriptional upregulation of xenobiotic metabolizing enzymes and drug transporter activity (including p-gp) by the pregnane X receptor (PXR in rodents, SXR in humans), a ligand-activated transcription factor.
Here I report that the PXR ligands pregnenolone carbonitrile (PCN) and chenodeoxycholic acid (CDCA) increased p-gp protein level and transport function in rat brain capillaries. Six hour exposure of isolated rat brain capillaries to 1-10 uM PCN or 10-25 uM CDCA significantly increased p-gp protein content of capillary plasma membranes; maximal stimulation was about 2-fold. Consistent with this, immunostaining with 2 different p-gp antibodies demonstrated substantially increased immunoreactivity at the luminal membrane of the capillary endothelial cells. Increased (2-fold) p-gp-mediated transport of a fluorescent cyclosporine A derivative into capillary lumens was also detected using confocal microscopy and image analysis. The increase in transport was abolished when an inhibitor of metabolism (NaCN) or a specific inhibitor of p-gp (PSC833) were added to the transport assay medium. No such increases in p-gp protein content or transport were found when capillaries were exposed to SXR-specific ligands (hyperforin, rifampicin), which have been shown not to activate rodent PXR.
This is the first demonstration that ligand-activated nuclear receptors regulate p-gp activity at the blood-brain barrier. Experiments are currently underway to measure the extent to which PXR activation in vivo alters barrier p-gp function. Because p-gp function is a major determinant of entry into the CNS for many classes of drugs, these findings have important clinical implications. First, since PXR is activated by a large number of drugs and metabolites, patients receiving multidrug therapies may exhibit increased blood-brain barrier function. Second, it is not clear to what extent blocking PXR will downregulate p-gp. Thus, PXR and SXR antagonists may provide a means to selectively and transiently open the barrier.

Laura L Baxter, Ph.D. - NHGRI
A whole genome mutagenesis screen for neural crest stem cell genes
We are performing a whole genome mutagenesis screen to discover genes important in neural crest development. N-ethyl-N-nitrosourea (ENU) is a potent mutagen that causes random point mutations in mouse spermatogonia, thus allowing the generation of heritable mutations. ENU mutagenesis can be targeted towards revealing additional genes in specific disease/developmental pathways by combining the ENU with a sensitization screen, where mutations are assessed on a strain background already destabilized by a known mutation. This is a powerful combination since a mutation that alone may not have a phenotype in the heterozygous state will be revealed when present along with another heterozygous mutation.
Our sensitized neural crest mutagenesis screen uses mice with a mutation in the transcription factor SOX10. SOX10 mutants exhibit developmental neural crest defects due to apoptosis of neural crest stem cells and abnormal transcription of genes required for cell fate specification. Heterozygous SOX10 mutants demonstrate the partial neural crest defects of hypopigmentation (fewer melanocytes) and megacolon (reduced enteric ganglia). We have shown the feasibility of using SOX10 heterozygotes for a sensitized screen, as synergistic enhancement of neural crest defects occurs when a Sox10 mutation (Sox10LacZ) occurs as a double heterozygote with a subset of known neural crest mutant loci. To identify new genes, the offspring of mutagenized mice and Sox10LacZ/+ mice are screened for a synergistic increase in hypopigmentation and megacolon. ENU mutations are linked to BALB/cJ alleles, as the mutagenized males are from the BALB/cJ inbred strain and Sox10LacZ/+ mice are from different strains. A panel of 44 genome-wide PCR markers with unique BALB/cJ polymorphisms will rapidly identify mutation map locations. To date this screen has identified two new mutant mice with enhanced hypopigmentation in combination with Sox10LacZ/+, and the genomic location of these mutations is being identified.
Through this screen, new loci that are important in neural crest development will be uncovered, and new alleles of previously identified genes will be generated for analysis of gene function. The mutagenized mice from this screen will also be used by other investigators, to create a frozen sperm resource and for an embryonic screen to identify genes important in early neural crest migration and survival.

Gil Ben-Menachem, Ph.D. - NICHD
A Newly Discovered Cholesteryl Galactoside from Borrelia Burgdorferi
Lyme disease, caused by the spirochete Borrelia burgdorferi, is recognized as a cause of common and serious systemic disease worldwide. It affects the skin, joints, nervous system, and heart, and in its chronic form, can result in autoimmune manifestations such as arthritis. Lyme disease is the number one vector borne disease, and about 20,000 new cases are reported each year in the US. Little information exists about the protective antigen/s or the host factor/s that confer immunity to B. burgdorferi. The licensed vaccine, composed of a derivative of an outer membrane protein, has been withdrawn by the manufacturer. Our initial aim was to isolate LPS from B. burgdorferi, for a potential vaccine use. Although LPS has been identified in several spirochaetales, such as Leptospira and Treponema, we could not find evidence for its presence in B. burgdorferi, nor were we able to detect markers of LPS such as KDO, Lipid A or 3-hydroxy fatty acids. However, we isolated and characterized, two major glycolipids from B. burgdorferi, designated as BbGL-I and BbGL-II. These compounds were purified by silica gel chromatography, and analyzed using GLC-MS, MALDI-TOF, FAB-MS, NMR spectrometry, and metabolic labeling. The structure of BbGL-I was determined as cholesteryl 6-O-acyl-beta-D-galactopyranoside, and that of BbGL-II as 1,2-di-O-acyl-3-O-alfa-D-galactopyranosyl-sn-glycerol. This is the first demonstration of a cholesteryl galactoside in bacteria. Surprisingly, despite their small size, BbGL-I and BbGL-II elicited antibodies in mice and rabbits, mostly of the IgM isotype. Antibodies to these glycolipids have also been found in sera of Lyme disease patients. The lack of LPS, the abundance of BbGL-I and BbGL-II in the bacterial membrane, their surface exposure (as indicated by fluorescence labeling), and their three-domain structure, lead us to suggest that these glycolipids may assume LPS function.

Tamar Ben-Yosef, PhD - NIDCD
Claudin 14: A Cation-Selective Tight Junction Protein Necessary for Cochlear Neurosensory Hair Cell Survival
Tight junctions (TJs) selectively modulate paracellular permeability between extracellular compartments and are also involved in maintaining cellular polarity. In the cochlear duct TJs are thought to separate the K+-rich endolymph from the Na+-rich perilymph. Recessive mutations of CLDN14, the human gene encoding claudin 14 TJ protein, cause profound, congenital deafness DFNB29, thus demonstrating the importance of claudin 14 for hearing. To explore the role of claudin 14 in the inner ear and in other tissues we created a mouse model for DFNB29 by a targeted deletion of Cldn14. In the targeted allele a lacZ cassette replaces the only Cldn14 coding exon, and is expressed under the Cldn14 promoter, thus serving as a reporter gene for Cldn14 expression. In Cldn14-lacZ heterozygous mice b-galactosidase activity was detected in cochlear inner and outer hair cells and supporting cells, in the vestibular sensory epithelium, in the collecting ducts of the kidney, and around the lobules of the liver. Cldn14-null homozygous mice are viable and fertile, with normal vestibular function. Auditory brainstem response analyses indicated that four weeks old Cldn14-/- mice have profound hearing loss, with thresholds elevated by 50 dB-SPL or more over all frequencies tested, compared to their heterozygous and wild type litter mates. Scanning electron microscopy, immunohistochemistry and X-gal staining revealed that during the second and third weeks of life of Cldn14-/- mice, cochlear hair cells undergo rapid degeneration, which progresses from the base to the apex of the cochlea. Monolayers of MDCK cells expressing claudin 14 show a 6-fold increase in the transepithelial electrical resistance by decreasing paracellular permeability for cations. In wild type mice, claudin 14 was immunolocalized at hair cell and supporting cell TJs. Our data suggest that the TJ complex at the apex of the reticular lamina requires claudin 14 as a cation-selective barrier to maintain the proper ionic composition of the fluid surrounding the basolateral surface of OHCs.

cyril berthet, PhD - NCI-CCR
CDK2 KNOCKOUT MICE ARE UNEXPECTEDLY VIABLE BUT STERILE AND DISPLAY GROWTH DEFECTS
The functions and the regulation of the cyclin-dependent kinase 2 (Cdk2) have been studied in vitro and in cell lines. Proposed functions of Cdk2 include promotion of entry into S-phase of cell cycle, initiation and maintenance of DNA replication, exit from S-phase/entry into G2 phase and degradation of the cell cycle inhibitor p27.
Cdk2 has been considered to be rate limiting for cell cycle progression in mammalian cells and an essential gene because expression of dominant negative Cdk2 mutant in human cell lines leads to cell cycle arrest. However recent results show that this mutant does not affect the proliferation of specific colon cancer cell lines. It appears to be essential to study Cdk2 functions in the context of a living animal to determine its contribution to the control of cell cycle.
Here, we report the generation and characterization of Cdk2 knockout mice carrying a deletion of the kinase domain. Unexpectedly, homozygote Cdk2-deficient mice are viable, indicating that Cdk2 is apparently not essential for cell viability. However, these mice are smaller than the wild-type and both male and female knockout mice are sterile, showing severe atrophy of the gonads. Moreover, a partial lethality of homozygote embryos is also observed. These observations raise the possibility that Cdk2 might be crucial for meiosis and important for a normal growth rate, though dispensable in somatic cells.
Our analysis of these growth defects focus now on embryonic development, meiotic arrest and organs implicated in immune response. These questions are addressed in multiple tissues using an in vivo assay of BrdU incorporation. Further studies will ask the questions of tumor formation and regulation of the cell cycle without Cdk2 activity and look for possible overlapping mechanisms. In order to learn more about the genetic interactions of Cdk2 with other cell cycle genes we crossed Cdk2-/- mice with other mutant mice. We have generated Cdk2+/-; Cdk4+/- and Cdk2+/-; p27-/- mice and expected soon to obtain the double homozygotes. This will help to investigate the functional overlap between these genes.
The functional overlap of the different Cdks is one of the important open questions in order to decide if inhibition of a single Cdk is sufficient for cell cycle arrest in an animal. Results could influence the decision for developing drugs specific for a single Cdk or for several Cdk family members at the same time.

