Rodent Models of Alzheimer's Disease: Past Misconceptions and Future Prospects.

Alzheimer's disease (AD) is a progressive neurodegenerative disease with no effective treatments, not least due to the lack of authentic animal models. Typically, rodent models recapitulate the effects but not causes of AD, such as cholinergic neuron loss: lesioning of cholinergic neurons mimics the cognitive decline reminiscent of AD but not its neuropathology. Alternative models rely on the overexpression of genes associated with familial AD, such as amyloid precursor protein, or have genetically amplified expression of mutant tau. Yet transgenic rodent models poorly replicate the neuropathogenesis and protein overexpression patterns of sporadic AD. Seeding rodents with amyloid or tau facilitates the formation of these pathologies but cannot account for their initial accumulation. Intracerebral infusion of proinflammatory agents offer an alternative model, but these fail to replicate the cause of AD. A novel model is therefore needed, perhaps similar to those used for Parkinson's disease, namely adult wildtype rodents with neuron-specific (dopaminergic) lesions within the same vulnerable brainstem nuclei, 'the isodendritic core', which are the first to degenerate in AD. Site-selective targeting of these nuclei in adult rodents may recapitulate the initial neurodegenerative processes in AD to faithfully mimic its pathogenesis and progression, ultimately leading to presymptomatic biomarkers and preventative therapies.

Evidence for a novel neuronal mechanism driving Alzheimer's disease, upstream of amyloid.

This perspective offers an alternative to the amyloid hypothesis in the etiology of Alzheimer's disease (AD). We review evidence for a novel signaling mechanism based on a little-known peptide, T14. T14 could drive neurodegeneration as an aberrantly activated process of plasticity selective to interconnecting subcortical nuclei, the isodendritic core, where cell loss starts at the pre-symptomatic stages of the disease. Each of these cell groups has the capacity to form T14, which can stimulate production of p-Tau and ß-amyloid, suggestive of an upstream driver of neurodegeneration. Moreover, results in an animal AD model show that antagonism of T14 with a cyclated variant, NBP14, prevents formation of ß-amyloid, and restores cognitive function to that of wild-type counterparts. Any diagnostic and/or therapeutic strategy based on T14-NBP14 awaits validation in clinical trials. However, an understanding of this novel signaling system could bring much-needed fresh insights into the progression of cell loss underlying AD. HIGHLIGHTS: The possible primary mechanism of neurodegeneration upstream of amyloid. Primary involvement of selectively vulnerable subcortical nuclei, isodendritic core. Bioactive peptide T14 trophic in development but toxic in context of mature brain. Potential for early-stage biomarker to detect Alzheimer's disease. Effective therapeutic halting neurodegeneration, validated already in 5XFAD mice.

A Novel Peptide Driving Neurodegeneration Appears Exclusively Linked to the α7 Nicotinic Acetylcholine Receptor.

T14, a 14mer peptide, is significantly increased in the pre-symptomatic Alzheimer's disease brain, and growing evidence implies its pivotal role in neurodegeneration. Here, we explore the subsequent intracellular events following binding of T14 to its target α7 nicotinic acetylcholine receptor (nAChR). Specifically, we test how various experimental manipulations of PC12 cells impact T14-induced functional outcomes. Three preparations were compared: (i) undifferentiated vs. NGF-differentiated cells; (ii) cells transfected with an overexpression of the target α7 nAChR vs. wild type cells; (iii) cells transfected with a mutant α7 nAChR containing a mutation in the G protein-binding cluster, vs. cells transfected with an overexpression of the target α7 nAChR, in three functional assays - calcium influx, cell viability, and acetylcholinesterase release. NGF-differentiated PC12 cells were less sensitive than undifferentiated cells to the concentration-dependent T14 treatment, in all the functional assays performed. The overexpression of α7 nAChR in PC12 cells promoted enhanced calcium influx when compared with the wild type PC12 cells. The α7345-348 A mutation effectively abolished the T14-triggered responses across all the readouts observed. The close relationship between T14 and the α7 nAChR was further evidenced in the more physiological preparation of ex vivo rat brain, where T30 increased α7 nAChR mRNA, and finally in human brain post-mortem, where levels of T14 and α7 nAChR exhibited a strong correlation, reflecting the progression of neurodegeneration. Taken together these data would make it hard to account for T14 binding to any other receptor, and thus interception at this binding site would make a very attractive and remarkably specific therapeutic strategy.

