Critical Appraisal of Artificial Intelligence-Enabled Imaging Tools Using the Levels of Evidence System

Clinical adoption of an artificial intelligence-enabled imaging tool requires critical appraisal of its life cycle from development to implementation by using a systematic, standardized, and objective approach that can verify both its technical and clinical efficacy. Toward this concerted effort, the ASFNR/ASNR Artificial Intelligence Workshop Technology Working Group is proposing a hierarchal evaluation system based on the quality, type, and amount of scientific evidence that the artificial intelligence-enabled tool can demonstrate for each component of its life cycle. The current proposal is modeled after the levels of evidence in medicine, with the uppermost level of the hierarchy showing the strongest evidence for potential impact on patient care and health care outcomes. The intended goal of establishing an evidence-based evaluation system is to encourage transparency, foster an understanding of the creation of artificial intelligence tools and the artificial intelligence decision-making process, and to report the relevant data on the efficacy of artificial intelligence tools that are developed. The proposed system is an essential step in working toward a more formalized, clinically validated, and regulated framework for the safe and effective deployment of artificial intelligence imaging applications that will be used in clinical practice.

Multi-Parametric Radiomic Model to Predict 1p/19q Co-Deletion in Patients with IDH-1 Mutant Glioma: Added Value to the T2-FLAIR Mismatch Sign

PURPOSE

The T2-FLAIR mismatch sign has shown promise in determining IDH mutant 1p/19q non-co-deleted gliomas with a high specificity and modest sensitivity. To develop a multi-parametric radiomic model using MRI to predict 1p/19q co-deletion status in patients with newly diagnosed IDH1 mutant glioma and to perform a comparative analysis to T2-FLAIR mismatch sign+.

METHODS

In this retrospective study, patients with diagnosis of IDH1 mutant gliomas with known 1p/19q status who had preoperative MRI were included. T2-FLAIR mismatch was evaluated independently by two board-certified neuroradiologists. Texture features were extracted from glioma segmentation of FLAIR images. eXtremeGradient Boosting (XGboost) classifiers were used for model development. Leave-one-out-cross-validation (LOOCV) and external validation performances were reported for both the training and external validation sets.

RESULTS

A total of 103 patients were included for model development and 18 patients for external testing validation. The diagnostic performance (sensitivity/specificity/accuracy) in the determination of the 1p/19q co-deletion status was 59%/83%/67% (training) and 62.5%/70.0%/66.3% (testing) for the T2-FLAIR mismatch sign. This was significantly improved (p = 0.04) using the radiomics model to 77.9%/82.8%/80.3% (training) and 87.5%/89.9%/88.8% (testing), respectively. The addition of radiomics as a computer-assisted tool resulted in significant (p = 0.02) improvement in the performance of the neuroradiologist with 13 additional corrected cases in comparison to just using the T2-FLAIR mismatch sign.

CONCLUSION

The proposed radiomic model provides much needed sensitivity to the highly specific T2-FLAIR mismatch sign in the determination of the 1p/19q non-co-deletion status and improves the overall diagnostic performance of neuroradiologists when used as an assistive tool.

Performance of high-resolution CT for detection and discrimination tasks related to stenotic lesions - A phantom study using model observers

BACKGROUND

Accurate detection and grading of atheromatous stenotic lesions within the cardiac, renal, and intracranial vasculature is imperative for early recognition of disease and guiding treatment strategies.

PURPOSE

In this work, a stenotic lesion phantom was used to compare high resolution and normal resolution modes on the same CT scanner in terms of detection and size discrimination performance.

MATERIALS AND METHODS

The phantom is comprised of three acrylic cylinders (each 15.0 cm in diameter and 1.3 cm thick) with a matching array of holes in each module. The outer two modules contain holes that are slightly larger than the corresponding hole in the central module to simulate stenotic narrowing in vasculature. The stack of modules was submerged in an iodine solution simulating contrast-enhanced stenotic lesions with a range of lumen diameters (1.32-10.08 mm) and stenosis severity (0%, 50%, 60%, 70%, and 80%). The phantom was imaged on the Canon Aquilion Precision high-resolution CT scanner in high-resolution (HR) mode (0.25 mm × 0.50 mm detector element size) and normal-resolution (NR) mode (0.50 mm × 0.50 mm) using 120 kV and two dose levels (14 and 21 mGy SSDE) with 30 repeat scans acquired for each combination_. Filtered back-projection (FBP) and a hybrid-iterative reconstruction (AIDR) were used with the FC18 kernel, as well as a deep learning algorithm (AiCE) which is only available for HR. A non-prewhitening model observer with an eye filter was implemented to quantify performance for detection and size discrimination tasks in the axial plane.