Sven Bilke, PhD - NCI-CCR
Cancer Genome Investigations using Comparative Genomic Hybridization on cDNA Microarrays
Genomic instabilities, including gain or loss of specific DNA regions, are commonly observed in tumors. It is generally believed that these alterations are at least partially responsible for oncogenesis and contribute to the phenotype of the tumor. Regions commonly changed in tumors may therefore harbor genes important for a specific phenotype. The precise localization of genomic changes and the identification of pathways leading to these alterations is therefore a key problem. Comparative genomic hybridization (CGH) based on cDNA arrays is a powerful tool that measures the relative copy number of cell DNA for several thousand genomic locations. While the spatial resolution of this method is far superior to more traditional approaches like metaphase CGH or most BAC arrays, the cDNA data contains somewhat more noise. We have developed an algorithm to improve the signal to noise ratio while retaining much of the spatial resolution. The basic idea is the integration of topological information in the data analysis: observing increased (or decreased) copy numbers at several adjacent DNA locations is a highly significant indication of amplification (or deletion), even if the individual signals are within noise. This algorithm was used to analyze cDNA based CGH data for 50 neuroblastoma tumors, one of the most frequent childhood cancers. Based on approximately 20,000 data-points, the effective spatial resolution in this dataset for the entire human genome is of the order 1.5 MBp. We identified known as well as so far undocumented genomic alterations. The information about altered regions was subsequently used to train a neural network on clinical stages of the samples. We show that the fingerprint of genomic alterations is sufficient to classify the tumors. This finding strongly supports the hypothesis that the driving mechanism behind tumor progression is the genomic changes. We further analyzed interactions between regions of genomic imbalance. One example for such an interaction is unbalanced gene transfer: an extra copy of a DNA segment A substitutes (and removes) a segment B at a different genomic location. Using correlation as wel as Bayesian -based approaches we identified several such interactions, some of which were novel and some that were previously published. These combined approaches of CGH on cDNA microarrays with the statistical techniques described provide powerful tools for the exploration of genomic imbalance in cancer.

Marita Bosticardo, Ph.D - NHGRI
Reconstitution of the IL-12 signaling pathway in IL-12 receptor beta-1 (IL-12Rb1) KO mice upon transplantation of retrovirally transduced hematopoietic stem cells
The functionality of the IL-12-mediated signaling pathway is critical for the elimination of intracellular pathogens. Patients carrying genetic defects in the IL-12 signaling pathway (IL-12 p40 or IL-12Rb1) indeed show increased susceptibility to weakly pathogenic strains of mycobacteria and salmonella. Infections can be treated by administration of IFN-g and antibiotics; however, reversion of patients' susceptibility by corrective gene transfer could be beneficial. We were able to restore the IL-12-mediated signaling pathway in PHA-activated T cells blasts of IL-12Rb1 deficient patients through retroviral- mediated gene correction. In addition, to test the feasibility and safety of retroviral-mediated gene correction in vivo, we established a murine model of gene therapy in IL-12Rb1-deficient mice. Lineage-negative cells isolated from bone marrow of IL-12Rb1-deficient mice were cultured for 5 days and transduced using a retroviral vector carrying the murine IL-12Rb1 cDNA. This resulted in a high percentage of IL-12Rb1+ cells (50-100%). Gene-corrected bone marrow cells were transplanted into lethally irradiated recipient mice, which were then analyzed at 8-16 weeks after the transplant. Treated mice did not show adverse effects upon gene therapy. We detected IL-12Rb1 expression in peripheral blood lymphocytes, splenocytes and thymocytes from the majority of treated mice. Moreover, splenocytes isolated from mice transplanted with gene-corrected bone marrow cells acquired the ability to respond to IL-12, as demonstrated by the phosphorylation of STAT4 and the production of IFN-g, which are completely impaired in IL-12Rb1 KO mice. These results have been confirmed in in vivo experiments of IFN-g production upon challenging with LPS from Salmonella Enteriditis. However, membrane expression of the IL-12Rb1 chain as well as the production of IFN-g in response to IL-12 or LPS where lower than those detected in wild type control mice. Additional experiments are needed to establish if the degree of gene correction reached in our experimental model is sufficient to overcome the immunodeficiency of IL-12Rb1 KO mice.

Susan J Boyd, M.D. - NIDA
Correlation between brain mu-opiate binding and cocaine use by cocaine users in treatment.
Most prior research on the neuropathophysiology of cocaine abuse has focused on the dopamine transporter. This laboratory is investigating the important mu-opiate inputs to the brain dopamine system. Prior Positron Emission Tomography (PET) studies from this laboratory have shown up-regulation of the endogenous mu-opioid receptor (mOR) in non-treatment-seeking cocaine users. Preliminary data in non-treatment seekers suggests mu-opiate binding may correlate with relapse to cocaine use following a period of enforced abstinence. This is the first study to correlate brain mu-opiate binding by PET imaging and treatment outcome in cocaine users in treatment. Methods: PET scans were obtained using [11C]-carfentanil, a selective mu-opioid receptor agonist, on admission of 12 subjects to outpatient treatment for cocaine abuse or dependence. Observed urine samples were obtained from subjects three times per week while they were in treatment and on the day of PET scanning. Subjects were largely male (67%), half were non-white (all but one African-American), with a mean age of 37.9 (SD = 5.2) years. Parametric images for binding potential (BP) of mOR were derived from 90- minute PET scan. Correlation between regional mOR binding and urine data was assessed by voxel-by-voxel analysis using SPM99. Results: Increased mOR binding in the left anterior cingulate gyrus was positively correlated with percent cocaine-positive urines while in treatment (r=0.82, p = 0.001). An inverse correlation between increased mOR binding and cocaine-positive urines was found in the right temporal cortex (r = -0.91 , p = 0.001). Including the cocaine urine status on the day of the PET scan as a covariate did not change the results. Conclusion: These preliminary results suggest a relationship between brain mu-opiate receptor density and treatment outcome. This influence may be mediated by interactions between brain dopamine and endogenous opioid systems. Further research on this relationship may open opportunities for improved treatment of cocaine addiction and prediction of treatment response.

Sabine K Breun, MS - NCI-CCR
IDENTIFICATION OF DC-SIGN SEQUENCES NECESSARY FOR HIV-1 TRANSMISSION AFTER VIRUS CAPTURE
Of the three DC-SIGN family molecules expressed in primates, only human DC-SIGN has been tested for HIV-1 transmission properties using primary cells. Because DC-SIGN mediated HIV-1 transmission varies depending on the cell type DC-SIGN is expressed on, we decided to more carefully examine a DC-SIGN-related molecule, human L-SIGN (DC-SIGNR, CD209L1), in cell systems that are permissive for DC-SIGN transmission function. L-SIGN is expressed in liver and lymph node endothelial cells and has also recently been identified as an attachment receptor for hepatitis C virus (HCV). Strikingly, we found that cells expressing human or chimpanzee DC-SIGN are 100-fold more efficient in HIV-1 transmission than cells expressing comparable levels of human or chimpanzee L-SIGN. Because L-SIGN is able to capture HIV-1 particles as efficiently as DC-SIGN, this DC-SIGN-specific effect in HIV-1 transmission occurs after virus binding. Although DC-SIGN and L-SIGN share 85 % sequence identity, there are notable differences in cytoplasmic and extracellular regions of the molecules. We thus constructed DC-SIGN and L-SIGN chimeric molecules and examined the chimeras in virus transmission assays. These experiments allowed us to define an extracellular region in DC-SIGN that is required for HIV-1 transmission. Curiously, these required amino acid residues would not be predicted to make direct contact with bound HIV-1 and, thus, possibly impact DC-SIGN structure, interaction with other molecules, or trafficking. Identification of this domain will likely help elucidate post-virus binding steps in the DC-SIGN-mediated transmission of HIV-1. It may also aid in the development of antiretrovirals to impair potential DC-SIGN-mediated mucosal transmission of HIV-1. Significantly, our results also question the role that L-SIGN may play in HIV-1 pathogenesis.

Elizabeth E Brown, MPH PhD - NCI-DCEG
Determinants of HHV8 viremia in the general population, Italy
Background: Human Herpesvirus 8 (HHV8) viremia is predictive of Kaposi's sarcoma (KS) development in patients with HIV-associated or iatrogenic immunosuppression. However, determinants of viremia in HHV8-infected members of the general population remain undefined. Methods: We assessed correlates of HHV8-DNA detected in peripheral blood mononuclear cells (PBMC) from 158 HHV8 antibody positive adults (males, 65%; median age, 75 years) from Sicily, Rome and Naples. We determined HHV8 antibody positivity to latent nuclear antigen (LANA) by immunofluorescence assay (IFA) and quantified HHV-8 DNA by TaqMan real-time PCR for the K6 gene. Hematologic parameters were determined with an automated cell counter. Analyses were adjusted for age, gender and geographic region by logistic regression to control for potential confounding. Results: Twenty-six (17%) subjects had detectable HHV8-DNA (median copies per million cells, 53; range, 13-2128). The presence of viremia did not vary by gender, age or geographic region. Adjusted for the potential confounders, HHV8 DNA was detected more frequently in participants who had more than 7 total residents in their childhood home (OR=3.7 [1.5-9.1]), more than 2 younger siblings (OR=2.6 [1.0-6.5]), current cardiovascular (OR=3.6 [1.3-9.7]) or renal disease (OR=3.1 [1.2-8.0]) and, among women, more than 1 lifetime sex partner (OR=29 [2.2-388]). Excluding 21 participants (4 cases and 17 controls) using corticosteroids or cancer therapeutics, HHV8 viremia was significantly more frequent among those with mild thrombocytopenia (<157 K/uL; OR=6.2 [2.0-19.1]) and mild normocytic anemia (<4.4 million erythrocytes per uL; OR=5.6 [1.9-16.8]). Conclusion: HHV8 viremia in elderly Italians is associated with childhood crowding, mild depression of RBC count and thrombocytopenia, perhaps indicative of early life infection and chronic inflammation. These risk factors are the first to be reported for HHV8 viremia in the general population. Additional studies are required to validate these relationships and determine their etiological significance.

David L Bryce, PhD - NIDDK
Use of Correlated Residual Dipolar Coupling Data in NMR Spectroscopy for the Determination of the Magnitude of the Molecular Alignment Tensor of Oriented Biopolymers in the Absence of Structural Information
Nuclear magnetic resonance (NMR) spectroscopy is the technique of choice for studying the structure and dynamics of small and moderately sized globular proteins and nucleic acids in solution. Traditional structure determination by NMR has primarily relied upon the measurement and interpretation of semiquantitative local restraints in isotropic solutions. Recently, it has been shown that by introducing a small degree of order into the orientational distribution of the biomolecules, e.g., by dissolving them in a dilute liquid crystalline solvent, residual anisotropic interactions which are lost in isotropic solutions, may be recovered. The most important of these interactions is the residual dipolar coupling (RDC) between pairs of nuclear spins. The RDCs act as novel and extremely sensitive reporters on local and long range molecular geometry, due to their dependence on internuclear distances and bond orientations within the biomolecular alignment tensor frame. To fully benefit from the RDC data, the degree of alignment of the biomolecules must be quantified. This is often accomplished by examining a histogram of the RDCs, which demands that a sufficiently large number of well-distributed RDCs be available. Such is not always the case, for example in selectively isotopically enriched samples, in some largely helical proteins, or in larger systems where slower molecular tumbling may reduce the number of accurate measurements which may be made. Here, we show that due to the known fixed geometry of certain molecular fragments, e.g., the peptide plane in proteins or the purine plane in DNA, the RDCs within such fragments are correlated. Equations describing this correlation have been developed on the basis of the symmetry properties of the dipolar coupling tensor. From the equations, an essentially infinite number of synthetic RDCs may be generated on the basis of limited correlated data, thereby offering the potential to drastically improve the accuracy of the alignment tensor measurement. The method has been tested on simulated and experimental data sets for several proteins including ubiquitin, GB1, and GB3. We conclude that by taking advantage of the correlated nature of RDC measurements, the degree of biomolecular alignment may be accurately quantified on the basis of relatively little experimental data. This work represents a step towards extending the applicability and improving the accuracy of biomolecular structure determination.

