Extracellular truncated tau causes early presynaptic dysfunction associated with Alzheimer's disease and other tauopathies

doi:10.18632/onco_target.17371

. 2017 Apr 22;8(39):64745-64778.

doi: 10.18632/onco_target.17371. eCollection 2017 Sep 12.

Extracellular truncated tau causes early presynaptic dysfunction associated with Alzheimer's disease and other tauopathies

Fulvio Florenzano¹, Corsetti Veronica², Gabriele Ciasca³, Maria Teresa Ciotti⁴, Anna Pittaluga⁵, Gunedalina Olivero⁵, Marco Feligioni^{1

6}, Filomena Iannuzzi¹, Valentina Latina², Michele Francesco Maria Sciacca⁷, Alessandro Sinopoli⁷, Danilo Milardi⁷, Giuseppe Pappalardo⁷, De Spirito Marco³, Massimiliano Papi³, Anna Atlante^{8

9}, Antonella Bobba^{8

9}, Antonella Borreca⁴, Pietro Calissano¹, Giuseppina Amadoro^{1

2}

Affiliations

¹ European Brain Research Institute, Rome, Italy.
² Institute of Translational Pharmacology, CNR, Rome, Italy.
³ Institute of Physics, Catholic University of the Sacred Heart, Largo F Vito 1, Rome, Italy.
⁴ Institute of Cellular Biology and Neuroscience, CNR, IRCSS Santa Lucia Foundation, Rome, Italy.
⁵ Department of Pharmacy, Pharmacology and Toxicology Section, University of Genoa, Genoa, Viale Cembrano, Italy.
⁶ Department of Neurorehabilitation Sciences, Casa Cura Policlinico, Milan, Italy.
⁷ Institute of Biostructures and Bioimaging, CNR, Catania, Italy.
⁸ Institute of Biomembranes and Bioenergetics, CNR, Bari, Italy.
⁹ Center of Excellence for Biomedical Research, University of Genoa, Genoa, Viale Benedetto XV, Italy.

PMID: 29029390
PMCID: PMC5630290
DOI: 10.18632/onco_target.17371

Extracellular truncated tau causes early presynaptic dysfunction associated with Alzheimer's disease and other tauopathies

Fulvio Florenzano et al. Onco_target. 2017.

. 2017 Apr 22;8(39):64745-64778.

doi: 10.18632/onco_target.17371. eCollection 2017 Sep 12.

Authors

Affiliations

¹ European Brain Research Institute, Rome, Italy.
² Institute of Translational Pharmacology, CNR, Rome, Italy.
³ Institute of Physics, Catholic University of the Sacred Heart, Largo F Vito 1, Rome, Italy.
⁴ Institute of Cellular Biology and Neuroscience, CNR, IRCSS Santa Lucia Foundation, Rome, Italy.
⁵ Department of Pharmacy, Pharmacology and Toxicology Section, University of Genoa, Genoa, Viale Cembrano, Italy.
⁶ Department of Neurorehabilitation Sciences, Casa Cura Policlinico, Milan, Italy.
⁷ Institute of Biostructures and Bioimaging, CNR, Catania, Italy.
⁸ Institute of Biomembranes and Bioenergetics, CNR, Bari, Italy.
⁹ Center of Excellence for Biomedical Research, University of Genoa, Genoa, Viale Benedetto XV, Italy.

PMID: 29029390
PMCID: PMC5630290
DOI: 10.18632/onco_target.17371

Abstract

The largest part of tau secreted from AD nerve terminals and released in cerebral spinal fluid (CSF) is C-terminally truncated, soluble and unaggregated supporting potential extracellular role(s) of NH₂ -derived fragments of protein on synaptic dysfunction underlying neurodegenerative tauopathies, including Alzheimer's disease (AD). Here we show that sub-toxic doses of extracellular-applied human NH₂ tau 26-44 (aka NH ₂ htau) -which is the minimal active moiety of neurotoxic 20-22kDa peptide accumulating in vivo at AD synapses and secreted into parenchyma- acutely provokes presynaptic deficit in K⁺ -evoked glutamate release on hippocampal synaptosomes along with alteration in local Ca²⁺ dynamics. Neuritic dystrophy, microtubules breakdown, deregulation in presynaptic proteins and loss of mitochondria located at nerve endings are detected in hippocampal cultures only after prolonged exposure to NH ₂ htau. The specificity of these biological effects is supported by the lack of any significant change, either on neuronal activity or on cellular integrity, shown by administration of its reverse sequence counterpart which behaves as an inactive control, likely due to a poor conformational flexibility which makes it unable to dynamically perturb biomembrane-like environments. Our results demonstrate that one of the AD-relevant, soluble and secreted N-terminally truncated tau forms can early contribute to pathology outside of neurons causing alterations in synaptic activity at presynaptic level, independently of overt neurodegeneration.