RESULTS

Detection performance improved with increasing diameter, dose, and for AIDR in comparison to FBP for a fixed resolution mode. Performance in the HR mode was generally higher than NR for the smaller lumen diameters (1-5 mm) with decreasing differences as the diameter increased. Performance in NR mode surpassed HR mode for lumen diameters greater than ∼4 mm and ∼5 mm for 14 mGy and 21 mGy, respectively. AiCE provided consistently higher detection performance compared with AIDR-FC18 (48% higher for a 6 mm lumen diameter). Discrimination performance increased with increasing nominal diameter, dose, and for larger differences in stenosis severity. When comparing discrimination performance in HR to NR modes, the largest relative differences occur at the smallest nominal diameters and smallest differences in stenosis severity. The AiCE reconstruction algorithm produced the highest overall discrimination performance values, and these were significantly higher than AIDR-FC18 for nominal diameters of 7.14 and 10.08 mm.

CONCLUSIONS

HR mode outperforms NR for detection up to a specific diameter and the results improve with AiCE and for higher dose levels. For the task of size discrimination, HR mode consistently outperforms NR if AIDR-FC18 is used for dose levels of at least 21 mGy, and the results improve with AiCE and for the smallest differences in stenosis severity investigated (50% vs. 60%). High-resolution CT appears to be beneficial for detecting smaller simulated lumen diameters (<5 mm) and is generally advantageous for discrimination tasks related to stenotic lesions, which inherently contain information at higher frequencies, given the right reconstruction algorithm and dose level.

POS1169 THE INFLAMMATION INDUCED BY MONOSODIUM URATE AND CALCIUM PYROPHOSPHATE CRYSTALS DEPENDS ON OSMOLARITY AND AQUAPORIN CHANNELS

Background

The inflammation induced by monosodium urate (MSU) and calcium pyrophosphate (CPP) crystals is driven by interleukin (IL)-1β production. This later relies on NLRP3 inflammasome which can be activated by variation of ion concentration.

Objectives

To assess the role of osmolarity and water flux in MSU and CPP crystal-induced inflammation.

Methods

In vitro, THP1 monocytes were stimulated by pyrogen-free synthetic MSU and CPP crystals in iso-, hypo- or hyperosmotic media. Cytokine production was quantified by ELISA in cell culture supernatants. Cell size was measured using video microscopy. The role of aquaporin channels was assessed by pharmacological inhibitor (mercury chloride, HgCl2). In vivo, murine air pouch model was used. MSU and CPP crystals were injected in air pouch of mice treated or not with HgCl2 or mannitol. Osmolarity of mouse sera and patient synovial fluids (SF) were measured using freezing point osmometer. The size of cells collected from SF was assessed with imageJ software.

Results

MSU and CPP crystal-induced IL-1β production was substantially reduced by HgCl2 treatment (MSU 4900 vs 880 pg/ml; CPP 10500 vs 980, p<0.0001) or when cells were cultured in hyperosmotic medium. MSU and CPP crystals induced a transient increase in cell size which was 1.6 and 1.5 bigger after 30 and 100 min of stimulation by MSU and CPP crystals, respectively. After 150 min of stimulation, cell size decreased to their baseline size. Cell size increase was abolished by HgCl2 or hyperosmotic medium. In vivo, MSU and CPP crystal-induced inflammation (assessed by cell infiltration, IL-1β and CXCL2 production in air pouch lavage) was drastically reduced by HgCl2 or mannitol treatment. The serum osmolarity was higher in mannitol-treated mice than untreated mice (320 vs 300 mmosm/L). In patients, cells collected from SF during CPP or MSU crystal-induced flares had a bigger size than cells collected from osteoarthitic SF. The osmolarity of MSU or CPP crystal-containing SF was lower than the osmolarity of osteoarthritic SF (270 vs 310 mmosm/L). Finally, the IL-1β concentration in SF was strongly correlated with cell size and SF osmolarity.

Conclusion

These results suggest that the variation of osmolarity plays central role in MSU and CPP crystal-induced inflammation. Deciphering how crystals modulate osmolarity will identify new therapeutic targets.

Disclosure of Interests

None declared

Artificial Intelligence in Head and Neck Imaging

Artificial intelligence (AI) can be applied to head and neck imaging to augment image quality and various clinical tasks including segmentation of tumor volumes, tumor characterization, tumor prognostication and treatment response, and prediction of metastatic lymph node disease. Head and neck oncology care is well positioned for the application of AI since treatment is guided by a wealth of information derived from CT, MRI, and PET imaging data. AI-based methods can integrate complex imaging, histologic, molecular, and clinical data to model tumor biology and behavior, and potentially identify associations, far beyond what conventional qualitative imaging can provide alone.