Silvia Buervenich, PhD - NIMH
High-throughput genetic analysis of nuclear receptor genes in bipolar affective disorder
Nuclear receptors (NR) are a superfamily of intracellular transcription factors that directly regulate gene expression. Animal studies have shown that defects in genes encoding NRs lead to observable developmental or behavioral disturbances, but often only when more than one gene in the NR superfamily is disrupted, consistent with the high redundancy typically observed for critical genetic pathways. For example, the co-occurrence of induced mutations in several retinoic-X receptor (RXR) genes leads to developmental disturbances in mice. Three NR genes map to established regions of genetic linkage to bipolar affective disorder (BPAD), most notably NR2E1 on chromosome 6q, recently linked to BPAD at a genome-wide significant level. To investigate a possible role for NR genes in BPAD, we have undertaken a study of 27 validated single-nucleotide polymorphisms (SNPs) that occur in genes encoding 3 NRs of the RXR family and 2 orphan NRs, NR2E1 and NR4A2. The overall experimental design involves a hypothesis-generating stage, where SNPs are genotyped in a sample of 150 cases and 150 matched controls, followed by a hypothesis-testing stage, where promising associations are tested in an independent sample of 400 probands and family-based controls. Multi-locus analysis allows testing of individual genes as well as interaction between genes. Promising association findings have so far been detected for individual SNPs in RXRgamma and NR2E1. Perhaps the most interesting finding so far is an association involving 2 SNPs in NR2E1 and one in RXRbeta , significant at the p<0.001 level in the case-control sample. If verified in the larger sample of trios, this finding would directly implicate these NR genes in BPAD, potentially opening new avenues of diagnosis and treatment.

Maja A Bumke, Dr. sc. nat. - NCI-CCR
Transcriptomic and proteomic analyses of gene expression in fibroblasts
In this work, gene expression in fibroblasts has been studied by proteomic and transcriptomic techniques, with particular attention to the modifications induced by extracellular pH changes. The pH of tissues varies during development, in physiological processes and in disease (e.g. cancer). However, the influence of extracellular pH on gene expression has been studied only for a limited number of genes. Primary fibroblasts were chosen as a model system, since these cells adapt their intracellular pH on the basis of the pH of the culture medium, and due to the fact, that fibroblasts are responsible for the synthesis of a large portion of extracellular matrix (ECM) components in health and disease. Normal human dermal fibroblasts were cultured at two different pH values (pH 6.7 and 7.5). The expression of secreted proteins, as well as the expression of mRNA transcripts were studied. The aim was to identify components of the ECM, which are differentially modulated by extracellular pH and might be aberrantly expressed in a tumoral environment, and to get a deeper insight in the regulation of the transcriptional program as a response to a pH change. For the proteomic analysis of the fibroblast cell culture model, two-dimensional gel electrophoresis (2D PAGE) was performed, and for protein visualization several staining and imaging techniques were compared. For the identification of protein spots from 2D gels, different analyses were performed: N-terminal Edman sequencing, Peptide Mass Fingerprinting using MALDI-TOF mass spectrometry, as well as microcapillary liquid chromatography - tandem mass spectrometry (µLC-MS/MS).
The transcriptional program of fibroblasts at the two different pH values was studied by quantitative RT-PCR and Northern blotting, and a genome-wide analysis was performed using Affymetrix GeneChip technology. Among the proteins found to be differentially expressed by fibroblasts at the two pH conditions, the ones with the strongest differences were agrin and tetranectin, and e.g. extracellular matrix protein ECM-2 and stanniocalcin STC-1. These and other upregulated proteins might be interesting candidates for further studies. The relevance to physiological and pathological processes of these findings will have to be confirmed, e.g. performing antibody selection followed by immunohistochemical analysis of healthy and diseased tissue samples, to evaluate their eventual differential expression.

Hongxue Cai, Ph.D - NIDCD
Motion analysis in the hemicochlea
The relative motions of cochlear structures remain unclear despite advances in imaging the vibrational patterns in the cochlea. Our goal is to develop a new, fast and efficient geometry-based approach to analyze the micromechanics in the cochlear duct. Optical flow techniques are often used to estimate velocity fields to represent motion in successive video images. Usually the method is mathematically ill-posed, because the single scalar equation representing the conservation of local intensity contains more than one unknown velocity component. Instead of regularizing the problem using optimization techniques, we formulate a well-posed problem for the gerbil hemicochlea preparation by introducing an in-plane incompressibility constraint, and then show that local brightness is also conserved. We solve the resulting system using a Lagrangian description of the conservation equations. With this approach, the displacement of isointensity contours on sequential images determines the normal component of velocity of an area element, while the tangential component is computed from the local constant area constraint. We have validated our method using pairs of images generated from our calculations of the vibrational deformation in a cross section of the organ of Corti and tectorial membrane in the mammalian cochlea, and quantified the superior performance of our method when complex artificial motion is applied to a noisy image obtained from the hemicochlea preparation. Our new method avoids the calculation of image spatiotemporal derivatives. It also avoids both the well known aperture problem and the need for high contrast edges. Finally, we applied our Lagrangian method to pairs of video images from the hemicochlea preparations. Our results show that the organ of Corti transforms basilar membrane motion into a shearing stimulus at the apices of outer hair cells. The relative shear movement is mainly induced by the radial motion of the reticular lamina, with little contribution from the tectorial membrane. Furthermore, the deformations within the organ of Corti and tectorial membrane are very complex, and we find a significant radial motion of the inner hair cell body, which is caused by the rotation of the tunnel of Corti about the foot plate of the inner pillar cell.

Frances S Calderon, Ph D - NIAAA
Docosahexaenoic acid promotes differentiation in hippocampal neurons
Docosahexanoic Acid (22:6n-3;DHA) is an essential fatty acid particularly enriched in brain. DHA specifically accumulates in brain during prenatal and postnatal periods paralleling important events in neurogenesis and synaptogenesis. Moreover, dietary n-3 fatty acid deficiency during this stage is associated with an impairment of learning and memory tasks, strongly suggesting that DHA may have an important role in neuronal development.
Previous studies in this laboratory have shown that DHA can increase the content of cellular phosphatidylserine enhancing signal transduction processes involving Raf and the PI3Kinase/Akt pathway, thus promoting neuronal survival. Differentiation processes could also be modulated trough these mechanisms.
This study is aimed to evaluate the possible effect of DHA on neuronal differentiation. Since the hippocampus is one of the brain structures mediating learning and memory functions, we have used hippocampal primary cultures obtained from E18 rats; cultures were grown in defined medium with 0 or 1.5uM DHA. As control we examined the effect of arachidonic acid (20:4n-6;AA), oleic acid (18:1n-9;OA) or docosapentaenoic acid (22:5n-6;DPA). DPA is an important control because in n-3 fatty acid deficiency situation, there is a dramatic decrease of DHA compensated by DPA, the closest n-6 fatty acid analogue of DHA. Neuronal differentiation was evaluated after 3-6 days in vitro (DIV) measuring neurite length with the NIH Image Software on neurons positively stained with MAP2, a somatodendritic marker. The results show that DHA uniquely increased the total neurite length/neuron. The observed DHA effect was specific as OA, AA or DPA did not have any effect on neurite extension. At 5 and 6 DIV DHA increased the total neurite length by increasing the individual neurite length and the number of branches/neuron. Moreover, hippocampal cells from animals depleted of DHA by feeding animals with n-3 fatty acid diet have shorter neurites and lower total neurite length than hippocampal cells from animals fed with n-3 fatty acid adequate diet. Our data are the first demonstration that DHA has a remarkable effect on morphological differentiation in hippocampal neurons, possibly providing a basis for the essential role of DHA in development and functions associated with learning and memory.

Sophie Candon, MD, PhD - NIAID
A Th2 TRANSGENIC MODEL OF AUTOIMMUNE GASTRITIS
Upon activation, CD4+ T lymphocytes differentiate into Th1 or Th2 cells. The Th1 subset produces interferon-gamma (IFN-g) and tumor necrosis factor-beta (TNF-b), while Th2 cells secrete interleukin (IL)-4, -5, -10, and -13. Th1 cells are thought to be the effectors in autoimmune diseases, while Th2 cells might be suppressive. Autoimmune gastritis is induced by thymectomy of BALB/c mice on day 3 of life. AIG is mediated by CD4+ T cells and seems to be mostly Th1-mediated. We previously isolated from a thymectomized mouse, an I-Ad-restricted T cell clone, TXA51, specific for residues (889-900) of the gastric H/K ATPase alpha-chain. TXA51 has a Th2 phenotype and is pathogenic upon transfer to nu/nu mice. We now have generated transgenic mice expressing the TCR from the TXA51 clone (A51 mice). The low proportions of CD4+ single positive thymocytes together with reduced frequency of CD4+CD8+ double positive thymocytes expressing high levels, as compared to control mice, suggest that transgenic thymocytes are inefficiently selected. Nevertheless, mature transgenic T cells, that reproduce the peptide specificity of the original TXA51 T cell clone, are found in the periphery. 59% of A51 mice spontaneously develop gastritis by 10 weeks of age. The disease is characterized by eosinophilic infiltration of the gastric mucosa and by high serum levels of IgE. A51 T cells are activated in the gastric lymph node and undergo Th2 differentiation in vivo since, upon in vitro restimulation under cytokine neutralizing conditions, they produce IL-4 but not IFN-g. Thus, the transgenic TCR conveys in vivo the Th2 phenotype of the TXA51 T cell clone and its pathogenicity. The marked differences in disease incidence and severity observed between the three A51 mouse lines generated correlate with differences in the level of expression of the transgenic TCR chains and the frequency of endogenous TCR alpha rearrangements. Whereas incidence and severity of disease might be influenced by the frequency of pathogenic T cells, the Th2 phenotype of the autoreactive response in vivo seem to be related to peculiaritities in the presentation of the target epitope and/or in intrinsic affinity of the TXA51 TCR for its MHC class II/peptide ligand. Further studies of these new transgenic lines should provide insight into the mechanisms involved in Th1/Th2 differentiation in the context of autoimmunity.