Keywords: Alzheimer’s disease; Gero_target; extracellular tau; neurodegeneration; synapse(s); tau cleavage.

PubMed Disclaimer

Conflict of interest statement

CONFLICTS OF INTEREST The authors declare that they have no actual or potential conflicts of interest and that these data are not published elsewhere. In addition all authors approve the study described in this report.

Figures

**Figure 1. NH₂htau 26-44 (i.e.NH₂htau) shows an intrinsically disordered structure and is mainly monomeric in water environment**
a.-b.-c.-d. CD spectra of NH₂htau26-44 (10µM ) at different pH(A), at different concentrations of SDS(B), in presence of large unilamellar vesicles (LUV) composed by POPC/POPS 100 µM in a 7/3 ratio at pH7.4 at increasing temperatures (c) and at different percentage of TFE(d) are shown. Notice that NH₂htau shows in aqueous solution a negative peak typical for an unfolded protein with a minimum at 200nm, indicating the general absence of local major conformational changes. e. Kinetics of fiber formation was measured by ThT binding assay and ThT fluorescence of Aβ1-40 20 µM (black curve), NH₂htau 20 µM in the absence (red curve) and presence of lipid LUV TLBE 200 µM (green curve), reverse control peptide (i.e reverse) 20 µM in the absence (blue curve) and presence of lipid LUV TLBE 200 µM (cyan curve) are reported. All samples were prepared in 10 mM phosphate buffer, pH 7.4, 100 mM NaCl. Time traces reported are the average of three experiments. Kinetic curves for Aβ1-40 are classically sigmoidal-shaped indicating an ongoing aggregation growth process while no noticeable differences in ThT emission intensity over time are contextually detected for NH₂htau and reverse peptides both in solution and in presence of lipid membrane mimetics. f.-g. NH₂htau and its reverse sequence were dissolved as described in material and methods and 150pmol of both peptides were analyzed on 15% SDS-PAGE in reducing conditions. After reversible Sypro Ruby protein staining of gel(f), Western blotting analysis was performed by probing with specific 12A12 monoclonal antibody directed against the extreme N-terminal 26-36 aa of human tau protein.Cropped representative WB is shown (g).

**Figure 2. NH₂htau is adsorbed by native neuronal plasma membranes under *in vitro* physiological conditions and deeply perturbs membrane-like lipid bilayers in a time-dependent manner**
a.-b.-c.-d.-e.-f.-g.-h. Fluorescence analysis (a-b-c-d) under non-permeabilizing conditions on hippocampal neurons (DIV15) in the presence or absence of the FITC-conjugated NH₂htau. Live primary cultures were 30’exposed to FITC-conjugated NH₂htau (1µM) (green channel) and then stained with TRITC-conjugated cholera toxin subunit b (red channel). Nuclei were stained with DAPI (blue channel). Merge image shows the composition of two fluorescence channels. Untreated cultures were used as negative control (e-f-g-h). Notice that NH₂htau displays a diffuse, dot-like pattern distributed both in close proximity to and away from somatic compartment (arrowheads and arrows point to non-synaptic and synaptic areas, respectively). Structures which are positive for components of lipid rafts but not for NH₂htau (asterisks) are also present. Scale bar:5 µm. i.-j. Fluorescence analysis of purified hippocampal synaptosomal fractions in the presence or absence of the FITC-conjugated NH₂htau. Isolated synaptosomes were dual-labeled (i) by staining with FM4-64 (50µM for 1’+ 30mMKCl for 90sec) (red channel) and with FITC-conjugated NH₂htau (1µM) for 10’(green channel). Untreated synaptosomes were used as negative control (j). Overlay image (yellow, arrowheads) indicates colocalization of NH₂htau and FM 4-64 in numerous granular spots. Few ring-like morphological structures (usually 1–3 µm in diameter) resembling spherical synaptosomes are labeled by NH₂htau but not by membrane-bound FM4-64 dye .Scale bar:10 µm. k. Western blotting analysis (n=2) of isolated membrane-containing fractions from primary hippocampal neurons (15 DIV) exposed for 1h to NH₂htau and its reverse (5µM). Untreated cultures and synthetic NH₂ 26-44 human tau peptide were used as controls. Immunoblotting was performed with 12A12 monoclonal antibody (26-36 aa) and with NR1 subunit antibody, to check the preparation purity. Cropped representative WB are shown. l.-m. DSC thermograms of LUV DMPC/DMPS 7/3 freshly prepared (black curve), LUV DMPC/DMPS 7/3 + NH₂htau or reverse 20 µM each t = 0 (red curve), LUV DMPC/DMPS 7/3 + NH₂htau or reverse 20 µM t = 24h (blue curve), LUV DMPC/DMPS 7/3 + NH₂htau or reverse 20 µM t = 48h (green curve), LUV DMPC/DMPS 7/3 + NH₂htau or reverse 20 µM each t = 72h (cyan curve). Values from deconvolution analysis of the peak profiles are reported in Table 1.