High-Resolution CT Imaging of the Temporal Bone: A Cadaveric Specimen Study

Objective  Super-high and ultra-high spatial resolution computed tomography (CT) imaging can be advantageous for detecting temporal bone pathology and guiding treatment strategies. Methods  Six temporal bone cadaveric specimens were used to evaluate the temporal bone microanatomic structures utilizing the following CT reconstruction modes: normal resolution (NR, 0.5-mm slice thickness, 512 ² matrix), high resolution (HR, 0.5-mm slice thickness, 1,024 ² matrix), super-high resolution (SHR, 0.25-mm slice thickness, 1,024 ² matrix), and ultra-high resolution (UHR, 0.25-mm slice thickness, 2,048 ² matrix). Noise and signal-to-noise ratio (SNR) for bone and air were measured at each reconstruction mode. Two observers assessed visualization of seven small anatomic structures using a 4-point scale at each reconstruction mode. Results  Noise was significantly higher and SNR significantly lower with increases in spatial resolution (NR, HR, and SHR). There was no statistical difference between SHR and UHR imaging with regard to noise and SNR. There was significantly improved visibility of all temporal bone osseous structures of interest with SHR and UHR imaging relative to NR imaging ( p  < 0.001) and most of the temporal bone osseous structures relative to HR imaging. There was no statistical difference in the subjective image quality between SHR and UHR imaging of the temporal bone ( p  ≥ 0.085). Conclusion  Super-high-resolution and ultra-high-resolution CT imaging results in significant improvement in image quality compared with normal-resolution and high-resolution CT imaging of the temporal bone. This preliminary study also demonstrates equivalency between super-high and ultra-high spatial resolution temporal bone CT imaging protocols for clinical use.

Cochlear Meniere's: A Distinct Clinical Entity With Isolated Cochlear Hydrops on High-Resolution MRI?

Objective: Describe the clinical characteristics of patients with isolated cochlear endolymphatic hydrops (EH). Study design: Clinical case series. Setting: Tertiary Neurotology referral clinic. Patients: All subjects presenting to a University Neurotology clinic during a 1-year period from July 2015 until August 2016 who had isolated cochlear EH on MRI. Patients with a history of temporal bone surgery prior to the MRI were excluded. Intervention: High-resolution delayed-intravenous contrast MRI. Main outcome measures: Audiometric and vestibular testing, clinical history analysis. Results: 10 subjects demonstrated isolated, unilateral cochlear hydrops on MRI. None of these patients met the criteria for Meniere's disease. Mean age of the group was 66.4 years and most were males (70%). Unilateral aural fullness (70%), tinnitus (80%), and hearing loss (90%) were frequently observed. Only one patient presented with unsteadiness (10%) and one patient had a single isolated spell of positional vertigo 1 month prior to the MRI (10%) but no further vertigo spells in the 4 years following the MRI. The mean PTA was 37.8 dB which was significantly decreased from the non-affected ear with PTA of 17.9 (p < 0.001). One patient developed vertiginous spells and unsteadiness 4 years after initial presentation and a repeat MRI revealed progression to utricular, saccular and cochlear hydrops. Vestibular testing was obtained in five patients with one patient presenting with 50% caloric paresis and all others normal. The most common treatment tried was acetazolamide in seven patients with 86% reporting subjective clinical improvement. Two out of the 10 patients had a history of migraine (20%). Conclusions: Patients with MRI exhibiting isolated cochlear EH present with predominantly auditory symptoms: mild to moderate low-frequency hearing loss, aural fullness, tinnitus without significant vertigo. Isolated cochlear hydrops is more common in males, average age in mid-60's and there is a low comorbidity of migraine headaches. This contrasts significantly with patients with isolated saccular hydrops on MRI from our prior studies. We believe that isolated cochlear EH with hearing loss but no vertigo is distinct from Meniere's disease or its variant delayed endolymphatic hydrops. We propose that cochlear Meniere's disease represents a distinct clinical entity that could be a variant of Meniere's disease.

Neuroimaging findings and pathophysiology of dorsal spinal arachnoid webs: illustrative case

BACKGROUND

Spinal arachnoid webs are uncommon and difficult to diagnose, especially because causative intradural transverse bands of arachnoid tissue are radiographically occult. Left untreated, arachnoid webs may cause progressive, debilitating, and permanent neurological dysfunction. Conversely, more than 90% of patients may experience rapid neurological recovery after resection, even with a prolonged duration of presenting symptoms. Indirect imaging signs such as spinal cord indentation and compression with cerebrospinal fluid (CSF) flow alteration provide crucial diagnostic clues that are critical in guiding appropriate management of such patients.

OBSERVATIONS

The authors reported a patient with no significant medical history who presented with back pain, progressive lower extremity weakness, gait ataxia, and bowel and bladder incontinence. They discussed multimodality imaging for determining the presence of arachnoid webs, including magnetic resonance imaging, phase-contrast CSF flow study, computed tomography myelography, and intraoperative ultrasound. They also discussed the detailed anatomy of the spinal subarachnoid space and a plausible pathophysiological mechanism for dorsal arachnoid webs.

LESSONS

The authors report on a patient who underwent comprehensive imaging evaluation detailing the arachnoid web and whose subsequent anatomical localization and surgical treatment resulted in a full neurological recovery.