Matilde Canelles, PhD - NIAID
The influence of the thymic environment on the CD4 versus CD8 T lineage decision
During development in the thymus, T cell precursors expressing CD4 and CD8 coreceptors (double positive, or DP) are positively selected to mature based on the ability of the T cell receptor (TCR) to interact with peptide-MHC complexes on thymic stromal cells. TCR specificity is intimately related to lineage commitment since DP thymocytes expressing TCRs that recognize class II MHC mature as CD4+CD8- (CD4 single positive, SP) T cells, whereas those expressing a TCR specific for MHC class I mature as CD4-CD8+ (CD8 SP) T cells. How TCR-MHC recognition is related to the CD4/CD8 decision has been an issue of much interest. Most studies of CD4 versus CD8 lineage commitment have focused on intracellular signaling events during thymic selection with little attention given to the role of the microenvironment, localization, topography, and migration. How and when thymocytes interact with each other and with thymic stromal cells during positive selection could have a significant impact on cell fate. In normal mice, relatively few precursors are selected for differentiation and migration from the cortex to the inner medullary region, where newly generated SP thymocytes appear. In fact, the proportion of precursors able to undergo selection may be crucial for normal thymic architecture, since the thymus of TCR transgenic mice shows more medullary regions than normal. In an attempt to examine development of thymocytes bearing transgenic TCR in a more normal thymic environment, we generated mixed hematopoietic stem cell chimeras to limit the number of thymocytes with selectable TCRs. We find that decreasing the frequency of selectable precursors generally improves positive selection. Moreover, decreasing the ratio of selectable to non-selectable precursors also affects lineage commitment. Whereas precursors with a class II-selectable TCR usually generate CD4 SP thymocytes in H-2b hosts, dilution of selectable precursors with precursors unable to undergo positive selection causes a significant fraction of class II-restricted thymocytes to mature as CD8 SP thymocytes. Examination of chimeric thymuses by confocal microscopy reveals that CD4 and CD8 SP can be segregated into distinct medullary regions. Of note, CD8 SP thymocytes predominate in medullas where surrounding cortical areas are high in non-selectable precursors. These results indicate that a thymic microenvironment high in non-selectable precursors favors CD8 development, irrespective of MHC specificity.

Jennifer L Cannons, PhD - NHGRI
The Role of SAP in the Regulation of Th2 Differentiation
X-linked Proliferative disease (XLP) is a human genetic disorder characterized by immune dysregulation with hyperactivated T cells, low antibody production and malignant lymphoma often triggered by EBV infection. XLP is associated with mutations that affect SAP (SLAM-associated protein), which interacts with the cytoplasmic tail of receptors related to SLAM (signaling lymphocyte activation molecule). SLAM has been reported to increase expression of the Th1 cytokine IFN-gamma. To determine whether alterations in cytokine production and differentiation contribute to the phenotype of XLP, SAP-deficient mice were generated. SAP-deficient mice exhibit features of XLP post-infection, including T cell hyperactivation and impaired B cell responses that are secondary to intrinsic T cell defects. Furthermore, following TCR engagement, we have found that SAP-deficient T cells exhibit a skewing towards a Th1 phenotype (increased IFN-gamma) with a dramatic defect in production of Th2 cytokines (IL-4). Analysis of SAPxIFN-gamma doubly deficient mice indicates that the Th2 defect is independent of Th1 skewing. SAP-deficient T cells show an impaired induction of the Th2 transcription factor GATA-3, whereas other Th2 transcription factors including NFAT and STAT6 as well as the Th1 inducing transcription factor T-bet appear to be expressed normally. Evidence from Das et al suggest that NF-kappaB p50 is critical for Th2 differentiation and GATA-3 expression. Following TCR stimulation, we find that SAP-deficient T cells demonstrate impaired IkappaB-alpha degradation as well as an imbalance of p50/p50 to p50/p65 dimers compared to wildtype T cells. In contrast, TCR-induced Ca++ mobilization and Erk activation are normal. Thus, SAP-deficient T cells appear to have a specific defect in GATA-3 induction associated with abnormal NF-kappaB activation.
To evaluate the effect of expression of wildtype SAP in T cells, CD2-SAP transgenic mice were generated. Consistent with the results obtained from the SAP-deficient T cells, anti-CD3 stimulation of CD2-SAP transgenic T cells demonstrate an increase in IL-4 production with a decline in IFN-gamma production. We are investigating the activation and expression of transcription factors from CD2-SAP T cells compared to SAP-deficient T cells. Thus, the expression of SAP plays a critical role in the regulation of Th cell differentiation: SAP-deficiency skews to a Th1 profile while overexpression of SAP biases toward Th2 differentiation.

Liu Cao, Ph.D. - NIDDK
Inactivation of Brca1 sensitizes to oxidative stress and compromises organ homeostasis
The breast cancer associated gene1 (BRCA1) forms a genome surveillance complex with multiple DNA damage repair proteins, and plays a central role in genomic stability and tumorigenesis. Recently we reported that loss of Brca1 full-length isoform (Brca1Delta11/Delta11) caused p53-mediated aging and tumorigenesis in animals. This finding further extends the genomic instability as an initiating factor in aging. Alternation in organism homeostasis is known as an important pathogenic factor for accelerating aging and cancer susceptibility. Stem/progenitor cells in somatic tissues govern homeostasis, while self-renew and differentiation of these cells is precisely regulated by genetic and environmental factors to avoid ageing and neoplasm in renewable organs. To further investigate the mechanisms underlying Brca1-mutation associated aging and tumorigenesis, we studied behavior of stem/progenitor cells in Brca1 mutant mice. The mutant mice showed less proliferating stem cells in multiple organs. In skin wound healing assay, we found that the mutant mice had less increase of proliferating stem cells in wound area, leading to decreased rate of wound healing. In newborn keratinocyte culture, mutant cells showed reduced proliferation and differentiation. On the other hand, the aging population of mutant mice (up to 6 months of age) displayed metaplasia, and hyperkeratosis in esophagus, and dysplasia and papilloma in forestomach, which were accelerated by the absence of p53. Molecular and histologic analyses revealed expansions of keratin 10 and 14 positive layers, overexpression cyclin D1, and b-catenin. As oxidative stress may have an important role in the pathogenesis of aging and carcinogenesis, we further explored the effect of oxidative stress in Brca1 mutant mice. We found that Brca1 mutant mice were significantly more sensitive to Paraquat induced lethality than controls. At cellular level, mutant cells accumulated high level of reactive oxygen species, and were hypersensitive to oxidative damage induced apoptosis (by H2O2 and Paraquat). These observations provide a molecular basis for the premature aging and tumorigenesis observed in Brca1-deficient mice. In sum, our results suggested that Brca1 prevents oxidative damage in renewable organ homeostasis. It links the environmental and genetic factors in carcinogenesis and aging, provides new insights into genomic injury in organism maintenance and tumorigenesis.

Emilia Caputo, Ph.D. - NIMH
Analysis of CD4 and FN interaction with GCDFP-15/gp17 by ProteinChip Technology
Gross cystic disease fluid protein (GCDFP-15), also known as prolactin-inducible protein (PIP) is a specific breast tumor marker. GCDFP-15/PIP is also identified as gp17 and/or seminal actin-binding protein (SABP) from seminal vesicles; and as extra-parotid glycoprotein (EP-GP) from salivary glands. It is an aspartyl-proteinase with specific fibronectin (FN)-degrading ability, suggesting its potential role in mammary tumor progression and fertilization. Other functions have been attributed to it based on its ability to interact with CD4, actin, and FN, although these are still under investigation. We demonstrated that GCDFP-15 and gp17 expressed in pathological and physiological tissues, respectively, showed different structural properties. This suggested that depending on its conformational state it could differently bind to these molecules and change its function(s).
We investigated the interaction of GCDFP-15 and gp17 with CD4 and FN, two of the well-known binding-partner molecules. We used protein chip technology, a biochemical approach based on two powerful techniques, chromatography (on protein chip surfaces) and mass spectrometry (on Surface-Enhanced Laser Desorption Ionization Time of Flight mass spectrometer). We created specific CD4 and FN surfaces, with CD4 and FN covalently attached. GCDFP-15 and gp17 were incubated on these surfaces and the species able to bind to CD4 or FN were directly detected by SELDI-TOF MS. We found that gp17 was mainly involved in the binding to CD4 as compared with its pathological counterpart and we identified the specific FN and CD4 binding-domains on GCDFP-15/gp17. This may provide a rationale for the effective modulation of its function(s).

Adela R Cardones, MD - NCI-CCR
CXCR4 enhances adhesion of B16 tumor cells to endothelial cells in vitro and in vivo via beta 1 integrin
The chemokine receptor, CXCR4, is expressed by human melanomas and its ligand, CXCL12, is frequently produced at sites of melanoma metastasis. We have previously shown that CXCR4-transduced B16 melanoma cells exhibited markedly increased pulmonary metastasis two weeks after tail vein inoculation of C57/B6 mice, and that one of the possible mechanisms of enhanced metastasis was increased adhesion to endothelial cells. Herein, we examine CXCR4-enhanced binding of B16 murine melanoma cells to endothelial cells (EC) and recombinant adhesion molecules in vitro in order to determine the role of tumor- and EC-derived adhesion molecules in tumor metastasis. By flow cytometry, unstimulated primary lung EC showed constitutive expression of VCAM-1 whereas skin-derived EC did not. All B16 cell lines tested showed constitutive expression of beta 1 (CD29), but not beta 2, integrins. Under physiologic shear stress conditions (1.5 dynes/cm2), CXCR4-B16 cells bound to lung EC at lea!
st two-fold more efficiently than to skin-derived EC (P=0.002). CXCR4-B16 arrest on VCAM-1-Ig-coated plates and TNF-a-stimulated EC was rapid, resistant to high shear stress, and showed no evidence of rolling prior to arrest. In vitro, CXCR4-B16 cell binding to EC was blocked by anti-CD29 and anti-CXCL12 mAb. In vivo, metastasis of CXCR4-B16 cells to murine lungs was strongly inhibited by anti-CXCL12 and two different anti-CD29 mAb, suggesting that EC-associated CXCL12 was required for efficient arrest of tumor cells. Finally, CXCR4-B16 exposed to CXCL12 rapidly increased binding affinity for soluble VCAM-1-Ig as detected by a flow cytometric assay. Although it is known that beta 1 integrins are involved in melanoma metastasis, our work demonstrates beta 1 integrins are regulated by chemokine receptors. Thus, beta 1 integrins play a critical role in CXCR4-mediated B16 tumor cell metastasis in vivo and may be a potential target for inhibition of tumor metastasis, particularly to the lung.