Antagonism of a key peptide 'T14' driving neurodegeneration: Evaluation of a next generation therapeutic.

T14, a 14mer peptide derived from the C-terminus of acetylcholinesterase (AChE) is a signalling molecule that could drive neurodegeneration via the alpha 7 nicotinic acetylcholine receptor. Its levels increase as Alzheimer's pathology progresses; however, a cyclic variant of the compound, NBP14, can block the effects of the endogenous linear counterpart in-vitro, ex vivo, and in vivo. Here, we explore the antagonistic potential of two 6mer peptides, NBP6A and NBP6B. These are smaller linear versions of NBP14, designed to be more effective by modifying the amino acid residues to enhance receptor blockade alongside other relevant solubility parameters. The peptides were tested in-vitro in PC12 cells on three parameters, calcium influx, cell viability, and AChE release, and ex vivo using voltage sensitive dye imaging (VSDI) in rat brain slices. Neither NBP6A nor NBP6B applied alone had any effect. In PC12 cells, NBP6B was identified as the more potent molecule since it demonstrated more effective blockade of T14 action on calcium influx, cell viability, and AChE release. NBP6B was then further evaluated using VSDI, where it proved twice as potent as NBP14 in blocking the action of T14. The improved effect of NBP6B in blocking the actions of T14, combined with its smaller size suggests that this variant could have even greater therapeutic potential than its original cyclic compound, for treating neurodegenerative disorders.

A novel peptide 'T14' reflects age and photo-aging in human skin.

T14 is a 14mer peptide derived from the C-terminus of acetylcholinesterase (AChE). Once cleaved, it is independently bioactive of the parent molecule and enhances calcium influx in different cell types, in a range of scenarios: it binds to an allosteric site selectively on the alpha-7 receptor, where it modulates calcium influx and is thus a potential trophic agent, as already reported in a range of normal developmental scenarios. However, if inappropriately activated, this erstwhile beneficial effect converts to a toxic one, resulting in pathologies as disparate as Alzheimer's and various metastatic cancers. Given that epidermal keratinocyte cells have the same ectodermal origin as brain cells, as well as expressing AChE and the alpha-7 receptor, we have explored whether T14 plays a comparable role. Here we report that the T14 immunoreactivity is detectable in human keratinocytes with levels inversely related to age: this decrease is even more apparent with chronic photo-exposure and thus accelerated skin aging. We conclude that T14, an agent promoting cell growth and renewal in other parts of the body, also operates in skin, Moreover, monitoring of keratinocyte T14 levels might offer further insights into the now well reported link between degenerative diseases and epidermal cell profile.

A Novel Bioactive Peptide, T14, Selectively Activates mTORC1 Signalling: Therapeutic Implications for Neurodegeneration and Other Rapamycin-Sensitive Applications.

T14 modulates calcium influx via the α-7 nicotinic acetylcholine receptor to regulate cell growth. Inappropriate triggering of this process has been implicated in Alzheimer's disease (AD) and cancer, whereas T14 blockade has proven therapeutic potential in in vitro, ex vivo and in vivo models of these pathologies. Mammalian target of rapamycin complex 1 (mTORC1) is critical for growth, however its hyperactivation is implicated in AD and cancer. T14 is a product of the longer 30mer-T30. Recent work shows that T30 drives neurite growth in the human SH-SY5Y cell line via the mTOR pathway. Here, we demonstrate that T30 induces an increase in mTORC1 in PC12 cells, and ex vivo rat brain slices containing substantia nigra, but not mTORC2. The increase in mTORC1 by T30 in PC12 cells is attenuated by its blocker, NBP14. Moreover, in post-mortem human midbrain, T14 levels correlate significantly with mTORC1. Silencing mTORC1 reverses the effects of T30 on PC12 cells measured via AChE release in undifferentiated PC12 cells, whilst silencing mTORC2 does not. This suggests that T14 acts selectively via mTORC1. T14 blockade offers a preferable alternative to currently available blockers of mTOR as it would enable selective blockade of mTORC1, thereby reducing side effects associated with generalised mTOR blockade.

Antagonising a novel toxin "T14" in Alzheimer's disease: Comparison of receptor blocker versus antibody effects in vitro.