A Comparison of ToxCast Test Results with In Vivo and Other In Vitro Endpoints for Neuro, Endocrine, and Developmental Toxicities: A Case Study Using Endosulfan and Methidathion

INTRODUCTION

The U.S. Environmental Protection Agency's (EPA's) Toxicity Forecaster (ToxCast) is a potential tool for chemical prioritization, hazard identification, and risk assessment. We conducted a case study to compare ToxCast data with endpoints from other in vitro and in vivo studies for two data-rich pesticides: endosulfan and methidathion.

METHODS

ToxCast assays for endocrine disruption, development (zebrafish), and neurotoxicity were qualitatively compared to traditional neurotoxicity, developmental and reproductive toxicity findings. We also used in vitro-in vivo extrapolation to convert half-maximal activity concentrations in active ToxCast assays to rat oral equivalent doses, and quantitatively compared these to the lowest observable effect level (LOEL) from in vivo studies.

RESULTS

Endosulfan was inactive for GABAA R, unlike in vivo; but active with dopamine transporter assays and was neurotoxic in zebrafish as expected. Methidathion was not active for these endpoints in vivo or in vitro. Acetylcholinesterase inhibition was ToxCast-inactive, although both pesticides are inhibitors in vivo. ToxCast results were generally inactive for endosulfan estrogen receptor agonism and androgen receptor antagonism unlike in vivo. Calculated oral equivalent doses for estrogen receptor and androgen receptor pathways and for zebrafish assays for both compounds were generally consistent with in vivo LOELs. Endosulfan showed neurotoxicity and both pesticides showed developmental effects in the zebrafish assays, although methidathion is not developmentally toxic in vivo.

CONCLUSIONS

ToxCast's predictions showed concordance on some endpoints and nonconcordance, consisting mainly of false inactives, in several critical endpoints, likely due to a lack of metabolic activation and limitations in assay design. Zebrafish assays were good predictors of developmental toxicity and neurotoxicity for endosulfan.

Anti-C1q antibodies in systemic lupus erythematosus

OBJECTIVE

Anti-C1q has been associated with systemic lupus erythematosus (SLE) and lupus nephritis in previous studies. We studied anti-C1q specificity for SLE (vs rheumatic disease controls) and the association with SLE manifestations in an international multicenter study.

METHODS

Information and blood samples were obtained in a cross-sectional study from patients with SLE (n = 308) and other rheumatologic diseases (n = 389) from 25 clinical sites (84% female, 68% Caucasian, 17% African descent, 8% Asian, 7% other). IgG anti-C1q against the collagen-like region was measured by ELISA.

RESULTS

Prevalence of anti-C1q was 28% (86/308) in patients with SLE and 13% (49/389) in controls (OR = 2.7, 95% CI: 1.8-4, p < 0.001). Anti-C1q was associated with proteinuria (OR = 3.0, 95% CI: 1.7-5.1, p < 0.001), red cell casts (OR = 2.6, 95% CI: 1.2-5.4, p = 0.015), anti-dsDNA (OR = 3.4, 95% CI: 1.9-6.1, p < 0.001) and anti-Smith (OR = 2.8, 95% CI: 1.5-5.0, p = 0.01). Anti-C1q was independently associated with renal involvement after adjustment for demographics, ANA, anti-dsDNA and low complement (OR = 2.3, 95% CI: 1.3-4.2, p < 0.01). Simultaneously positive anti-C1q, anti-dsDNA and low complement was strongly associated with renal involvement (OR = 14.9, 95% CI: 5.8-38.4, p < 0.01).

CONCLUSIONS

Anti-C1q was more common in patients with SLE and those of Asian race/ethnicity. We confirmed a significant association of anti-C1q with renal involvement, independent of demographics and other serologies. Anti-C1q in combination with anti-dsDNA and low complement was the strongest serological association with renal involvement. These data support the usefulness of anti-C1q in SLE, especially in lupus nephritis.

P3-16-08: A Phase 2, Randomized Open-Label Study of Iniparib, Administered Either Weekly or Twice-Weekly in Combination with Gemcitabine Plus Carboplatin in Patients with mTNBC

Background

Iniparib (BSI-201) is an investigational anticancer agent whose precise mechanism of action is under active investigation. In breast cancer cell lines and xenograft models of triple-negative breast cancer (TNBC), iniparib exhibits anti-proliferative activity and potentiates the cell cycle effects of some DNA damaging agents. In a randomized, open-label phase 2 study in pts with metastatic TNBC (mTNBC), iniparib combined with gemcitabine (G) and carboplatin (C) (GC) improved efficacy outcomes compared with GC alone. A confirmatory phase 3 study with GCI failed to meet pre-specified criteria for PFS and OS; however, an exploratory subset analysis demonstrated a potential benefit amongst 2nd/3rd line pts (O'Shaughnessy et al. ASCO 2011). Here we report results of a randomized phase 2 study (NCT01045304) in pts with mTNBC, which assesses efficacy and pharmacokinetics (PK) of iniparib administered either biw or qw in combination with GC.