**Figure 3. Acute administration of sub-toxic doses of NH₂htau induces perturbation in regulation of K⁺-evoked intracellular calcium dynamics in isolated hippocampal synaptosomes**
Changes in K⁺-evoked Ca²⁺ levels induced by NH₂htau and its reverse counterpart were assessed in Fluo-3 loaded hippocampal synaptosomes exposed to NH₂htau (1µM), its reverse peptide (1µM), and saline (CTRL) 5 min before KCl (30mM) stimulation. a.-d.-g. Upper row: wide field images of treated synaptosomes. Baseline pre-stimulus is the image immediately before the fluorescence increase towards the peak. Peak is the image showing the highest fluorescence intensity. Baseline post-stimulus is the image immediately after that the post-stimulus baseline is established. Lower row: false color image (fire palette; Spectral Difference Image) showing the fluorescence intensity of the difference image. Brighter spots correspond to functional presynaptic terminals. Difference image is the subtraction of the baseline pre-stimulus image from the peak one. b.-e.-h. Normalized traces of fluorescence intensity spatial profiles derived from the line shown in a, d, and g. Green line refers to the baseline pre-stimulus image. Red line refers to the peak image. Cyan line refers to the baseline post-stimulus image. Note the relative fluorescence height intensity of the peak image (red line) which appears spatially distributed to partially fused multiple sites. c.-f.-i. Representative traces of fluorescence time courses derived from the three experimental groups. KCl black arrows indicate the time of the stimulus addition. j. Comparison, among the three experimental groups, of the average percentages of the fluorescence intensity difference between the peak and the baseline pre-stimulus values. Note the statistically significant increase of calculated value from NH₂htau in comparison to those from the reverse and CTRL groups (one-way ANOVA followed by Bonferroni post-hoc test ***p<0,0001 versus NH₂htau). k. kinetic analysis of the fluo-3 intensity time-course: baseline to peak, i.e. time for the rate of signal rise; peak to baseline, i.e. time for the rate of signal decay; total duration, i.e. the overall time from baseline pre-stimulus to baseline post-stimulus. Data were derived from five independent experiments. In each experiment two coverslips for each experimental group were analyzed. Values are means of at least four independent recordings and statistically significant differences were calculated by one-way ANOVA followed by Bonferroni post-hoc test (**p<0,01; ***p<0,0001 versus NH₂htau). Scale bar: 20µm.