Michelle A Carey, PhD - NIEHS
Attenuated Immune Response and Enhanced Mortality following Influenza Virus Infection in Cyclooxygenase-2 Null Mice
Influenza is a significant cause of morbidity/mortality worldwide despite extensive research and vaccine availability. Immunocompromised individuals are particularly susceptible. Cyclooxygenase (COX)-1 and COX-2 metabolize arachidonic acid to prostaglandins which have profound immunomodulatory properties. Non-steroidal anti-inflammatory drugs (NSAIDs) inhibit COX and are widely used. In recent years, selective COX-2 inhibitors have been extensively marketed. The objective of our study was to investigate the role of COX-1 and COX-2 on the host immune response to influenza viral infection. Mice (Wt, COX-1 null, COX-2 null) were infected intranasally with 200 pfu of Hong Kong influenza A (H3N2) virus. Control mice were sham infected. Body weight and temperature were measured daily as indicators of morbidity. Groups of animals were sacrificed on days 1, 4 and 6 post-infection and bronchoalveolar lavage fluid (BALF) was analyzed for the presence of inflammatory cells, and cytokines important in proinflammatory and anti-viral responses. Wt and COX-1 null mice, but not COX-2 null mice, developed significant hyperthermia on day 1 following infection (p<0.05). Weight loss was evident in COX-1 null and Wt mice on day 1, whereas COX-2 null mice showed no significant weight loss until day 3. Thereafter, the rate of weight decline in the COX-2 null mice rapidly accelerated and by day 5 exceeded that in Wt animals (18% vs 13%, p<0.05). Fifty percent of COX-2 null mice died between days 5-6 whereas there was no mortality in Wt or COX-1 null animals (p<0.05). Levels of the proinflammatory cytokines TNF-alpha and IL-1beta (day 6) and of the anti-viral cytokine IFN-gamma (day 6) were dramatically reduced in BALF from COX-2 null mice relative to Wt and COX-1 null mice (TNF-alpha: 32 vs 72 and 54 pg/ml; IL-1-beta: 8 vs 31 and 17 pg/ml; IFN-gamma: 454 vs 1039 and 1010 pg/ml, p<0.05). Recruitment of neutrophils and macrophages to the airways of infected mice was markedly attenuated in COX-2 null mice compared to Wt and COX-1 null mice on day 4 (neutrophils: 4.9X10E5 vs 6.9X10E5 and 13.9X10E5, macrophages: 7.4X10E5 vs 10.2X10E5 and 11.0X10E5, p<0.05). In conclusion, lack of hyperthermia, retarded early weight loss, reduced BALF proinflammatory and antiviral cytokines, decreased lung inflammatory cell recruitment and enhanced mortality in the COX-2 null mice suggests that COX-2 products are important in mediating the acute phase host response to influenza viral infection.

Guibin Chen, MD - NHLBI
Differential Gene Expression Profile of Hematopoietic Progenitor Cells from Paroxysmal Nocturnal Hemoglobinuria (PNH) Patients
Paroxysmal nocturnal hemoglobinuria (PNH) is a clonal hematopoietic stem cell disorder in which an acquired PIG-A gene mutation leads to deficient bioassembly of glycophosphatidylinositol (GPI) and failure to express GPI-anchored proteins (GPI-AP) on the surface of blood cells. Despite elucidation of the biochemical and molecular defects in PNH, the pathophysiology of clonal expansion of defective cells, especially in the setting of immune-mediated marrow failure, remains unexplained. In pursuit of evidence of intrinsic differences between GPI-AP-deficient and Á¡normalÁ± cells, we determined gene expression profiles in CD34 BMCs from healthy donors and PNH patients, using the Affymetrix GeneChip Hu95KA version 2 arrays. Individual PNH patientsÁø CD34 BMCs were sorted into two populations: GPI-AP-Á¡normalÁ± and GPI-AP-deficient by flow cytomety. RNA was extracted and pooled before synthesized of cDNA. Two cycles of cDNA synthesis were utilized due to one did not have sufficient cRNA for micoarray analysis even after pooled. Data analysis was performed using Affymetrix and GeneSpring software. In parallel with functional tissue culture studies, the GPI-AP-Á¡normalÁ± cells from PNH patients showed up-regulation of genes involved in apoptosis and the immune response, including APO-1, TRAIL, TRAIL receptor, TNF, and interferon-gamma. Cell cycle inhibitory genes (p21 and p57) also were increased, but genes associated with anti-apoptosis function (BAX and DFF45) and hematopoietic proliferation and differentiation (Flt-3 and STK-1) were decreased in these cells. In contrast, the GPI-AP-deficient cells, representing the PNH clone, appeared more similar to CD34 cells obtained from healthy individuals. We have chosen ten genes using real-time RT-PCR and flow cytometer to verify the findings of the microarray in individuals of eight PNH patients. The result corresponded to the findings of the GeneChip. Our results first confirm at a molecular level that Á¡normalÁ± cells in the PNH BM undergo apoptosis preferentially compared to GPI-AP-deficient cells. These results implicate an environmental influence on hematopoietic cell proliferation, in which the PNH clone evades immune attack and destruction, while normal cells suffer a stress response followed by programmed cell death.

Xiaoli Chen, Ph.D. - NIDDK
Characterization of the secretory proteome of adipose cells using a global proteomics approach
The critical endocrine function of adipose cells has been recognized. It has been demonstrated that adipose cells play a central role in the pathogenesis of obesity and its related diseases by secreting signaling molecules. Dysregulation of adipose cell secretion may be the initial defect responsible for the development of disorders in other tissues/organs. However, the signaling cascades are not well understood and cannot be completely explained with the currently known adipose-derived factors. The purpose of this study was to identify the proteome of secretory proteins from adipose cells using a 2D-LC-MS/MS approach. Stromal-vascular (S-V) and adipose cells from the epididymal, omental, and inguinal fat depots were isolated from rats at age 5-6 weeks, and cultured in a serum-free DMEM in the presence or absence of insulin for 48 h. Culture medium was concentrated and secreted proteins were separated on RP-HPLC, followed by LC-MS/MS using a QTOF 2 mass spectrometer. Data were analyzed using a MASCOT search engine. In addition, identity as a secretory protein was validated by searching for the presence of a N-terminal signal peptide. From comparison to S-V cells, we could clearly show that mature adipose cells function as secretory cells, and a large number of secreted proteins were found. The profile of secreted proteins varies among the three depots examined. To 15 proteins that have already been shown to be secreted by adipose cells, we have compiled a group of at least 40 known secretory proteins previously not associated with adipose cells, and another group of novel proteins. Sequence prediction of signal peptide and transmembrane domains demonstrates that among the identified proteins at least 15 are novel potential endocrine factors secreted by adipose cells. Examination by RT-PCR of adipose depot-specific expression of selected proteins shows that CD14, CD59 antigen, epididymal secretory protein, and adrenomedullin are expressed in adipose cells from all three depots, while we found one potential endocrine factor is specifically expressed in the epididymal adipose depot. Comparison with the microarray results from a study of human adipose cells from insulin resistant patients demonstrates that mRNA expression of more than half of the identified secretory proteins is up- or down-regulated in insulin-resistant states. Our findings lead to an improved understanding of the role of adipose cells in pathophysiological states.

XIU CHEN, Ph.D - NCI-CCR
Cyclin D-Cdk4 and Cyclin E-Cdk2 Regulate the JAK/STAT Signal Transduction Pathway in Drosophila
The JAK/STAT signal transduction pathway regulates many developmental processes in Drosophila. However, the functional mechanism of this pathway is poorly understood. We identify the Drosophila cyclin-dependent kinase 4 (Cdk4), which exhibits embryonic mutant phenotypes identical to those in the Hopscotch/JAK kinase and stat92E/STAT mutations. Specific genetic interactions between Cdk4 and hop mutations suggest that Cdk4 functions downstream of the HOP tyrosine kinase. We further show that Cyclin D-Cdk4 (as well as Cyclin E-Cdk2) binds and regulates STAT92E protein stability. STAT92E regulates gene expression for various biological processes, including the endocycle S phase. These data suggest that Cyclin D-Cdk4 and Cyclin E-Cdk2 play more versatile roles in Drosophila development.

Connie Cheung, PhD - NCI-CCR
Role of human PPARalpha in the induction of peroxisome proliferation in PPARalpha-deficient mouse liver
Fibrates are hypolipidemic drugs that are widely used to lower plasma lipid concentrations, particularly in correcting the dyslipidemia that accompanies type II diabetes and so reduce the risk of developing heart disease. They exert their effects by activating peroxisome proliferator-activated receptor alpha (PPARalpha), which regulates the transcription of a number of genes involved in hepatic lipid metabolism. Sustained activation of PPARalpha leads to the development of liver tumors in rats and mice, however humans are refractory to induction of peroxisome proliferation. It is hypothesized that humans are resistant to the adverse effects of synthetic PPARalpha ligands due to low levels of hepatic expression of PPARalpha compared to rodent species. To examine the mechanism of species differences in response to PPARalpha ligands, humanized mice were generated expressing human PPARalpha in a mouse PPARalpha knockout background. Mice expressing human PPARalpha were produced using cDNA expression with a Tet-OFF system in which expression of the transgene can be regulated by exposure to doxycycline. Northern and western blot analyses revealed hPPARalpha mRNA and protein expression in the liver, which was at a level comparable to mouse PPARalpha wildtype mice. Administration of doxycycline (dox) in the drinking water reduced this level in a dose dependent manner; 2mug/ml dox completely suppressed hepatic hPPARalpha expression. Mice were treated for 2 weeks with the PPARalpha ligands Wy-14,643 and fenofibrate. The humanized PPARalpha mice showed marked induction of a number of peroxisomal, microsomal and mitochondrial fatty acid oxidation enzymes, hepatomegaly and lowering of serum triglycerides in a manner similar to that of mouse PPARalpha wildtype mice, albeit to a lower extent. However, these effects were reduced with increasing dox exposure. These data establish that in this mouse model, human PPARalpha is functional and can induce peroxisome proliferation within the liver. Moreover, the levels of this receptor appear to be important in determining the species - specific response to PPARalpha ligands. Further analyses of this mouse model, such as long-term carcinogenicity studies with synthetic PPARalpha ligands, will contribute to understanding the mechanism of action of peroxisome proliferators and the species differences in biological activity and carcinogenicity mediated by them.

Thierry Cheutin, ph D - NCI-CCR
Mechanisms of maintenance of heterochromatin in yeast and mammals
Epigenetic modifications are a fundamental mechanism for control of gene expression. Recently, the tri-methylation of histone H3 on the K9 residue has been correlated with gene inactivation and has been shown to increase the affinity of heterochromatin protein 1 (HP1) for histone H3. HP1 is involved in epigenetic silencing and has been proposed to be a structural protein stabilizing heterochromatin. Epigenetic silencing through histone H3 methylation of K9 is an evolutionary conserved mechanism present from S. pombe to human. By performing dynamic studies on living mammalian cells, we demonstrate that HP1 is highly mobile inside stable heterochromatin domains. To compare the behavior of HP1 between mammalian cells and yeast, we have now investigated the mobility of swi6, the yeast homolog of HP1. In contrast to mammalian cells, the heterochromatin domains are highly mobile in the yeast nucleus. Regardless, fluorescence recovery after photobleaching indicated that swi6 is as highly mobile as HP1. The mobility of swi6 as well as the mobility of heterochromatin domains depends on the physiological state of the yeast. The mobility of swi6 was not affected by loss of clr1, clr2, clr3, but was strongly dependent on clr4 and rik1. Taken together, these results demonstrate that swi6, as HP1, only transiently interacts with heterochromatin in vivo, and support a stochastic model of HP1/swi6 binding to heterochromatin. These observations provide novel insights into mechanisms of heterochromatin maintenance.