Patients and methods: Eligible pts (N=163; median age 49 yrs) had documented and measurable TNBC, ECOG PS 0–1, normal organ/marrow function, and had received ≤2 prior chemotherapy (CT) regimens for metastatic disease. Pts were randomized (1:1) to receive G (1,000 mg/m2, IV, d 1, 8) plus C (AUC 2, IV, d 1, 8) and iniparib either biw (5.6 mg/kg, IV d 1,4,8,11) or qw (11.2 mg/kg, IV d 1,8) on a 21 d cycle. Pts were stratified according to prior CT for mTNBC (0 vs. 1–2). The primary efficacy endpoint was overall response rate (ORR; CR + PR); secondary endpoints included: clinical benefit rate (CBR; CR + PR + SD for 24 weeks), PFS, OS and PK.

Results: At the time of analysis, 23% of patients were still on treatment. The median number of cycles administered per patient was 6 in both arms; exposure to iniparib was identical. Safety data are not fully validated. All pts experienced at least 1 treatment emergent adverse event (TEAE). Grade (Gr) ≥3 TEAEs occurred in 94% and 85% of pts in the biw and qw arms, respectively. TEAEs Gr ≥3 occurring in ≥5% of pts regardless of relationship to study drug (biw vs qw) are as follows: blood and lymphatic 71% vs 67%; hepatobiliary 7.5% vs 9.8%; asthenia/fatigue 7.5% vs 11%; GI 8.8% vs 8.5%; infections 7.5% vs 3.7%; respiratory, thoracic and mediastinal 5% vs 8.5%, metabolism and nutrition 4% vs 6%. For response data see table.

Particle embolization of the bilateral superior and inferior alveolar arteries for life threatening dental socket hemorrhage

Particle embolization for epistaxis and intra-oral hemorrhage are performed on an as needed basis by neurointerventionalists. A case is presented of massive oral hemorrhage associated with end stage liver disease coagulopathy after tooth extraction of bilateral maxillary and mandibular molars. A man in his fifties with end stage liver disease who presented for evaluation of a syncopal episode was determined to be in hemorrhagic shock from 2 days of persistent oral bleeding after elective tooth extractions. Conservative management with multiple blood transfusion products, packing and vasoconstrictive spray was ineffective. Microcatheter angiograms of the alveolar arteries demonstrated blood and contrast pooling within the sockets of the extracted teeth. Selection of the bilateral supplying alveolar and inferior alveolar artery branches was achieved followed by polyvinyl alcohol particle embolization (250-355 μm). Polyvinyl alcohol particle embolization of dental socket hemorrhages is technically feasible and effective.

Extracranial sources of S100B do not affect serum levels

S100B, established as prevalent protein of the central nervous system, is a peripheral biomarker for blood-brain barrier disruption and often also a marker of brain injury. However, reports of extracranial sources of S100B, especially from adipose tissue, may confound its interpretation in the clinical setting. The objective of this study was to characterize the tissue specificity of S100B and assess how extracranial sources of S100B affect serum levels. The extracranial sources of S100B were determined by analyzing nine different types of human tissues by ELISA and Western blot. In addition, brain and adipose tissue were further analyzed by mass spectrometry. A study of 200 subjects was undertaken to determine the relationship between body mass index (BMI) and S100B serum levels. We also measured the levels of S100B homo- and heterodimers in serum quantitatively after blood-brain barrier disruption. Analysis of human tissues by ELISA and Western blot revealed variable levels of S100B expression. By ELISA, brain tissue expressed the highest S100B levels. Similarly, Western blot measurements revealed that brain tissue expressed high levels of S100B but comparable levels were found in skeletal muscle. Mass spectrometry of brain and adipose tissue confirmed the presence of S100B but also revealed the presence of S100A1. The analysis of 200 subjects revealed no statistically significant relationship between BMI and S100B levels. The main species of S100B released from the brain was the B-B homodimer. Our results show that extracranial sources of S100B do not affect serum levels. Thus, the diagnostic value of S100B and its negative predictive value in neurological diseases in intact subjects (without traumatic brain or bodily injury from accident or surgery) are not compromised in the clinical setting.

Ligand binding promotes prion protein aggregation--role of the octapeptide repeats

Aggregation of the normal cellular prion protein, PrP, is important in the pathogenesis of prion disease. PrP binds glycosaminoglycan (GAG) and divalent cations, such as Cu(2+) and Zn(2+). Here, we report our findings that GAG and Cu(2+) promote the aggregation of recombinant human PrP (rPrP). The normal cellular prion protein has five octapeptide repeats. In the presence of either GAG or Cu(2+), mutant rPrPs with eight or ten octapeptide repeats are more aggregation prone, exhibit faster kinetics and form larger aggregates than wild-type PrP. When the GAG-binding motif, KKRPK, is deleted the effect of GAG but not that of Cu(2+) is abolished. By contrast, when the Cu(2+)-binding motif, the octapeptide-repeat region, is deleted, neither GAG nor Cu(2+) is able to promote aggregation. Therefore, the octapeptide-repeat region is critical in the aggregation of rPrP, irrespective of the promoting ligand. Furthermore, aggregation of rPrP in the presence of GAG is blocked with anti-PrP mAbs, whereas none of the tested anti-PrP mAbs block Cu(2+)-promoted aggregation. However, a mAb that is specific for an epitope at the N-terminus enhances aggregation in the presence of either GAG or Cu(2+). Therefore, although binding of either GAG or Cu(2+) promotes the aggregation of rPrP, their aggregation processes are different, suggesting multiple pathways of rPrP aggregation.

