FREE ELECTRONIC LIBRARY - Abstract, dissertation, book

Pages:     | 1 |   ...   | 3 | 4 || 6 | 7 |   ...   | 13 |

«Der Naturwissenschaftlichen Fakultät der Friedrich-Alexander-Universität Erlangen-Nürnberg zur Erlangung des Doktorgrades Dr. rer. nat. vorgelegt ...»

-- [ Page 5 ] --

Moreover, it can be shown analytically that the distribution of nearest obstacle distances in a random line networks follows a Rayleigh distribution even when the network is anisotropic (SI).

Figure 3.1.

8: Rayleigh distribution of nearest obstacle distances. (A) The distribution of nearest obstacle distances in simulated line networks is determined for 3 different line densities by computing the Euclidean distance to the nearest solid phase pixel (inset). Regardless of line density, the resulting probability density calculated for 105 test points randomly placed in the liquid phase (dots) follows a Rayleigh distribution (solid lines). (B) Probability densities of nearest obstacle distances calculated for an isotropic line network (black) and two anisotropic networks with cut-off angles of 40° and 55°. In all cases, the probability distribution of the nearest obstacle distance follows a Rayleigh distribution (solid lines).

–  –  –

albeit with a right shifted peak and a broader distribution compared to the isotropic network (Fig. 3.1.8 B), as expected. Therefore, the characteristic mean value rmean of the Rayleigh distribution fitted to the p(rnod) distribution of nearest obstacle distances can be taken as a measure of the average pore size of the network imaged with either CRM or CFM.

This independence of the Rayleigh distribution from fiber anisotropy offers the opportunity to reconstruct the unbiased network pore size from CRM and SHG images and to correct for the blind spot effect. It can be analytically shown that rmean,unbiased of an isotropic network and rmean,biased of the same network but with invisible fibers above a cut-off angle are related by a factor that is simply the square root of the fraction of the visible fibers (SI). For a cut-off angle of 51°, this correction factor is √cos(51°) = 0.793.

The correction factor is constant for different network densities and only depends on the

optical cut-off angle θcut of the imaging system:

(2) √ At first glance, an over-estimation of the pore size by 25% from uncorrected confocal reflection images may not seem dramatic. However, several recent reports have demonstrated that cell migration in a porous environment critically depends on pore size, with a surprisingly sharp cut-off below which cells cannot migrate [23, 24]. A 25% change in pore size around a mean diameter of 2 µm, for instance, was associated with a more than 10-fold change in transmigration efficiency [23].

Comparison with the covering radius transform

To compare the pore sizes obtained using the nearest obstacle distance method with the pore sizes obtained using the established covering radius transform (CRT) method [64], the scaling factor between the two measures was numerically determined for simulated isotropic networks (SI). The average pore radius given by the covering radius transform, rCRT, is by a factor of f = 1.82 larger than rmean,unbiased obtained with the nearest obstacle distance method.


(3) √ For random isotropic networks, the factor f relating the pore sizes obtained with the two methods is independent of the imaging setup and the cut-off angle.

Application to collagen data To test whether these analytical and numerical results also hold for real biopolymer networks, we applied our method to a set of collagen gels polymerized at different monomer concentrations imaged with the 20x dip in objective (NA=1.0). For each gel, we obtained a CRM and CFM data set and measured the distributions of nearest obstacle distances. For all concentrations imaged, the rnod distributions followed Rayleigh distributions for both CRM and CFM. As expected, the CRM data showed consistently larger pores than the CFM data (Fig. 3.1.9 A-D). With decreasing collagen concentrations, these distributions broadened and the maxima shifted to larger pore sizes. The quality of the Rayleigh fit was very high, with correlation coefficients between r2=0.987 for 2.4 mg/ml collagen gels and r2=0.997 for 0.3 mg/ml collagen gels.

The CRM data of fibrin gels and the SHG data of collagen gels also followed a Rayleigh distribution, thus further confirming that the nearest obstacle distance of random biopolymer networks is Rayleigh distributed, irrespective of imaging mode or protein composition.

We then computed rmean,biased by fitting a Rayleigh distribution (Eq. 1) to the anisotropic CRM data, and predicted the unbiased distribution of nearest obstacle distances for an isotropic network according to Eq. (2). The Rayleigh distribution predicted from the CRM data and the measured distribution from CFM data show excellent agreement for all collagen concentrations (Fig. 3.1.9 A-D), confirming the validity of our method.

We next predicted the covering radius transform pore size, rCRT, for different collagen gels from the fit of the Rayleigh distribution to the confocal reflection data according to Eq. (3). The predicted rCRT was then compared to the directly measured rCRT from the confocal fluorescence images. We found a close agreement between predicted and directly measured pore sizes for the different collagen concentrations (Fig. 3.1.9 E), confirming the validity of Eq. (3) for converting the different pore size measures.