Frederick T Chin, PhD - NIMH
PET IMAGING OF MOUSE BRAIN KINETICS OF [18F]FEM-IMPY, A PROSPECTIVE RADIOLIGAND FOR b-AMYLOID - EFFECT OF DEUTERATION
The b-amyloid radioligand, [125I]6-iodo-2-(4'-N,N-dimethylamino)phenyl-imidazo[1,2-a]pyridine ([125I]IMPY), has high uptake into normal mouse brain (7.2% dose/g; peak at 2-5 min) and fast clearance (< 0.35% dose/g at 2 h after i.v. injection). Brain radioactivity in transgenic mice (Tg2576) bearing brain amyloid deposits is 3.3 times higher at 4 h after injection than in age-matched controls. Since [125I]IMPY has such favorable radioligand behavior, we have investigated two fluorine-18 (t_=110 min) labeled analogs of IMPY, namely [18F]2-fluoroethyl-IMPY ([18F]FEM-IMPY) and [18F]2-(fluoro-(D4-ethyl))-IMPY (D4[18F]FEM-IMPY), as potential radioligands for PET, initially focussing on measurement of the kinetics of these radioligands in normal mouse brain. FEM-IMPY has moderately high affinity (KD = 27 ± 8 nM) for Ab-aggregates in vitro. The uptake of radioactivity into mouse brain after intravenous injection of each radioligand was high (~ 6% ID/g at ~ 0.5-1.2 min) and similar to [125I]IMPY (7.2% dose/g at 2 min). However, in contrast to the known single exponential washout of [125I]IMPY, the clearance of brain radioactivity from these radioligands was bi-phasic, being rapid over the first 20 min and then very slow. Residual brain radioactivity for [18F]FEM-IMPY was 4.3% ID/g and significantly lower for D4[18F]FEM-IMPY (3.2% ID/g; p < 0.01) at 1.5 h after injection. In the PET experiments with [18F]FEM-IMPY there was substantial bone uptake of radioactivity (8.2% ID/g at 1.5 h after injection) presumed to be [18F]fluoride. D4[18F]FEM-IMPY gave significantly lower uptake of radioactivity (5.2% ID/g; p < 0.01). This is the first observation of a deuterium isotope effect on the biodistribution of an [18F]N-fluoroethylated compound and is being investigated further. The methodology developed here will be applicable to evaluation of these radioligands by PET in transgenic amyloid bearing mice, and to further 18F-labeled b-amyloid radioligands that we are developing with a view to avoiding rapid defluoridation and residual radioactivity in brain. Species effects on kinetics and metabolism are also to be investigated.

Lina S Correa-Cerro, MD, PhD - NICHD
Characterization of a hypomorphic Smith-Lemli-Opitz syndrome mouse model and its use in testing therapeutic interventions
Smith-Lemli-Opitz syndrome (SLOS) is an autosomal recessive, multiple malformation syndrome due to an inborn error of cholesterol synthesis. Mutations in the gene encoding the 7-dehydrocholesterol reductase (DHCR7), the last enzyme of the cholesterol biosynthetic pathway, cause SLOS. Consistent with the role of DHCR7 in reducing 7-dehydrocholesterol (DHC) to cholesterol, SLOS patients typically have reduced levels of cholesterol and increased 7-DHC levels. Clinical findings in SLOS include facial anomalies, limb defects including syndactyly of the second and third toes, abnormal behavioral, and failure to thrive. The 278C>T (T93M) is the most common missense mutation. Mouse models with a null disruption (delta) have been found to be lethal. To investigate therapeutic interventions in SLOS, we generated a hypomorphic SLOS mouse model. Using targeted homologous recombination, we inserted a point mutation into Dhcr7 to produce a T93M missense allele. T93M/T93M mice were viable and phenotypically normal. However the compound heterozygote, T93M/Delta had 2-3 toes syndactyly. Of note, this is the most frequently observed physical finding in SLOS patients. Sterol profiles were obtained from a set of tissues from one day and 6 weeks old mice. We observed high levels of 7-DHC and low cholesterol level in both genotypes. 7-DHC levels were higher in the T93M/Delta compared to T93M/T93M mice consistent with the more severe phenotype. Cholesterol supplementation has been used to treat human SLOS patients; however, therapeutic efficiency has not been completely established. Thus, we are studying the effect of dietary cholesterol supplementation on the biochemical and behavioral phenotype of T93M/Delta mice. Preliminary results, using the vertical pole and hanging wire tests to evaluate motor coordination, suggest that cholesterol supplementation may be beneficial. Pathological studies of mice on cholesterol supplementation versus regular diet, did not show major differences. This data supports the long-term safety of dietary cholesterol supplementation in SLOS.
We are currently performing the biochemistry analysis of sterol profiles in different tissues from mice under cholesterol supplemented versus the regular diet in both mutants and controls to investigate for correction of the biochemical abnormality.
Thus, the T93M/Delta hypomorphic mouse is a good model system to study the effects of pharmacological interventions on abnormal sterols in SLOS.

Diana Cozma, M.D - NICHD
Lathosterolosis, an Inborn Error of Murine Cholesterol Synthesis with Multiple Defects in Gastrointestinal Tract
Inborn errors of cholesterol synthesis represent an important group of human malformations syndromes. The prototypical example is the Smith-Lemli-Opitz syndrome (SLOS) which is due to a deficiency of 3 beta-hydroxysterol reductase (DHCR7) activity. Other syndromes in this group include desmosterolosis, CHILD syndrome, CDPX2 and lathosterolosis.
Lathosterol 5-desaturase (Sc5d) catalyzes the conversion of lathosterol to yield 7-dehydrocholesterol (7DHC). In the subsequent enzymatic step, 7DHC is reduced by 7-dehydroxysterol reductase (DHCR7) to yield cholesterol.
We have generated a mouse model of lathosterolosis by targeted disruption of the murine Sc5d gene. In part, these efforts were initiated to examine whether some of the phenotypic features of SLOS are due to a lack of cholesterol or due to increased 7DHC. Sc5D -/- pups were stillborn, had craniofacial defects including cleft palate and micrognathia, and limb patterning defects. Tissue lathosterol levels were increased, whereas cholesterol levels were decreased. Histopathological evaluation of 18.5 day embryos revealed multiple gastrointestinal defects. In Sc5d -/- embryos the stomach lumen contained multiple sloughed cells, intestinal hypoplasia was observed, and a distinct cloaca was identified in 3/5 mutant embryos. TUNEL staining of the glandular stomach showed marked apoptosis in mutant embryos.
Many malformations found in Sc5d -/- are consistent with impaired hedgehog signaling. Sonic Hedgehog (Shh) signaling has been shown to be important for gastrointestinal tract development and our previous work has shown that Shh signaling is impaired in Sc5d -/- fibroblasts. To address Shh signaling we characterized the expression of Shh and its receptor Patched (Ptc) in the lathosterolosis mouse stomachs by whole amount in situ hybridization. No difference in expression Shh or Ptc at E=12.5 and E=14.5 days was detected. However expression of both Shh and Ptc were decreased at E=16.5 and E=18.5 days in Sc5d -/- stomachs embryos. This data suggests that impaired Shh signaling may underlie the gastric malformation present in the lathosterolosis mice.

Alain Dabdoub, PhD - NIDCD
Protein kinase C regulates inner hair cell numbers in the mammalian cochlea
The determination of specific cellular phenotypes is a crucial step during development especially within structures that are comprised of highly ordered patterns. The sensory epithelium of the mammalian cochlea is comprised of a highly ordered cellular pattern, including a single row of inner hair cells and three rows of outer hair cells that extend along the basal-to-apical axis of the cochlea. These mechanosensory hair cells are responsible for sound perception and their cellular pattern represents one of the most highly ordered structures in any vertebrate system; however, the factors that regulate the precise specification and number of cellular phenotype in the cochlea are unknown.
Major aspects of development, proliferation, and differentiation are regulated by the phosphorylation states of proteins through the action of protein kinases. Consequently, one factor playing crucial roles in cellular signal transduction and a common target of several ubiquitous transduction cascades is protein kinase C (PKC), a family of phospholipid-dependent serine/threonine kinases that phosphorylate a variety of target proteins. To determine the role of PKC during the development of the cochlea, PKC was either inhibited or activated in embryonic day 13 mouse explant cochleae in vitro. PKC inhibition, by either of two highly selective inhibitors, significantly increased the number of cells that developed as inner hair cells and resulted in large regions of supernumerary inner hair cells (up to 8 rows instead of one row). A control compound for PKC inhibition had no effect thus establishing the specificity of the result. To confirm that the additional cells were in fact hair cells, the level of expression of mRNA for the hair cell specific gene Math1 was quantified for control and PKC inhibited cultures by real-time quantitative PCR. Math1 expression increased after 24 and 48 hrs of PKC inhibition.
Activation of PKC with either of two distinct and potent activators resulted in a decrease in the number of inner hair cells. Furthermore, sensitivity to PKC was temporally limited; application of PKC activators or inhibitors on E15 or later had no effect on hair cell number.
These results provide the first link between PKC modulation and determination of inner hair cells in the cochlea and demonstrate that the number of cells with the potential to develop as inner hair cells is significantly greater than the number of cells that normally develop.

Kamal Datta, MD - CC
Molecular Mapping of a Key Lethal DNA Lesion Produced by Radiation: A Structurally Complex Radiation-Induced DNA Double-Strand Break.
Radiation therapy is one of the most common and successful methods used to treat many cancers. The killing effect of radiation is attributed to DNA double-strand break (DSB) production in genomic DNA. A range of structural complexity is predicted for radiation-induced DSBs, and DSB lethality is a function of structural complexity. Thus, knowledge of complex DSB structure is needed to understand the biochemical mechanisms of radiation lethality and DSB repair, and to possibly manipulate these events to enhance radiotherapy effectiveness. However, to date, no radiation-induced DSB has ever been isolated and analyzed at the molecular level. To address this problem we created a model site-specific complex DSB in a plasmid by triplex-forming oligonucleotide-targeted 125-I decay, in the presence and absence of DMSO, a radical scavenger. A 32 bp fragment upstream of the DSB was isolated and probed with the E. coli DNA repair enzymes, endonuclease IV (endoIV), endonuclease III (endoIII), and formamidopyramidine-DNA glycosylase (Fpg), which recognize apurinic/apyrimidinic (AP) sites, and pyrimidine and purine base damage respectively. Fragments were analyzed by phosphoimagery and densitometry following resolution in 20 percent denaturing PAGE. Both strands of the damaged fragment were analyzed, as were an internal control fragment isolated from a position 1000 bp from the break site and an undamaged control fragment isolated from undamaged plasmid. No significant difference in DSB yield was observed with, or without DMSO. EndoIV treatment of the upper strand indicated AP sites clustered within 3 bases downstream and 7 bases upstream of the targeted base. Repeated experiments consistently detected an AP site 4 bases upstream of the 125I target base. EndoIV analysis of the lower strand showed similar AP site clustering. In contrast, endoIII and Fpg demonstrated base damage clustering near the end opposite the break site. No enzyme sensitivity was observed with the controls. GC/MS identified base damage as 8-hydroguanine, 8-hydroxyadenine, and 5-hydroxycytosine. In contrast to DSB yield, base damage yield and the enzyme sensitivity pattern were dependent upon DMSO, with a greater yield in its absence. In conclusion we show AP sites associated with complex 125-I-induced DSB are clustered near the break site, while base damage is clustered away from the DSB. Also, radical scavenging does not affect DSB yield, but it does alter the yield of associated damage.