A 14mer peptide, T14, is a possible signaling molecule driving neurodegeneration. Its levels are doubled in the Alzheimer brain, but its effects can be blocked at the target alpha-7 receptor by a cyclised variant, 'NBP14', which has beneficial effects, in a transgenic mouse model, on the behavioral and histochemical profile. Since the antagonism of T14 has evident therapeutic potential, we explore here an alternative method of preventing its action by comparing the efficacy of NBP14 with a proprietorial polyclonal antibody against T14, 'Ab-19', at inhibiting three distinct effects of the peptide in PC12 cells: calcium influx, cell viability and compensatory acetylcholinesterase (AChE) release. None of these three parameters was affected by either blocking agent when applied alone. However, both NBP14 and the Ab-19 exhibited a dose-dependent profile against the actions of T14 in all three scenarios: the least sensitive effect observed was in the lower dose range, for both the antibody and the receptor blocker, in antagonizing T14-triggered release of AChE: this parameter is interpreted as indirect compensation for the T14-induced compromise of cell viability, triggered by the enhanced influx of calcium through the initial binding of the peptide to an allosteric site on the alpha-7 receptor. As such, it is the most delayed and indirect index of T14 action and thus the least relatively impacted by lowest doses of either NBP14 or Ab-19. In all three scenarios however the effects of T14 are successfully offset by either agent and thus offer two potentially very different therapies against Alzheimer's disease.

Characterization of a Bioactive Peptide T14 in the Human and Rodent Substantia Nigra: Implications for Neurodegenerative Disease.

The substantia nigra is generally considered to show significant cell loss not only in Parkinson's but also in Alzheimer's disease, conditions that share several neuropathological traits. An interesting feature of this nucleus is that the pars compacta dopaminergic neurons contain acetylcholinesterase (AChE). Independent of its enzymatic role, this protein is released from pars reticulata dendrites, with effects that have been observed in vitro, ex vivo and in vivo. The part of the molecule responsible for these actions has been identified as a 14-mer peptide, T14, cleaved from the AChE C-terminus and acting at an allosteric site on alpha-7 nicotinic receptors, with consequences implicated in neurodegeneration. Here, we show that free T14 is co-localized with tyrosine hydroxylase in rodent pars compacta neurons. In brains with Alzheimer's pathology, the T14 immunoreactivity in these neurons increases in density as their number decreases with the progression of the disease. To explore the functional implications of raised T14 levels in the substantia nigra, the effect of exogenous peptide on electrically evoked neuronal activation was tested in rat brain slices using optical imaging with a voltage-sensitive dye (Di-4-ANEPPS). A significant reduction in the activation response was observed; this was blocked by the cyclized variant of T14, NBP14. In contrast, no such effect of the peptide was seen in the striatum, a region lacking the T14 target, alpha-7 receptors. These findings add to the accumulating evidence that T14 is a key signaling molecule in neurodegenerative disorders and that its antagonist NBP14 has therapeutic potential.

When a trophic process turns toxic: Alzheimer's disease as an aberrant recapitulation of a developmental mechanism.

Here we review the idea that Alzheimer's disease (AD) results from aberrant activation of a normal developmental mechanism. This process operates in primarily vulnerable, subcortical nuclei with a distinguishing embryological provenance: the basal rather than the alar plate. All cells are dependent for growth on calcium influx yet these neurons retain a sensitivity to trophic factors into maturity. However, as the brain matures this action becomes detrimental such that the trophic process could turn toxic if triggered in adult brain, in retaliation to an initial insult. The signalling molecule driving this trophic-toxic mechanism is a 14mer peptide (T14) that acts on the alpha-7 receptor to enhance calcium entry, inducing excitotoxicity and proliferation of the receptor, perpetuating a feedforward cycle of neurodegeneration including production of beta-amyloid and p-tau. The T14 system has been previously unrecognised as a basic biological process, yet its pharmaceutical manipulation could have valuable clinical applications.

A novel process driving Alzheimer's disease validated in a mouse model: Therapeutic potential.