Figure 3.1.

9: Blind spot correction. (A) From the binarized reflection data, the nearest obstacle distance is determined (green squares) and fitted with a Rayleigh distribution (grey line). Rescaling this distribution with a correction factor according to Equ.(2) yields a prediction for the distribution of nearest obstacle distance for CFM data, as indicated by the red arrows. The distribution of the nearest obstacle distances calculated from a binarized fluorescent data set obtained for the same collagen sample (blue squares) is correctly predicted by the re-scaled Rayleigh curve (this curve is not a fit to the measurements). (B-D) Repeating this procedure for different collagen concentrations from 0.3 mg/ml to 2.4 mg/ml gave excellent agreement between predicted and measured distributions. (E) Average pore size rmean, biased for different collagen concentrations calculated from reflection data (black line), blind-spot corrected (rmean,unbiased, blue dashed line) with Equ. (2), and converted to the corresponding mean CRT values (rCRT, red dashed line) with Equ.(3). All predicted values are in good agreement with the measured data from fluorescent images (solid lines). The error bars indicate the standard deviation between different fields of view (n=8) of the same samples.

Relationship between pore size and protein concentration

Our data show that collagen gels polymerized at higher monomer concentrations have smaller pore sizes, consistent with previous findings [16, 23] (Fig. 3.1.3 and Tab.3.1.2).

Moreover, by analyzing the relationship between protein concentration and pore size for both, collagen and fibrin, it is possible to address the question whether higher monomer


concentrations are stoichiometrically incorporated into the fibers, and whether the incorporated monomers contribute to increased fiber length or instead to increased fiber thickness.

We can take advantage of the fact that the scaling factor rmean of the Rayleigh function

–  –  –

If protein monomers are stoichiometrically incorporated into fibers, and if they contribute only to fiber lengthening and not to fiber thickening, we expect that the pore size increases with concentration cm according to rmean ~cm-1/2. The pore sizes determined for collagen polymerized at concentrations between 0.6 and 2.4 mg/ml and for fibrin polymerized at concentrations between 0.125 and 8 mg/ml closely follow this prediction (Fig. 3.1.10). However, for very high fibrin concentrations and accordingly small pore sizes, the relationship deviates from the expected square root dependency (Fig. 3.1.10, grey area), suggesting that pore sizes below 1 µm cannot be reliably resolved.

Collagen polymerization is known to be strongly influenced by the source of the collagen (36) but also by pH and temperature [15, 16, 75, 76], which differ between protocols from different laboratories. Previous studies [77] have reported a stronger relationships between collagen monomer concentration and pore size, according to rmean ~cm-1. Our gels were polymerized at 37° degrees and with a pH of 10, however, which leads to a faster polymerization [76] compared to the more widely used pH of 7 or lower [16, 17, 76], and therefore may have contributed to these differences.

Furthermore, it has been suggested that fluorescent labeling of collagen might affect its polymerization behavior [78]. Even though we do not see systematic pore size differences between fluorescently labeled and unlabeled collagen, our CRM-based method avoids this issue altogether.


-0.5 4 Collagen Pore size [µm] Fibrin

-0.5 0.5 1 2 4 8 Concentration [mg/ml] Figure 3.1.

10: Concentration dependence of the average pore size of fibrin (red) and collagen (black) gels. Average pore size rmean,unbiased reconstructed from 3D CRM data vs.

protein concentration. The error bars indicate the standard deviation of the mean from different collagen gels (n≥3) and different fields of view (n≥5 for each gel). Both data sets are in good agreement with a power law with an exponent of -0.5 (gey dashed line), as expected for stoichiometrically polymerizing networks where monomer addition contributes to fiber lengthening but not to fiber thickening. The grey area illustrates the range (r 1 µm) where the pore sizes cannot be reliably measured.

Collagen Pore size [µm], control Pore size [µm], glutaraldehyde concentration

0.3 mg/ml

0.6 mg/ml

1.2 mg/ml

2.4 mg/ml Table 3.1.2: Average pore size dependence on collagen concentration and glutaraldehyde treatment. Average pore size rmean,unbiased reconstructed from 3D CRM data. With increasing concentration, the pore size of the network decreases. Glutaraldehyde treatment does not change the average pore size. The error bars indicate the standard deviation of the mean from different collagen gels (n≥3) and different fields of view (n≥5 for each gel).


Relationship between pore size and polymerization temperature

As mentioned before the pore size of a self-assembling biopolymer network is crucially dependent on the surrounding conditions during the polymerization [16, 41, 79]. As such, the pH of the collagen solution and the polymerization temperature play a significant role. To characterize the influence of the polymerization temperature on our collagen gels, we varied the polymerization temperature in the incubator. All other conditions (pH of the solution, CO2 level and humidity) remained constant. However, due to the properties of our incubators, we could not set the temperature below 22°C.