Gloria L David, PhD - NIEHS
NQO1 and GSTM1 Polymorphisms and Childhood Asthma in a High Ozone Area: Mexico City
Asthma, a chronic inflammatory disorder of the airways, is the most common chronic disease of childhood in the US. The prevalence of asthma in the US has increased steadily over the past two decades, especially among children. The interaction of genetic and environmental factors is important in determining asthma susceptibility. Exposure to ozone, a potent oxidant, has been clearly related to asthma severity, and recent evidence suggests that chronic exposure to ambient ozone may lead to the development of asthma in adults and children. Response to ozone varies between individuals, and the pattern of variability is consistent with genetic susceptibility (high inter-individual variability combined with low intra-individual variability). Ozone imposes an oxidative burden on the lung; therefore, polymorphisms in genes involved in response to oxidative stress, such as antioxidant enzymes, may play a role in susceptibility to asthma. NAD(P)H: quinone oxidoreductase (NQO1) and glutathione S-transferase mu (GSTM1) are phase II enzymes important in response to oxidative stress, such as occurs during exposure to ozone. We examined the relationship between functionally significant polymorphisms in NQO1 (Pro187Ser) and GSTM1 (homozygous deletion) and asthma risk in children with a high lifetime exposure to ozone. We enrolled asthmatic children from the allergy referral clinic at a public pediatric hospital in Mexico City, along with their parents. Children and parents provided a blood sample as a source of DNA. We assayed for the Pro187Ser polymorphism in NQO1 using a PCR-RFLP assay and for the presence of GSTM1 by PCR. We used a log-linear model to analyze data from 218 available triads. We did not find strong evidence of an association between NQO1 genotype alone and asthma risk. However, a significant reduction in asthma risk was observed among individuals carrying at least one Ser allele for NQO1 who also had homozygous deletion of GSTM1 [Relative Risk (RR) = 0.4, 95% CI 0.2-0.8]. The P value for difference in RR for NQO1 by GSTM1 genotype = 0.013. These data are consistent with a protective effect of the NQO1 Ser allele in GSTM1-null children with high ozone exposure. Our results emphasize the importance of evaluating gene-gene and gene-environment interactions when studying susceptibility to complex diseases, such as asthma.

Valeria de Mello-Coelho, Ph.D. - NIA
Molecular and Histological Alterations within the Aging Thymus: Evidence for preadipocyte-like cells as a novel thymic microenvironmental component
Age-associated thymic involution is characterized by a progressive reduction of the thymic weight and volume, fat deposition and diminished thymopoiesis resulting in a systemic decline in T cell immunity. While several theories have been forwarded regarding the mechanisms of age-associated thymic atrophy, little is known about the alterations in gene expression that occur within the thymus during aging. In an effort to define such changes, total RNA was obtained from the thymi of mice of varying ages and hybridized with a cDNA microarray containing 1,152 known genes. Approximately 80 genes were upregulated more than three folds while 4 genes were down regulated in the thymi of 4, 6, 12 and 18 month-old mice when compared with 2 month-old animals. The majority of these genes encompass proteins associated with cellular activation, differentiation, metabolism, and interestingly, adipocyte differentiation. The increased expression of BDNF, NGF-R, SOCS-2 and LIF noted in the microarray analyses was further substantiated using real time RT-PCR and immunohistology. Immunohistological analysis of the aging thymi revealed that proteins corresponding to these genes are expressed by thymic stromal cells and/or thymocytes. In addition, an age-associated hyperexpression of BDNF, NGF-R as well as LIF, which is known to promote thymic atrophy, was detected in a unique multivacuolated cell population in the septa adjacent to mature adipocytes and within the thymic parenchyma. These novel multivacuolated cells exhibited a phenotype similar to adipocytes undergoing differentiation based on their morphology, specific staining with Oil Red O, and co-expression of PPAR-gamma, resistin and the fibroblast marker ER-TR7. Moreover, these preadipocyte-like cells were found to increase in number with progressive aging. Overall, these findings demonstrate that significant changes in gene expression occur within the thymus with age. The inverse correlation between the loss of thymocytes and increased presence of preadipocyte-like cells with advancing age suggests that adipogenic cells populations may not simply be innocent bystanders but actual participants in the thymic involution process.

Angelo DelParigi, M.D. - NIDDK
Reduction in hypothalamic gray matter in obese individuals with and without Prader-Willi Syndrome
Prader-Willi syndrome (PWS) is a genetic disorder characterized by hyperphagia and massive obesity as well as short stature, muscular hypotonia, and cognitive impairment. Most patients have growth hormone deficiency and hypogonadotropic hypogonadism, suggesting hypothalamic-pituitary dysfunction. However, very little is known about the hypothalamus in PWS. To test the hypothesis that hypothalamic abnormalities are present in PWS, we used voxel-based morphometry (VBM) of T1-weighted magnetic resonance images to compare 7 male subjects with genetically confirmed PWS [age 26 +/- 7y, percent body fat (DEXA) 36 +/- 11, mean +/- SD] and 18 healthy male subjects (controls, 29 +/- 9y, 22 +/- 12 percent). Groups did not differ in age, but PWS were fatter than controls (p=0.02). VBM analysis was performed using statistical parametric mapping (SPM99, Welcome Department of Cognitive Neurology, London, UK) with procedures optimized to remove artifacts related to non-brain tissue. Brain scans were transformed into coordinates of a standard brain atlas [Talairach J and Tournoux P, 1988] and segmented into gray matter images. A probability map of group differences in regional gray matter concentration was then generated. PWS subjects showed a lower gray matter concentration than controls in the hypothalamic region (peak-voxel, x=2, y=-4, z=-14; z-score=5.68, p=0.002, corrected for multiple comparisons over the whole brain volume). This difference remained significant after adjusting for percent body fat (F-value=14.25, D.F.=21, p=0.001). Average gray matter concentration in a 6mm-radius sphere centered in the hypothalamus (x=0, y=0, z=-11) was negatively related to adiposity in controls (r=-0.64, p=0.004), but not in PWS (p=0.99). Here we report the first in-vivo evidence of a structural abnormality of the hypothalamus in PWS. Our finding is consistent with a report of a loss of oxytocin-containing neurons in the paraventricular nucleus of the hypothalamus in 5 PWS brain donors (Swaab DF et al., 1995). An unexpected result of our study was the negative correlation between gray matter hypothalamic concentration and adiposity. Ablation of hypothalamic neurons has been used to induce weight gain in animals, but no evidence of a structurally abnormal hypothalamus has been reported in obesity of unknown origin. In conclusion, our data indicate that obesity is associated with low hypothalamic gray matter, a defect that is especially pronounced in individuals with PWS.

Bonnie J Deroo, PhD - NIEHS
Estrogen treatment reduces expression of thioredoxin interacting protein in the mouse uterus through an estrogen-receptor independent mechanism
Thioredoxin (TXN) is a multifunctional protein involved in the redox regulation of many essential cellular functions and processes. The reducing activity of TXN is negatively regulated by Thioredoxin Interacting Protein (TXNIP). Mice carrying a TXNIP mutation are predisposed to coronary artery disease, and TXNIP has been shown to suppress metastasis in human melanoma. The thioredoxin system also plays a critical role in reproductive processes and implantation. While the upregulation of TXN expression by estrogen in the female reproductive tract has been previously documented, the role of TXNIP has not been described and its context in estrogen action is not understood. During microarray studies to identify novel estrogen-regulated genes in the mouse uterus, we observed that, in contrast to upregulation of TXN mRNA levels, an acute dose of estrogen rapidly reduced the expression of TXNIP. These results were confirmed by reverse transcription, real-time PCR, and Northern Blot. Estrogen reduced TXNIP mRNA levels by 86% after two hours, and this level of repression was maintained up to 12h. To ascertain if the downregulation of TXNIP required the estrogen receptor (ER), mice were co-treated with the ER-antagonist, ICI 182,780. ICI was unable to block TXNIP repression by estrogen, suggesting that the ER is not required for repression of this gene. In support of this hypothesis, repression of TXNIP was investigated in vitro in an endometrial carcinoma cell line lacking ER alpha. Estrogen reduced TXNIP mRNA levels by 30-40% as early as thirty minutes after treatment in the absence of ER alpha. We also determined the levels of thioredoxin 2 (TXN2) and thioredoxin reductase 1 (TXNR1) in the mouse uterus after estrogen treatment. The estrogen-mediated reduction in TXNIP mRNA correlated with an increase in both TXN2 and TXNR1, and ICI blocked this increase. Thus, we have identified a novel rapid ER-independent response in the mammalian uterus, and have shown that estrogen regulation of three key members of the thioredoxin family may involve both ER-alpha dependent and independent mechanisms. These data support an important role for estrogen action in regulating the redox state of the uterine tissue signaling components during the initial stages of estrogen action in the mammalian uterus.

Daniel P Dickstein, MD - NIMH
Neuropsychological Performance in Pediatric Bipolar Disorder
Background: Considerable controversy exists about whether pediatric and adult onset Bipolar Disorder (BPD) are different phenotypes of the same illness. While considerable work has focused on the diagnosis, less work examines the pathophysiology of pediatric BPD. We used a neuropsychological battery previously used in adults with BPD to examine neuropsychological function in children with BPD.
Method: The Cambridge Neuropsychological Test Automated Battery (CANTAB) is a computerized battery designed to evaluate several cognitive domains, including visual attention and memory. Our pediatric BPD sample met strict DSM-IV criteria for BPD, including one manic episode with grandiosity lasting at least 4 days. Controls were matched for age and gender.
Results: Pediatric BPD and control groups each consisted of 21 children (15 boys and 6 girls; mean age 12 years in both groups). Our sample was euthymic with Young Mania Rating Scale-Parent mean of 9.19 +/- 8.14. Children's Depression Rating Scale-Parent mean was 27.19 + 9.42, and -Child was 23.67 +/- 7.85. Children's Depression Inventory mean was 9 +/- 8.47. Groups did not differ by IQ. All patients were taking at least 1 psychotropic medication, and 57% met criteria for co-morbid Attention Deficit Hyperactivity Disorder (ADHD). CANTAB results showed significant impairment (p<0.05) in pediatric BPD subjects in comparison to controls on measures of attentional set-shifting (Intra-Dimensional/Extra-Dimensional shift) and visuospatial working memory (Pattern Recognition Memory and Spatial Span). BPD performance was not significantly different from controls on the remainder of the CANTAB tests including: Motor Screening, Spatial Working Memory, Stockings of Cambridge, and Spatial Recognition Memory. Further analysis showed no significant correlations existed between neuropsychological performance and mania ratings, co-morbid ADHD, or psychotropic medications.
Conclusions: CANTAB data presented here in pediatric BPD fit well within the broader framework of known neurocognitive deficits in adults BPD. Neuropsychological performance in our pediatric BPD sample is more similar to that in adult BPD than to that in childhood ADHD, with deficiencies in attentional set-shifting and visuospatial working memory. In concert with previous lesion studies in adults and imaging studies of adult BPD, our results suggest altered PFC function in pediatric BPD, especially orbitofrontal cortex.