Human prion proteins with pathogenic mutations share common conformational changes resulting in enhanced binding to glycosaminoglycans

Mutation in the prion gene PRNP accounts for 10-15% of human prion diseases. However, little is known about the mechanisms by which mutant prion proteins (PrPs) cause disease. Here we investigated the effects of 10 different pathogenic mutations on the conformation and ligand-binding activity of recombinant human PrP (rPrP). We found that mutant rPrPs react more strongly with N terminus-specific antibodies, indicative of a more exposed N terminus. The N terminus of PrP contains a glycosaminoglycan (GAG)-binding motif. Binding of GAG is important in prion disease. Accordingly, all mutant rPrPs bind more GAG, and GAG promotes the aggregation of mutant rPrPs more efficiently than wild-type recombinant normal cellular PrP (rPrP(C)). Furthermore, point mutations in PRNP also cause conformational changes in the region between residues 109 and 136, resulting in the exposure of a second, normally buried, GAG-binding motif. Importantly, brain-derived PrP from transgenic mice, which express a pathogenic mutant with nine extra octapeptide repeats, also binds more strongly to GAG than wild-type PrP(C). Thus, several rPrPs with distinct pathogenic mutations have common conformational changes, which enhance binding to GAG. These changes may contribute to the pathogenesis of inherited prion diseases.

HLA-DR3 transgenic mice immunized with adenovirus encoding the thyrotropin receptor: T cell epitopes and functional analysis of the CD40 Graves' polymorphism

The major histocompatibility (MHC) molecule HLA-DR3 is a susceptibility gene for Graves' disease (GD) in Caucasians. Mice lacking murine MHC and expressing human HLA-DR3 develop thyrotropin receptor (TSHR) antibodies and sometimes hyperthyroidism after vaccination with TSHR-DNA. MHC molecules present peptides processed from antigens to T cells. Therefore, we used DR3-transgenic mice to investigate recognition of TSHR ectodomain peptides. After immunization with TSHR A-subunit adenovirus (A-subunit-Ad) but not control-adenovirus (Control-Ad), splenocytes from DR3 mice responded to A-subunit protein in culture by producing interferon-gamma (IFN-gamma). When challenged with 29 overlapping TSHR peptides, splenocytes from A-subunit-Ad- or Control-Ad-immunized mice responded to several peptides. However, in splenocytes from A-subunit-Ad but not Control-Ad mice, a peptide containing TSHR residues 142-161 induced significantly more IFN-gamma than the same splenocytes in medium alone. Immunized DR3 mice also permitted testing the TSHR-specific function of the CD40 single nucleotide polymorphism (C vs. T) associated with GD. Of three human DR3 human Epstein-Barr virus lines (EBVL), two had C in both alleles (CC) and one was CT. However, these EBVL presented peptides poorly and there was no difference between CC vs. CT EBVL in peptide presentation to splenocytes from immunized mice. A peptide corresponding to residues 145-163 is one of seven suggested to be important in GD based on HLA-binding affinities, T-epitope algorithms, and human studies. Consequently, as in human GD, TSHR amino acids 142-161 appear to include a major T cell epitope in HLA-DR3 transgenic mice immunized with A-subunit-Ad.

Targeted expression of the human thyrotropin receptor A-subunit to the mouse thyroid: insight into overcoming the lack of response to A-subunit adenovirus immunization

The thyrotropin receptor (TSHR), the major autoantigen in Graves' disease, is posttranslationally modified by intramolecular cleavage to form disulfide-linked A- and B-subunits. Because Graves' hyperthyroidism is preferentially induced in BALB/c mice using adenovirus encoding the free A-subunit rather than full-length human TSHR, the shed A-subunit appears to drive the disease-associated autoimmune response. To further investigate this phenomenon, we generated transgenic mice with the human A-subunit targeted to the thyroid. Founder transgenic mice had normal thyroid function and were backcrossed to BALB/c. The A-subunit mRNA expression was confirmed in thyroid tissue. Unlike wild-type littermates, transgenic mice immunized with low-dose A-subunit adenovirus failed to develop TSHR Abs, hyperthyroidism, or splenocyte responses to TSHR Ag. Conventional immunization with A-subunit protein and adjuvants induced TSHR Abs lacking the characteristics of human autoantibodies. Unresponsiveness was partially overcome using high-dose, full-length human TSHR adenovirus. Although of low titer, these induced Abs recognized the N terminus of the A-subunit, and splenocytes responded to A-subunit peptides. Therefore, "non-self" regions in the B-subunit did not contribute to inducing responses. Indeed, transgenic mice immunized with high-dose A-subunit adenovirus developed TSHR Abs with thyrotropin-binding inhibitory activity, although at lower titers than wild-type littermates, suggesting down-regulation in the transgenic mice. In conclusion, in mice expressing a human A-subunit transgene in the thyroid, non-self human B-subunit epitopes are not necessary to induce responses to the A-subunit. Our findings raise the possibility that autoimmunity to the TSHR in humans may not involve epitopes on a cross-reacting protein, but rather, strong adjuvant signals provided in bystander immune responses.