The results showed that the average pore size was decreasing with an increase in temperature, which could be explained by a faster polymerization. The effect was independent of the collagen monomer concentration used in the experiments (Fig 3.1.11). For 1.2 mg/ml collagen and 2.4 mg/ml collagen gels, the decrease showed even a similar trend. However, there is no stoichiometric theory how exactly the polymerization temperature influences the network assembling. Therefore we desisted from fitting any curve to the measured data. Moreover, due to the fact that we could not pre-estimate the pore size when the polymerization temperature was changed we did not pursue an experiment where also cells would be seeded in or on those collagen gels.

1.2 mg/ml 2.4 mg/ml Pore Size [µm]

Temperature [°C] Figure 3.1.

11: Polymerization temperature dependence of the average pore size of collagen gels. Average pore size rmean,unbiased of a 1.2 mg/ml gel (blue stars) and a 2.4 mg/ml gel (red circles); reconstructed from 3D CRM data vs. the temperature during polymerization. The error bars indicate the standard deviation of the mean from different collagen gels (n≥4) and different fields of view (n≥5 for each gel). Both data sets showed a decrease in pore size with increasing polymerization temperature. However, any mathematical function could be fitted to the curves


and because no theory of dependence known so far, we resigned on fitting a function to the curves.

Relationship between pore size and a chemical cross-linker We have shown that the pore size of reconstituted collagen networks depend on the collagen monomer concentration and polymerization conditions including temperature.

Others have shown, that the pore size also depends on the pH [15, 16, 19, 51, 78].

However, the pore size of a reconstituted collagen gel should not change if the gel is treated with a cross-linker. To test this hypothesis, we polymerized eight dishes of 1.2 mg/ml collagen gels and treated four of them with the cross-linker glutaraldehyde. We used a 0.2% solution of glutaraldehyde in PBS for the cross-linking procedure, which already saturates the free binding sites on the collagen fibrils [80, 81]. After the treatment we obtained the pore size of all four of these gels. As expected we did not see a difference between the pore sizes of the treated and the untreated gels. We repeated the experiment with 3 more sample sets on different days. For all gels, regardless on which day they were produced, we did see no difference in the resulting pore size (Fig.

3.1.12, A). Repeating this experiment on different samples prepared on different days showed highly reproducible results, with a variation in pore sizes of ± 0.15 µm (Fig.

3.1.12, A, black line). Note that if a new batch of collagen solution was used the resulting pore size can vary up to ± 2 µm 1. Therefore, if a series of experiments is planned with collagen gels, enough collagen solution from the same batch should be stored.

Pages:     | 1 |   ...   | 3 | 4 || 6 | 7 |   ...   | 13 |

Similar works:

«Verfluchtes Gold Copyright by Anton Lennartz 2006 Die beste Zeit während einer Expedition erlebe ich wenn ein anstrengender Tag zu Ende geht. Man sitzt mit Freunden am Campfeuer, während man sich unterhält, genießt man ein leckeres Essen. Mond und Sterne sind klar und nah. Hin und wieder ruft ein Nachtvogel. In solchen Momenten ist mir klar, dass Gold und Erfolg hier nicht das Wesentliche, sondern nur die Zutaten sind. Das Gold liegt immer noch da. Die Erkenntnis, dass es vorerst unberührt...»

«K I S E B B KÖZLEMÉNYEK — ELŐADÁSOK A MIKROBIOLÓGIA MAGYAR MESTEREI* Emlékezés Aujeszky Aladárra születésének 100. évfordulóján írta: K A R A S S Z O N D É N E S (Budapest) Nec virtute fuit, clarisve potentius armis Quam lingua, Latium... Laurea facundis cesserunt arma togatis. (Horatius) X I X. század második felének nagy bakteriológiai felfedezései hazánkban A élénk visszhangot keltettek. Igaz, erre szellemi örökségünk is kötelezett, hiszen innen i n d u l t...»

«Alternative Services The Book of Alternative Services of the Anglican Church of Canada with the Revised Common Lectionary Anglican Book Centre Toronto, Canada © Copyright 1985 by the General Synod of the Anglican Church of Canada Published by the Anglican Book Centre, 600 Jarvis Street, Toronto, Ontario, Canada, M4Y 2J6 All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying,...»

«The Impact of Post-9/11 Airport Security Measures on the Demand for Air Travel∗ Garrick Blalock Vrinda Kadiyali Daniel H. Simon April 30, 2007 Abstract We examine the impact of post-9/11 airport security measures on air travel in the U.S. Using five years of data on passenger volume, we evaluate the effects of the implementation of baggage screening and the federalization of passenger screening on the demand for air travel. These two congressionally mandated measures are the most visible...»