**Figure 4. NH₂htau acutely inhibits the K⁺-stimulated presynaptic vesicles release and glutamate exocytosis in purified synptosomal preparations**
a.-b.-c.-d.-e.-f.-g.-h. K⁺-induced destaining of FM1-43 dye on isolated nerve terminals from mature (15 DIV) hippocampal neurons exposed to NH₂htau (1µM), its reverse peptide (1µM), and saline (CTRL) 5 min before KCl (30mM) stimulus addition. a.-c.-e. Upper row: wide field images of treated synaptosomes. Lower row: false color vertically corresponding images (fire palette) which show the fluorescence intensity. Brighter spots correspond to functional presynaptic terminals. First is the image before KCl administration. Last is the image when the after-stimulus baseline is established. Difference is subtraction of the last image from the first one. Note that the yellow color is below the saturation level (i.e.white color). b.-d.-f. Representative traces of destaining time courses derived from the three experimental groups. g. Normalized, aligned and averaged fluorescence intensity traces derived from the three experimental groups (25 traces each) plus two saline(-KCl) additional controls. One, in the absence of both treatments and KCl stimulus, representing the fluorescence bleaching rate of our experimental setting. The other one, in the presence of NH₂htau (1µM) and with saline added, showing that NH₂htau alone is not able to induce significant destaining effects. Trend lines (black lines) superimposed to fluorescence intensity values of three experimental groups were calculated by polynomial fitting. h. Comparison of the average destaining percentage of the fluorescence intensity among the three experimental groups. In each experiment(n=5) two coverslips for each experimental group were analyzed. Values are means of at least three independent recordings and statistically significant differences were by one-way ANOVA followed by Bonferroni post-hoc test (**p<0,01versus NH₂htau). Scale bar:15µm. i. The overall dose-effect of the NH₂htau action (1-10-100nM) on glutamate release was evaluated by high-sensitive radioactive-based measure of depolarization-evoked release of [3H]D-Asp. Reverse sequence, used at the highest concentration(100nM) and saline-exposed untreated controls were also included. The K⁺-evoked tritium overlow is expressed as % of 12 mM KCl-evoked [3H]D-aspartate overflow versus saline-exposed untreated ctrl.Values are means of at least five independent experiments and data were considered statistically significant for p < 0.05 at least (**p < 0,01; ***p < 0,0001 versus saline-exposed untreated ctrl, one-way ANOVA followed by Bonferroni post-hoc test).

**Figure 5. Long-term application of NH₂htau induces a marked and selective loss of exocytotic presynaptic vesicles proteins in cultured hippocampal neurons**
a.-b. Western blotting analysis (n=12) was carried out on equal amounts of total protein extract (40µg) from mature hippocampal primary neurons (DIV15) exposed for 48h to increasing subtoxic concentration (1-2µM) of NH₂htau and its reverse control sequence. Immunoblots (a) were probed with antibodies against several presynaptic- (α-synuclein, synapsin I, synaptosomal-associated protein 25 SNAP-25, synaptophysin, vesicular glutamate transporter 1 vGLUT1 , synaptic vesicle protein 2 SV2, dynamin, synaptotagmin) and post-synaptic markers (N-Methyl-D-aspartate NMDA Receptor Subunit NR1, postsynaptic density protein 95 PSD95) and against not-synaptic proteins located in trans-Golgi network and endoplasmic reticulum (golgin-97 and calnexin). Cropped representative WB are shown. Densitometric quantification of immunoreactivity levels (b) was calculated by normalizing the values on the β-actin intensity and expressed as ratio respect to corresponding ctrl values.Values are means of at least nine independent experiments and statistically significant differences were calculated by one-way ANOVA followed by Bonferroni post-hoc test (*p < 0,05; **p<0,01; ***p<0,0001 versus untreated ctrl).

Figure 6. Distortion of the dendritic tree, microtubule breakdown and mitochondria loss occur in concomitance with decline of presynaptic proteins density in living hippocampal neurons chronically exposed to NH₂htau
a.-b.-c. Confocal microscopy analysis of double immunofluorescence carried out on mature hippocampal primary neurons (DIV15) exposed for 48h to NH₂htau and its reverse control sequence (1µM). Merge images show the overlay of the three fluorescence channels, Differential Interference Contrast (DIC; gray channel) enables the visualization of the neuritic network, DIC Merge is the composition of the three fluorescence and of the DIC channels. (a): presynaptic synaptophysin (green channel) and dendritic MAP-2 (red channel). Nuclei (blue) were stained with Hoechst 33258 (0.5 mg/ml). Arrowheads and arrows point to MAP2- positive neurites of larger and smaller caliber, respectively. (b): presynaptic α-synuclein (green channel) and neuron-specific cytoskeletal beta III tubulin (red channel). Arrowheads and arrows point to beta III tubulin-positive neurites of larger and smaller caliber, respectively. (c): presynaptic synapsin I (green channel) and mitochondrial marker COX I (red channel). Arrowheads point to synapsin I-labeled presynaptic spots and arrows point to COX I -positive mitochondrial structures. In the merge, DIC and DIC-Merge channels, arrowheads and arrows appear in opposition to give evidence to mitochondria resident at juxstaposed presynaptic sites. Asterisks mark typical punctuate structures immunoreactive for both synapsin I and COX I (yellow dots) representing mitochondria which are localized to presynaptic sites (i.e.synaptic mitochondria). Note the loss of double-stained synapsin I/COX I puncta and the decrease of juxstaposed presynaptic sites/mitochondria in the NH₂htau-treated cultures. Images are representative of at least three independent experiments. Scale bar: A=20 µm ;B-C=10 µm.