Relationship between thyroid peroxidase T cell epitope restriction and antibody recognition of the autoantibody immunodominant region in human leukocyte antigen DR3 transgenic mice

We investigated the relationship between thyroid peroxidase (TPO) antibody and T lymphocyte epitopes in TPO-adenovirus (TPO-Ad) immunized BALB/c mice and mice transgenic for the human class II molecule DR3 associated with human thyroid autoimmunity. TPO autoantibodies are largely restricted to an immunodominant region (IDR). BALB/c mice immunized with fewer (10(7) vs. 10(9)) TPO-Ad particles developed TPO antibodies with lower titers that displayed greater restriction to the IDR. However, as with higher-dose TPO-Ad immunization, T cell epitopes (assessed by splenocyte interferon-gamma response to TPO in vitro) were highly diverse and variable in different animals. In contrast, DR3 mice immunized the higher TPO-Ad dose (10(9) particles) had high TPO antibody levels that showed relative focus on the IDR. Moreover, T cell epitopes recognized by splenocytes from DR3 mice showed greater restriction than BALB/c mice. Antibody affinities for TPO were higher in DR3 than in BALB/c mice. The present study indicates that weak TPO-Ad immunization of BALB/c mice (with consequent low TPO antibody titers) is required for enhanced IDR focus yet is not associated with T cell epitopic restriction. Humanized DR3 transgenic mice, despite stronger TPO-Ad immunization, develop higher titer TPO antibodies that do focus on the autoantibody IDR with T cells that recognize a more limited range of TPO peptides. These data suggest a relationship between major histocompatibility complex class II molecules and the development of antibodies to the IDR, a feature of human thyroid autoimmunity.

Nedd4 regulates ubiquitination and stability of the guanine-nucleotide exchange factor CNrasGEF

Cyclic nucleotide ras GEF (CNrasGEF) is a guanine-nucleotide exchange factor previously isolated in a screen for Nedd4-WW domain interacting proteins (Pham, N., Cheglakov, I., Koch, C. A., de Hoog, C. L., Moran, M. F., and Rotin, D. (2000) Curr. Biol. 10, 555-558). It activates Ras in a cAMP-dependent manner and Rap-1 independent of cAMP. Here we show that CNrasGEF is a likely substrate of the ubiquitin protein ligase Nedd4. CNrasGEF possesses two PY motifs at its C terminus that are responsible for binding to Nedd4 in vitro. Moreover, Nedd4 and CNrasGEF co-immunoprecipitate from 293T cells expressing ectopic CNrasGEF and endogenous Nedd4, and this co-immunoprecipitation is abrogated in PY motif-mutated CNrasGEF (CNrasGEFDelta2PY). CNrasGEF is ubiquitinated in cells, and this ubiquitination is augmented upon overexpression of wt-Nedd4 but is inhibited in cells overexpressing a catalytically inactive Nedd4 (Nedd4(CS)) or in cells expressing CNrasGEFDelta2PY, which cannot bind Nedd4. Moreover, pulse-chase experiments have demonstrated that the half-life of CNrasGEF is reduced 5-fold (from approximately 10 to approximately 2 h) in cells co-expressing Nedd4 with CNrasGEF but not with CNrasGEFDelta2PY (t(0.5) approximately 14 h). CNrasGEF is also stabilized in cells co-expressing Nedd4(CS) or following treatment with lactacystin, indicating proteasomal degradation of this protein. Deletion/mutation of the CDC25 domain to abrogate Ras (or Rap-1) binding leads to impaired ubiquitination of CNrasGEF, suggesting that such binding is critical for ubiquitination. Treatment of cells with the cAMP analogue 8-bromo-cAMP does not affect the ability of CNrasGEF to bind Nedd4 nor its level of ubiquitination, suggesting that Ras binding per se and not its activation is the critical step in triggering ubiquitination of CNrasGEF. These results suggest that CNrasGEF is a substrate for Nedd4, which regulates its ubiquitination and stability in cells.