«February 2016 The Center gives people of all faiths, backgrounds, and ages the space to dream, grow, and become fully alive—physically, mentally, The Chicago Southwest Suburban Community emotionally, and spiritually. Parish and Community Center Foundation Sunday services 12:30pm and 4:30pm “A place for celebration, enrichment and healing” http://www.thecenterpalos.org 708-361-3650 MISS KAY WE LOVE YOU! The first time I met Miss Kay was back in 1992 when I arrived at The Center. It was at...»

«ZOOLOGICAL MUSEUM OF MOSCOW STATE UNIVERSITY K.G. Mikhailov The spiders (Arachnida: Aranei) of Russia and adjacent countries: a non-annotated checklist Arthropoda Selecta. Supplement No. 3. KMK Scientific Press Ltd. Moscow L 2013 Mikhailov K.G.1 The spiders (Arachnida: Aranei) of Russia and adjacent countries: a nonannotated checklist. Arthropoda Selecta. Supplement No.3. Moscow: KMK Scientific Press Ltd. 262 p. Zoological Museum MGU, Bolshaya Nikitskaya Str. 6, Moscow 125009 Russia. E-mail:...»

«Gemeinsame Tagung der WKOR und der GOR Wintertagung 2016 Magdeburg, 28. 30. Januar 2016 Mit finanzieller Unterstutzung von: ¨ Wintertagung der WKOR 2016 Inhaltsverzeichnis Inhaltsverzeichnis...................................................................... 2 Stadtplan................................................................................»

«Emil-Fischer-Schule Berlin Staatliche Fachschule für Lebensmitteltechnik Fachrichtung Bäckereitechnik Technikerarbeit Unterschiede zwischen von Hand hergestellten und maschinell gerührten Sauerteigen von Tom Bennet Rötz Inhaltsverzeichnis 1 Einleitung 1.1 Allgemeiner Teil 1.2 Grundlagen der Sauerteigherstellung 1.3 Zielbeschreibung 2 Versuchsanordnung 2.1 Allgemein 2.2 Zeitlicher Ablauf 2.3 Sauerteigherstellung 2.4 Brotrezepturen und Führungsbedingungen 3 Geräte 3.1 Sauerteiganlagen 3.2...»

«FACHHOCHSCHULE KÖLN Interkulturelle Öffnung der Internationalen Jugendarbeit Gesamtbericht der wissenschaftlichen Begleitung zum Modellprojekt JiVE „Jugendarbeit international Vielfalt erleben“ Projektleitung Prof. Dr. Andreas Thimmel Autor/-innen Yasmine Chehata Katrin Riß Andreas Thimmel FACHHOCHSCHULE KÖLN Zitation: Thimmel, Andreas / Chehata, Yasmine /Riß, Katrin (2011): Interkulturelle Öffnung der Internationalen Jugendarbeit. Gesamtbericht der wissenschaftlichen Begleitung zum...»

«Würdig?? Wohl kaum!! Bibelstudie von Pastor F. M. Riley Lukas Kapitel 21, Vers 36 „Darum wacht jederzeit und bittet, dass ihr GEWÜRDIGT werdet, diesem allem zu entfliehen, was geschehen soll, und vor dem Sohn des Menschen zu stehen!“ EINFÜHRUNG Alle paar Tage bekomme ich eine Email, einen Brief oder einen Telefonanruf von jemandem, der mir obigen Bibelvers zitiert (oder falsch zitiert) und seine Hoffnung zum Ausdruck bringt, dass er „für würdig befunden wird, bei der Entrückung...»

«A HIPPOKRATÉSZI CORPUS PHTHISIOLOGIÁJA írta: D r. MIHÓCZY LÁSZLÓ H ippokratész igen sok helyen és meglehetős részletességgel fog­ lalkozik a tüdővésszel qpdtßeec. A részletes kórleírás, a leg­ finomabb árnyalatokat is figyelembe v e v ő kórismézés, a meg­ f o n t o l t prognosztikai meggondolások és pontos therápiás uta­ sítások legszebb példáit a tüdővészről szóló leírásokban adja. Tekintsük át azokat az ismereteket, amelyeket a tüdővész...»

«Zu wem sollen wir beten? Von Werner Gitt 1 Einleitung Auf der Erde leben zur Zeit 6,52 Milliarden Menschen (Stand: Januar 2006). Nach der Bibel wissen wir, dass alle Menschen Geschöpfe Gottes (nicht Kinder, wie oft fälschlich gesagt wird!) sind. Der größte Teil der Menschheit betet. Es war noch zur Zeit der Sowjetunion, da musste jeder Student dieses riesigen Landes eine Atheistik-Vorlesung besuchen. Von einem Professor für Atheistik ist bekannt geworden, dass er vor Beginn jeder Vorlesung...»

<<  HOME   |    CONTACTS
2016 www.abstract.xlibx.info - Free e-library - Abstract, dissertation, book

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.