**Figure 7. NH₂htau is more extended than its reverse counterpart, hinting at different conformational flexibility**
From top to bottom of the panel are reported: scattering profiles of NH₂htau (a) and its reverse sequence (b) and the red continuous lines indicate the EOM fit of the experimental data; Kratky plots of the two peptides (c, d for NH₂htau and its reverse control sequence, respectively); Guinier plot of the two peptides (e, f for NH₂htau (a) and its reverse sequence, respectively); Number-weighted hydrodynamic radius distributions P_N(r) (g, h for NH₂htau and its reverse control sequence, respectively); time evolution of the average number-weighted hydrodynamic radius (lower panels of fig g-h for NH₂htau and its reverse sequence, respectively).

**Figure 8. Ensembles of structures populated by NH₂htau in aqueous solution differ from those by its reverse sequence**
A. EOM starting pool of conformations (gray shaded curves) and selected ensembles (filled square) for NH₂htau (a) and its reverse counterpart (b). B. contact maps (C_α–C_α distance between all pairs of residues) of four highly probable conformations, namely: a highly probable NH₂htau/reverse sequence conformation extracted from the peak at low R_G values (a-c) and high R_G value (b-d). Distances between different residues are expressed in Å (a cut-off distance at 14 Å was used). In the figure insets, a snapshot of the corresponding configuration is reported.

See this image and copyright information in PMC

Cited by

Tau Secretion: Good and Bad for Neurons.
Pernègre C, Duquette A, Leclerc N. Pernègre C, et al. Front Neurosci. 2019 Jun 26;13:649. doi: 10.3389/fnins.2019.00649. eCollection 2019. Front Neurosci. 2019. PMID: 31293374 Free PMC article. Review.
A walk through tau therapeutic strategies.
Jadhav S, Avila J, Schöll M, Kovacs GG, Kövari E, Skrabana R, Evans LD, Kontsekova E, Malawska B, de Silva R, Buee L, Zilka N. Jadhav S, et al. Acta Neuropathol Commun. 2019 Feb 15;7(1):22. doi: 10.1186/s40478-019-0664-z. Acta Neuropathol Commun. 2019. PMID: 30767766 Free PMC article. Review.
A Walk in the Memory, from the First Functional Approach up to Its Regulatory Role of Mitochondrial Bioenergetic Flow in Health and Disease: Focus on the Adenine Nucleotide Translocator.
Atlante A, Valenti D. Atlante A, et al. Int J Mol Sci. 2021 Apr 17;22(8):4164. doi: 10.3390/ijms22084164. Int J Mol Sci. 2021. PMID: 33920595 Free PMC article. Review.
Passive immunotherapy for N-truncated tau ameliorates the cognitive deficits in two mouse Alzheimer's disease models.
Corsetti V, Borreca A, Latina V, Giacovazzo G, Pignataro A, Krashia P, Natale F, Cocco S, Rinaudo M, Malerba F, Florio R, Ciarapica R, Coccurello R, D'Amelio M, Ammassari-Teule M, Grassi C, Calissano P, Amadoro G. Corsetti V, et al. Brain Commun. 2020 Apr 6;2(1):fcaa039. doi: 10.1093/braincomms/fcaa039. eCollection 2020. Brain Commun. 2020. PMID: 32954296 Free PMC article.
Pharmacophore-based models for therapeutic drugs against phosphorylated tau in Alzheimer's disease.
Pradeepkiran JA, Reddy AP, Reddy PH. Pradeepkiran JA, et al. Drug Discov Today. 2019 Feb;24(2):616-623. doi: 10.1016/j.drudis.2018.11.005. Epub 2018 Nov 16. Drug Discov Today. 2019. PMID: 30453058 Free PMC article. Review.