Introduction: The neuronal mechanism driving Alzheimer's disease (AD) is incompletely understood. Methods: Immunohistochemistry, pharmacology, biochemistry, and behavioral testing are employed in two pathological contexts-AD and a transgenic mouse model-to investigate T14, a 14mer peptide, as a key signaling molecule in the neuropathology. Results: T14 increases in AD brains as the disease progresses and is conspicuous in 5XFAD mice, where its immunoreactivity corresponds to that seen in AD: neurons immunoreactive for T14 in proximity to T14-immunoreactive plaques. NBP14 is a cyclized version of T14, which dose-dependently displaces binding of its linear counterpart to alpha-7 nicotinic receptors in AD brains. In 5XFAD mice, intranasal NBP14 for 14 weeks decreases brain amyloid and restores novel object recognition to that in wild-types. Discussion: These findings indicate that the T14 system, for which the signaling pathway is described here, contributes to the neuropathological process and that NBP14 warrants consideration for its therapeutic potential.

Protective and reversal actions of a novel peptidomimetic against a pivotal toxin implicated in Alzheimer's disease.

Despite the many attempts to understand the aetiology of Alzheimer's disease, the basic mechanisms accounting for the progressive cycle of neuronal loss are still unknown. Previous work has suggested that the pivotal molecule mediating neurodegeneration could be an independently acting peptide cleaved from acetylcholinesterase. This previously unidentified agent acts as a signalling molecule in selectively vulnerable groups of cells where erstwhile developmental mechanisms are activated inappropriately to have a toxic effect in the context of the mature brain. We have previously shown that the toxic actions of this peptide, whose level is doubled in the Alzheimer brain, can be blocked by a cyclised variant (NBP14). However, the size and properties of NBP14 would render it unlikely as a feasible therapeutic candidate. Here therefore we test a synthetic peptidomimetic (NB-0193), modelled on the binding of NBP14 to the target alpha-7 nicotinic receptor, and benchmarked against it to screen for reversal effects using real-time optical imaging in rat brain slices. The blocking action of NB-0193 was confirmed by testing its effect against peptide-induced calcium influx in cell cultures, where it showed a dose-dependent profile over a trophic-toxic range. Moreover, NB-0193 presented promising pharmacokinetic characteristics and could therefore prompt a new therapeutic approach against Alzheimer's disease.

A Multidisciplinary Approach Reveals an Age-Dependent Expression of a Novel Bioactive Peptide, Already Involved in Neurodegeneration, in the Postnatal Rat Forebrain.

The basal forebrain has received much attention due to its involvement in multiple cognitive functions, but little is known about the basic neuronal mechanisms underlying its development, nor those mediating its primary role in Alzheimer’s disease. We have previously suggested that a novel 14-mer peptide, ‘T14’, could play a pivotal role in Alzheimer’s disease, via reactivation of a developmental signaling pathway. In this study, we have characterized T14 in the context of post-natal rat brain development, using a combination of different techniques. Ex-vivo rat brain slices containing the basal forebrain, at different stages of development, were used to investigate large-scale neuronal network activity in real time with voltage-sensitive dye imaging. Subsequent Western blot analysis revealed the expression profile of endogenous T14, its target alpha7 nicotinic receptor and the familiar markers of Alzheimer’s: amyloid beta and phosphorylated Tau. Results indicated maximal neuronal activity at the earliest ages during development, reflected in a concomitant profile of T14 peptide levels and related proteins. In conclusion, these findings show that the peptide, already implicated in neurodegenerative events, has an age-dependent expression, suggesting a possible contribution to the physiological mechanisms underlying brain maturation.

Modulatory Effects of a Novel Cyclized Peptide in Reducing the Expression of Markers Linked to Alzheimer's Disease.

Despite many studies attempt to identify the primary mechanisms underlying neurodegeneration in Alzheimer's disease (AD), the key events still remain elusive. We have previously shown that a peptide cleaved from the acetylcholinesterase (AChE) C-terminus (T14) can play a pivotal role as a signaling molecule in neurodegeneration, via its interaction with the α7 nicotinic acetylcholine receptor. The main goal of this study is to determine whether a cyclized variant (NBP14) of the toxic AChE-derived peptide can antagonize the effects of its linear counterpart, T14, in modulating well-known markers linked to neurodegeneration. We investigate this hypothesis applying NBP14 on ex-vivo rat brain slices containing the basal forebrain. Western blot analysis revealed an inhibitory action of NBP14 on naturally occurring T14 peptide, as well as on endogenous amyloid beta, whereas the expression of the nicotinic receptor and phosphorylated Tau was relatively unaffected. These results further confirm the neurotoxic properties of the AChE-peptide and show for the first time in an ex-vivo preparation the possible neuroprotective activity of NBP14, over a protracted period of hours, indicating that T14 pathway may offer a new prospect for therapeutic intervention in AD pathobiology.