Antony S Dimitrov, Ph.D. - NCI-CCR
The Role of the Fusion Peptide and Membrane-proximal Domain in HIV-1 Envelope Glycoprotein-mediated Membrane Fusion
The N-terminal fusion peptide and the interfacial sequence preceding the transmembrane anchor of HIV-1 gp41 are required for viral fusion. Studies with synthetic peptides indicated that these regions function by destabilizing membranes, which is regarded as a crucial step in the membrane fusion reaction. However, it is not clear whether membrane destabilization is induced by these sequences in the intact gp41. We address this question by examining fusion and destabilization of membranes expressing HIV-1 IIIB wild type Env and its following mutant Envs: 1) the Glu residue at position 2 of gp41 fusion peptide is substituted for Val (V2E); 2) the sequence DELTA 665-682 in the membrane-proximal domain is deleted. The process of membrane destabilization was monitored by the influx of Sytox, an impermeant fluorescent dye, into the Env expressing cells following the interaction with CD4CXCR4 complexes and fusion was monitored by observing dye transfer between Env-expressing cells and appropriate target cells. We also monitored the conformational changes in the Envs following their interactions with CD4 and CXCR4 by immunofluorescence using an anti-gp41 mAb that reacts with the six-helix bundle. In contrast to wild type, both Env mutants did not mediate cell fusion. The V2E Env did not mediate membrane destabilization. However, the Env with an unmodified fusion peptide but with a deletion of the sequence DELTA 665-682 in the membrane-proximal domain did mediate membrane destabilization. The WT and both mutant Envs undergo conformational changes detected by the anti-gp41 six-helix bundle mAbs. Our results suggest that in intact HIV-1 Env the membrane proximal domain is not required for membrane perturbations, but rather enables the bending of gp41 that is required for viral and target membranes to come together. Moreover, the observation that the DELTA 665-683 Env self-inserts its fusion peptide but does not cause fusion suggests that self-insertion of the fusion peptide is not sufficient for HIV-1 Env-mediated fusion.

Vishwa D Dixit, DVM, PhD - NIA
Ghrelin is expressed and secreted from human T lymphocytes and inhibits activation and leptin-induced inflammatory cytokine expression via functional Growth Hormone Secretagogue Receptors (GHS-R)
Ghrelin, a recently described 28 amino acid endogenous ligand for G protein-coupled seven transmembrane growth hormone secretagogue receptor (GHS-R), is produced primarily by the stomach serving as a potent circulating orexigen controlling energy expenditure, adiposity and GH secretion. While GHS-R mRNA is expressed in variety of tissues including lymphoid organs, the functional relevance of these receptors in human immune system remains to be established. Here, we report that GHS-R and Ghrelin are expressed in human T cells and specifically localize within GM1-associated lipid rafts. Ghrelin mRNA expression within human T cells revealed a greater than five-fold induction upon T-cell receptor ligation. In addition, the 117 amino acid pre-pro-form of ghrelin co-localizes within the Golgi apparatus, where it is presumably cleaved and subsequently secreted. Interestingly, ghrelin specifically inhibited the protein and mRNA expression of proinflammatory anorexigenic cytokines, IL-1 beta, IL-6 and TNF-alpha via functional GHS-R on the surface of human mononuclear and T cells. Given the believed mutually antagonistic effects of ghrelin and leptin on food intake and adiposity, the potential regulatory role of these hormones in controlling proinflammatory cytokine expression was also examined. Interestingly, ghrelin treatment resulted in a dose-dependent inhibition of leptin-induced proinflammatory cytokine expression. In addition, leptin failed to exert any significant effects on ghrelin protein or gene expression within human T cell cultures. More interestingly, we found that leptin treatment resulted in a significant increase in GHS-R mRNA expression by human T cells, suggesting the possible existence of regulatory feedback system by which these metabolic hormones control each other's immune effects. Finally, in a murine endotoxemia model, ghrelin inhibited proinflammatory mediators, IL-1 alpha, IL-1 beta, IL-6, TNF-alpha, IFN-gamma, IL-12, GCSF, KC and MIP-1 alpha without affecting IL-10, GMCSF and RANTES in serum. Moreover, ghrelin treatment significantly inhibited mRNA expression of IL-1 beta, IL-6 and TNF-alpha in spleen and liver of endotoxin- challenged mice and also attenuated anorexia. This is the first report describing a functional role for ghrelin and GHS-R within the immune system and also supports the potential use of ghrelin and GHS-R agonists in the management of wasting associated with chronic inflammation and cancer.

Arpad Dobolyi, PhD - NIMH
Identification of a new descending anatomical system and its involvement in peptidergic facilitation of spinal nociceptive processing
The descending control of nociceptive processing in the spinal cord (SC), first proposed as the gating theory, has become a fundamental part of our thinking about sensory information processing and has major clinical implications for analgesia. Many brain regions are known to exert descending inhibition, however, the existence and significance of anatomical systems devoted to descending facilitation remained a major question.
Tuberoinfundibular peptide of 39 residues (TIP39) was purified in our laboratory as ligand of the G-protein coupled parathyroid hormone 2 receptor (PTH2R). Since PTH2R is abundant in lamina II of the SC, where peripheral nociceptive afferent fibers terminate, we investigated the nociceptive function of TIP39 in the SC. We first demonstrated that TIP39 fibers were present in the dorsal horn and the dorsolateral funiculus of the SC using immunocytochemistry. Using in situ hybridization, we also found that TIP39 is synthesized in only two regions in the nervous system, the posterior thalamus and the lateral pons. Lesions in the lateral pons but not in the posterior thalamus eliminated TIP39 fibers from the SC, thus we conclude that fibers of TIP39 cells located in the lateral pons descend in the dorsolateral funiculus and terminate in the dorsal horn of the SC. The precise localization of TIP39 cells in the lateral pons was examined by cytoarchitectonic and double immunocytochemical studies. TIP39 cells are located just dorsolateral to the A7 noradrenergic cell group in the caudal part of an area designated as the medial paralemniscal nucleus. This area has not yet been demonstrated to participate in the descending control of the SC, thus we identified a new descending modulatory system. We also found that TIP39 cells do not possess anatomical markers of catecholamines, acetyl-choline, GABA or glutamate suggesting that the major transmitter of the descending system is TIP39. TIP39 injected intrathecally into the SC dramatically decreased latencies (increased pain) in paw-pressure and tail-flick assays showing its involvement in both mechanically and thermally induced nociceptions; the role of endogenous TIP39 was further supported by opposite actions of anti-TIP39 antibody in the assays. Thus, we propose that the function of the newly identified descending TIP39ergic system is the facilitation of spinal nociceptive processing. Our study also provides a basis for developing novel drugs with selective anti-nociceptive action.

Miroslav Dundr, Ph.D. - NCI-CCR
DYNAMICS OF CAJAL BODY COMPONENTS IN LIVING CELLS: Functional link between Cajal Body and the nucleolus
Cajal bodies (CBs) are subnuclear domains that are highly enriched in numerous RNA processing factors, including snRNPs. In most cell types CBs colocalized with gems, which contain the survival of motor neurons (SMN) protein complex, which is required for the biogenesis of small nucle(o)lar RNPs. Recent evidence demonstrates that CBs are sites of assembly and maturation of small RNPs prior to their accumulation and function within other subnuclear domains. Residual CBs containing only nucleolar CB component but lacking spliceosomal snRNPs and the SMN complex are formed in coilin knockout cells indicating strong link between CB and the nucleolus function. We have analyzed the kinetics of multiple CB components in living cells using photobleaching microscopy. We have identified several kinetically distinct classes of CB components. GFP-coilin, GFP-SMN and Gemin3-GFP each exhibited long residence times on the order of minutes, consistent with their in vivo interaction and possible role in the structural maintenance of CBs. The nucleolar components of CBs, fibrillarin, U3 snoRNP-55kD, GAR1, Nopp140 and B23 have significantly shorter residence times in CBs on the order of tens of seconds. In contrast, spliceosomal snRNP proteins SmB, SmD1 and SART3 were transiently associated with CBs. These observations reveal that functionally separate classes of CB components exhibit different residence times in CBs, suggesting their involvement in distinct CB functions. Coilin deletion mutants (1-92; 1-161) only briefly associate with CBs in contrast to mutants (1-248, 1-315, 1-482) which accumulate also in the nucleoli and exhibit similar residence time in CBs and in the nucleoli. GFP-SMN and fibrillarin-GFP exhibit similar dissociation kinetics from CBs in coilin knockout cells, where CBs are separated from gems, as in normal cells. Similarly, GFP-coilin shows the same dissociation kinetics in CBs when they are separated from gems by expression of a coilin ∆RG mutant. These data indicate that the physical separation of CBs and gems does not markedly affect the kinetics of their individual components and suggests that these subdomains are functional independently. We suggest that CBs and gems are merged into one structure to combine their related tasks in biogenesis and maturation of sn(o)RNPs more efficiently than if they were physically separated.

Julie M Duque, B.Sc - NINDS
Practice of a unilateral finger movement alters the cortical movement representation of the opposite finger.
Introduction: Training consisting of repetition of a unilateral finger movement over a short period of time leads to plastic changes in the cortical movement representation of the trained finger. Besides, it has been suggested that the ability to execute a unilateral finger movement is acquired through the development of inhibitory interhemispheric interactions between homologous finger representations.
Purpose: In the present study, we tested the hypothesis that repetition of a unilateral finger movement modifies the cortical movement representation of the opposite, untrained, finger in the motor cortex ipsilateral to the trained finger.
Methods: Transcranial magnetic stimulation (TMS) was applied before training to each motor cortex at the optimal position to elicit an isolated index finger movement. The direction of TMS-evoked movements was recorded with an accelerometer. In all subjects (n=9), TMS elicited abduction in the non-dominant index finger (inclusion criterion). Motor training consisted of an abduction of the dominant index finger repeated at 1Hz for 30 minutes. After training, TMS was re-applied at the same intensities and locations. The outcome measure was the number of TMS-evoked movements falling in the training target zone (TTZ), defined as a +/- 45-degree quadrant centered on the mean training direction, for the trained finger and, in the mirror-TTZ (quadrant centered on the mirror training direction) for the untrained finger, after training relative to the pre-training condition.
Results: Motor training led (1) in the trained finger, to an increase in the number of TMS-evoked movements in the TTZ (paired t test, p=0.035) and (2) in the untrained finger, to a decrease in the number of TMS-evoked movements in the mirror-TTZ (paired t test, p<0.001). The cortical excitability of the trained agonist muscle increased, whereas it decreased for the homologous untrained muscle.
Conclusions: These results indicate that the training of a given unilateral finger movement also alters the cortical movement representation of the opposite, untrained, finger. The plastic changes are specifically related to the kinematics of the trained movement and suggest that performance of a unilateral finger movement requires suppression of the mirror motion in the opposite finger.

Paul G Egla