See all "Cited by" articles

References

1. Hanger DP, Lau DH, Phillips EC, Bondulich MK, Guo T, Woodward BW, Pooler AM, Noble W. Intracellular and extracellular roles for tau in neurodegenerative disease. J Alzheimers Dis. 2014;40:S37–45. - PubMed
1. Mohamed NV, Herrou T, Plouffe V, Piperno N, Leclerc N. Spreading of tau pathology in Alzheimer’s disease by cell-to-cell transmission. Eur J Neurosci. 2013;37:1939–1948. - PubMed
1. Clavaguera F, Grueninger F, Tolnay M. Intercellular transfer of tau aggregates and spreading of tau pathology: Implications for therapeutic strategies. Neuropharmacology. 2014;76:9–15. - PubMed
1. Sigurdsson EM. Immunotherapy _targeting pathological tau protein in Alzheimer’s disease and related tauopathies. J Alzheimers Dis. 2008;15:157–168. - PMC - PubMed
1. Pedersen JT, Sigurdsson EM. Tau immunotherapy for Alzheimer’s disease. Trends Mol Med. 2015;21:394–402. - PubMed

LinkOut - more resources

Full Text Sources
Other Literature Sources
- The Lens - Patent Citations Database
- scite Smart Citations
Miscellaneous
- NCI CPTAC Assay Portal

[1] Hanger DP, Lau DH, Phillips EC, Bondulich MK, Guo T, Woodward BW, Pooler AM, Noble W. Intracellular and extracellular roles for tau in neurodegenerative disease. J Alzheimers Dis. 2014;40:S37–45. - PubMed

[2] Hanger DP, Lau DH, Phillips EC, Bondulich MK, Guo T, Woodward BW, Pooler AM, Noble W. Intracellular and extracellular roles for tau in neurodegenerative disease. J Alzheimers Dis. 2014;40:S37–45. - PubMed

[3] Mohamed NV, Herrou T, Plouffe V, Piperno N, Leclerc N. Spreading of tau pathology in Alzheimer’s disease by cell-to-cell transmission. Eur J Neurosci. 2013;37:1939–1948. - PubMed

[4] Mohamed NV, Herrou T, Plouffe V, Piperno N, Leclerc N. Spreading of tau pathology in Alzheimer’s disease by cell-to-cell transmission. Eur J Neurosci. 2013;37:1939–1948. - PubMed

[5] Clavaguera F, Grueninger F, Tolnay M. Intercellular transfer of tau aggregates and spreading of tau pathology: Implications for therapeutic strategies. Neuropharmacology. 2014;76:9–15. - PubMed

[6] Clavaguera F, Grueninger F, Tolnay M. Intercellular transfer of tau aggregates and spreading of tau pathology: Implications for therapeutic strategies. Neuropharmacology. 2014;76:9–15. - PubMed

[7] Sigurdsson EM. Immunotherapy _targeting pathological tau protein in Alzheimer’s disease and related tauopathies. J Alzheimers Dis. 2008;15:157–168. - PMC - PubMed

[8] Sigurdsson EM. Immunotherapy _targeting pathological tau protein in Alzheimer’s disease and related tauopathies. J Alzheimers Dis. 2008;15:157–168. - PMC - PubMed

[9] Pedersen JT, Sigurdsson EM. Tau immunotherapy for Alzheimer’s disease. Trends Mol Med. 2015;21:394–402. - PubMed

[10] Pedersen JT, Sigurdsson EM. Tau immunotherapy for Alzheimer’s disease. Trends Mol Med. 2015;21:394–402. - PubMed

Save citation to file

Email citation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

Extracellular truncated tau causes early presynaptic dysfunction associated with Alzheimer's disease and other tauopathies

Affiliations

Extracellular truncated tau causes early presynaptic dysfunction associated with Alzheimer's disease and other tauopathies

Authors

Affiliations

Abstract

Conflict of interest statement

Figures

Similar articles

Cited by

References

LinkOut - more resources

Full Text Sources

Other Literature Sources

Miscellaneous

Abstract

Conflict of interest statement

Figures

Similar articles

Cited by

References

Related information

LinkOut - more resources

Full Text Sources

Other Literature Sources

Miscellaneous