An Alternative Approach to Study Primary Events in Neurodegeneration Using Ex Vivo Rat Brain Slices.

Despite numerous studies that attempt to develop reliable animal models which reflecting the primary processes underlying neurodegeneration, very few have been widely accepted. Here, we propose a new procedure adapted from the well-known ex vivo brain slice technique, which offers a closer in vivo-like scenario than in vitro preparations, for investigating the early events triggering cell degeneration, as observed in Alzheimer's disease (AD). This variation consists of simple and easily reproducible steps, which enable preservation of the anatomical cytoarchitecture of the selected brain region and its local functionality in a physiological milieu. Different anatomical areas can be obtained from the same brain, providing the opportunity to perform multiple experiments with the treatments in question in a site-, dose-, and time-dependent manner. Potential limitations which could affect the outcomes related to this methodology are related to the conservation of the tissue, i.e., the maintenance of its anatomical integrity during the slicing and incubation steps and the section thickness, which can influence the biochemical and immunohistochemical analysis. This approach can be employed for different purposes, such as exploring molecular mechanisms involved in physiological or pathological conditions, drug screening, or dose-response assays. Finally, this protocol could also reduce the number of animals employed in behavioral studies. The application reported here has been recently described and tested for the first time on ex vivo rat brain slices containing the basal forebrain (BF), which is one of the cerebral regions primarily affected in AD. Specifically, it has been demonstrated that the administration of a toxic peptide derived from the C-terminus of acetylcholinesterase (AChE) could prompt an AD-like profile, triggering, along the antero-posterior axis of the BF, a differential expression of proteins altered in AD, such as the alpha7 nicotinic receptor (α7-nAChR), phosphorylated Tau (p-Tau), and amyloid beta (Aß).

A Novel Ex Vivo Model to Investigate the Underlying Mechanisms in Alzheimer's Disease.

Currently there is no widely accepted animal model reproducing the full pathological profile of Alzheimer's disease (AD), since the basic mechanisms of neurodegeneration are still poorly understood. We have proposed that the interaction between the α7 nicotinic acetylcholine receptor (α7-nAChR) and a recently discovered toxic peptide, cleaved from the acetylcholinesterase (AChE) C-terminus, could account for the aberrant processes occurring in AD. In this article we describe a new application on ex vivo model procedure, which combines the advantages of both in vivo and in vitro preparations, to study the effects of the AChE-derived peptide on the rat basal forebrain (BF). Western blot analysis showed that the levels of α7-nAChR, p-Tau and Aß are differentially expressed upon the AChE-peptide administration, in a selective site-dependent manner. In conclusion, this methodology demonstrates the action of a novel peptide in triggering an AD-like phenotype and proposes a new ex vivo approach for manipulating and monitoring neurochemical processes contributing to neurodegeneration, in a time-dependent and site-specific manner.

Optical imaging of the rat brain suggests a previously missing link between top-down and bottom-up nervous system function.

Optical imaging with voltage-sensitive dyes enables the visualization of extensive yet highly transient coalitions of neurons (assemblies) operating throughout the brain on a subsecond time scale. We suggest that operating at the mesoscale level of brain organization, neuronal assemblies may provide a functional link between "bottom-up" cellular mechanisms and "top-down" cognitive ones within anatomically defined regions. We demonstrate in ex vivo rat brain slices how varying spatiotemporal dynamics of assemblies reveal differences not previously appreciated between: different stages of development in cortical versus subcortical brain areas, different sensory modalities (hearing versus vision), different classes of psychoactive drugs (anesthetics versus analgesics), different effects of anesthesia linked to hyperbaric conditions and, in vivo, depths of anesthesia. The strategy of voltage-sensitive dye imaging is therefore as powerful as it is versatile and as such can now be applied to the evaluation of neurochemical signaling systems and the screening of related new drugs, as well as to mathematical modeling and, eventually, even theories of consciousness.

Tumor cell migration is inhibited by a novel therapeutic strategy antagonizing the alpha-7 receptor.