The guanine nucleotide exchange factor CNrasGEF activates ras in response to cAMP and cGMP

Small GTPase proteins such as Ras are key regulators of cellular proliferation and are activated by guanine nucleotide exchange/releasing factors (GEFs/GRFs). Three classes of Ras GRFs have been identified to date, represented by Sos1/2, Ras-GRF1/2 and Ras-GRP. Here, we describe a novel candidate Ras activator, cyclic nucleotide rasGEF (CNrasGEF), which contains CDC25, Ras exchange motif (REM), Ras-association (RA), PDZ and cNMP (cAMP/cGMP) binding (cNMP-BD) domains, two PY motifs and a carboxy-terminal SxV sequence. CNrasGEF can activate Ras in vitro, and it binds cAMP directly via its cNMP-BD. In cells, CNrasGEF activates Ras in response to elevation of intracellular cAMP or cGMP, or treatment with their analogues 8-Br-cAMP or 8-Br-cGMP, independently of protein kinases A and G (PKA and PKG). This activation is prevented in CNrasGEF lacking its CDC25 domain or cNMP-BD. CNrasGEF can also activate the small GTPase Rap1 in cells, but this activation is constitutive and independent of cAMP. CNrasGEF is expressed mainly in the brain and is localized at the plasma membrane, a localization dependent on the presence of intact PDZ domain but not the SxV sequence. These results suggest that CNrasGEF may directly connect cAMP-generating pathways or cGMP-generating pathways to Ras.

Managed care, school health programs, and adolescent health services: opportunities for health promotion

The rapid expansion of managed care creates opportunities and dilemmas for those involved in school health and adolescent health promotion. Managed care organizations (MCOs), public health agencies, and school and adolescent health providers share certain common goals and priorities including an emphasis on prevention, cost-effectiveness, and quality of care--and a willingness to explore innovative approaches to health promotion and disease prevention. However, MCOs often face conflicting challenges, balancing the goals of cost containment and investment in prevention. In considering support for school health programs, MCOs will be interested in evidence about the effectiveness of services in improving health and/or reducing medical expenditures. Mechanisms for improving prevention efforts within MCOs include quality assurance systems to monitor the performance of health plans, practice guidelines from professional organizations, and the contracting process between payers and health care providers. Development of partnerships between MCOs and schools will be a challenge given competing priorities, variation in managed care arrangements, structural differences between MCOs and schools, and variability in services provided by school health programs.

Cross-regulatory roles of interleukin (IL)-12 and IL-10 in atherosclerosis

T cell cytokines are known to play a major role in determining protection and pathology in infectious disease. It has recently become clear that IL-12 is a key inducer of the type 1 T cell cytokine pattern characterized by production of IFN-gamma. Conversely, IL-10 down-regulates IL-12 production and type 1 cytokine responses. We have investigated whether IL-12 and IL-10 might be involved in a chronic inflammatory reaction, atherosclerosis. In atherosclerotic plaques, we found strong expression of IFN-gamma but not IL-4 mRNAs as compared to normal arteries. IL-12 p40 mRNA and IL-12 p70 protein were also found to be abundant in atherosclerotic plaques. IL-12 was induced in monocytes in vitro in response to highly oxidized LDL but not minimally modified LDL. The cross-regulatory role of IL-10 was indicated by the expression of IL-10 in some atherosclerotic lesions, and the demonstration that exogenous rIL-10 inhibited LDL-induced IL-12 release. These data suggest that the balance between IL-12 and IL-10 production contributes to the level of immune-mediated tissue injury in atherosclerotsis.

Identification of a transcriptional activator-binding element in the 27-kilodalton zein promoter, the -300 element

By utilizing a homologous transient-expression system, we have shown that a 58-bp sequence from the gamma-class 27-kDa zein promoter, spanning from -307 to -250 relative to the transcription start site, confers a high level of transcriptional activity on a truncated plant viral promoter. The transcriptional activity mediated by the 58-bp sequence is orientation independent, and it is further enhanced as a result of its multimerization. A similarly high level of transcriptional activity was also observed in protoplasts isolated from leaf tissue-derived maize suspension cells. In vitro binding and DNase I footprinting assays with nuclear protein prepared from cultured endosperm cells revealed the sequence-specific binding of a nuclear factor(s) to a 16-nucleotide sequence present in the 58-bp region. The nuclear factor binding sequence includes the -300 element, a cis-acting element highly conserved among different zein genes and many other cereal storage protein genes. A 23-bp oligonucleotide sequence containing the nuclear factor binding site is sufficient for binding the nuclear factor in vitro. It also confers a high level of transcriptional activity in vivo, but in an orientation-dependent manner. Four nucleotide substitutions in the -300 element drastically reduced binding and transcriptional activation by the nuclear factor. The same nuclear factor is abundant in the developing kernel endosperm and binds to the -300 element region of the 27-kDa or the alpha-class zein promoter. These results suggest that the highly conserved -300 element is involved in the common regulatory mechanisms mediating the coordinated expression of the zein genes.