A 14mer peptide (T14) derived from the C-terminus of acetylcholinesterase (AChE) selectively activates metastatic breast cancer cells via the alpha-7 nicotinic receptor (α7 nAChR). This naturally occurring peptide is also present in brain, is elevated in Alzheimer's disease, and is antagonised by a cyclized variant (NBP-14). Here we investigated the effects of NBP-14 in six different cancer cell lines, primary leukemia B-cells and normal B-cells. All cells tested expressed α7 nAChR, intracellular and extracellular T14. However, NBP-14 showed low toxicity and weak anti-proliferative effects in the majority of the cell lines and was even less toxic in normal B-cells when compared to primary chronic lymphocytic leukemia cells (P < 0.001). Given the potential role of T14 peptide in metastasis, we next investigated the effects of NBP-14 on tumor cell migration, where it caused a dose-dependent reduction. The extent of NBP-14 inhibition positively correlated with the migration of the cells (r2 = 0.45; P = 0.06). Furthermore, NBP-14 preferentially inhibited the migration of primary leukemia cells when compared with normal B-cells (P = 0.0002); when the normal B-cell data was excluded, this correlation was strengthened (r2 = 0.80; P = 0.006). Importantly, the constitutive α7 nAChR expression positively correlated with intracellular T14 levels (r2 = 0.91; P = 0.0003) and inversely correlated with extracellular T14 levels in the cell culture supernatants (r2 = -0.79; P = 0.034). However, in the presence of NBP-14, α7 nAChR expression was reduced (P = 0.04) and the most migratory cells showed the largest reduction in expression. In conclusion, NBP-14-mediated antagonism of the α7 nAChR offers a novel therapeutic strategy with the potential to inhibit tumor cell migration.

Don't forget the dose: Improving computed tomography dosing for pediatric appendicitis.

BACKGROUND: A pediatric computed tomography (CT) radiation dose reduction program was implemented throughout our children's associated hospital system in 2010. We hypothesized that the CT dose received for evaluation of appendicitis in children would be significantly higher among the 40 referral, nonmember hospitals (NMH) than the 9 member hospitals (MH). METHODS: Preoperative CTs of pediatric (<18years) appendectomy patients between April 2012 and April 2015 were reviewed. Size specific dose estimate (SSDE), an approximation of absorbed dose incorporating patient diameter, and Effective Dose (ED) were calculated for each scan. RESULTS: 1128 (65%) of 1736 appendectomy patients underwent preoperative CT. 936 patients seen at and 102 children evaluated at NMH had dosing and patient diameter data for analysis. SSDE and ED were significantly higher with greater variance at NMH across all ages (all p<0.05, Figure). NMH's SSDE and ED also exceeded reference levels. CONCLUSION: Radiation exposure in CT scans for evaluation of pediatric appendicitis is significantly higher and more variable in NMH. A proactive approach to reduce dose, in addition to frequency, of CT scans in pediatric patients is essential. LEVEL OF EVIDENCE: Level III.

Prevalence of Pancreatic Steatosis at a Pediatric Tertiary Care Center.

OBJECTIVES: Pancreatic steatosis in adults has been proposed to be associated with obesity; however, data on pancreatic steatosis in children are lacking. Our study aimed to measure the prevalence of pancreatic steatosis in children and to examine its association with obesity and nonalcoholic fatty liver disease. METHODS: This is a retrospective chart review study of 232 patients 2 to 18 years old who underwent abdominal computed tomographic imaging in the emergency department or inpatient ward within a 1-year time span and from whom demographics, anthropometrics, and medical history were obtained. Our radiologist determined mean Hounsfield unit (HU) measurements for the pancreas, liver, and spleen. A difference of -20 between the pancreas and spleen (psHU) and between the liver and spleen was used to determine fatty infiltration. RESULTS: Of the 232 patients, 11.5% had a psHU less than -20. The prevalence of pancreatic steatosis was more than double among obese children (19%) than that in nonobese groups (8%). There is a significant correlation between the psHU and liver-spleen HU (r = 0.50, P < 0.001). CONCLUSIONS: Pancreatic steatosis was identified in 10% of the study population and is associated with obesity. Also, pancreatic steatosis is significantly associated with nonalcoholic fatty liver disease. This is the first study assessing the prevalence of pancreatic steatosis in children.