Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “quantitative model”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,207 records · Page 67Linked to original sources

Functional capillary density: an indicator of tissue perfusion?

Functional capillary density (FCD) is one of the parameters obtained by intravital microscopy using epi-illumination of the tissue surface or trans-illumination of thin tissue layers. FCD, defined as the length of red cell-perfused capillaries per observation area (cm-1), has been used as an indicator of the quality of tissue perfusion in various animal models. Quantitative analysis of FCD in randomly selected regions of the tissue is performed by means of a computer-assisted video analysis system which allows calculation of the length of RBC-perfused capillaries. Basically, two different mathematical approaches can be employed: the first approach is based on the addition of the distances between two neighboring points (pixels) on the video screen (Pythagorean principle). The second approach uses the superimposition of a grid system that allows estimation of the capillary length by counting the number of intersections between the capillaries and the grid lines (stereological approach). The immanent error has been calculated in our laboratory to be +/- 1% with the Pythagorean and +/- 5% with the stereological method. Beside these systematic errors of computerized measurement, the individual (user-dependent) errors occurring during recognition and redrawing of the capillaries on the video image with use of a digitizing tablet are in the range of +/- 10% (intraindividual) and +/- 70% (interindividual) for the recognition and +/- 3% (interindividual) for the redrawing procedure. Our studies indicate that the errors resulting from the use of a computer-assisted calculation (Pythagorean or stereological approach) or the user-assisted redrawing of the capillaries are negligible when compared to the errors made during recognition of the capillaries on the video screen. The methods are applied for assessment of FCD in two different microcirculation models of skin muscle and pancreas yielding highly reproducible, user-independent results under physiologic conditions and the pathophysiologic conditions of ischemia-reperfusion.

Animals↗

Steady-state regulation of the human DNA mismatch repair system.

Steady-state levels of human DNA mismatch repair (MMR) transcripts and proteins were measured in MMR-proficient and -deficient cell lines by the newly developed competitive quantitative reverse transcription- polymerase chain reaction and Western analysis normalized with purified proteins. In MMR-proficient cells, hMSH2 is the most abundant MMR protein and is expressed 3 to 5 times more than hMLH1. The hMLH1 protein was expressed 1.5 to 2.5 times more than hPMS2. Steady-state levels of mRNA expression correlated well with protein expression. hMSH2-mutated LoVo cells did not express detectable hMSH3 or hMSH6 proteins. Similarly, hMLH1-mutated HCT116 cells did not express detectable hMLH1 or hPMS2 protein, whereas in hMLH1-restored HCT116+ch3 cells, hPMS2 protein was re-expressed. In hMSH6-mutated HCT15 cells, both hMSH3 protein and mRNA were increased. In SV40-transformed lung fibroblasts, all MMR mRNAs and proteins examined were expressed at levels 1.5-5-fold higher than in their nontransformed counterpart. The steady-state levels of MMR proteins indicate that substantially more hMutS proteins, which are involved in DNA mismatch recognition, are present in comparison with the hMutL proteins. Stability of hMSH3 and hMSH6 proteins appears to depend upon the presence of the hMSH2 protein, and, similarly, the stability of the hPMS2 protein depends upon hMLH1. When the hMSH6 is mutationally inactivated, hMSH3 increases by both transcriptional up-regulation and enhanced protein stability. A balanced up-regulation of all of the components was seen after viral transformation in a fibroblast model. Quantitative changes of the MMR components are a potential mechanism to modify the DNA MMR capabilities of a cell.

Cell Transformation, Viral↗

Equilibrium DNA binding of Sac7d protein from the hyperthermophile Sulfolobus acidocaldarius: fluorescence and circular dichroism studies.

The thermodynamics of the binding of the Sac7d protein of Sulfolobus acidocaldarius to double-stranded DNA has been characterized using spectroscopic signals arising from both the protein and the DNA. Ligand binding density function analysis has been used to demonstrate that the fractional change in protein intrinsic tryptophan fluorescence quenching that occurs upon DNA binding is equal to the fraction of protein bound. Reverse titration data have been fit directly to the McGhee-von Hippel model [McGhee, J., & von Hippel, P. (1974) J. Mol. Biol. 86, 469-489] using nonlinear regression. Sac7d binds noncooperatively to poly(dGdC) x poly(dGdC) with an intrinsic affinity of 6.5 x 10(6) M(-1) and a site size of 4 base pairs in 1 mM KH2PO4 and 50 mM KC1 (pH 6.8). Some binding sequence preference is noted, with the binding to poly(dIdC) x poly(dIdC) over 10-fold stronger than to poly(DAdT) x poly(dAdT). The binding is largely driven by the polyelectrolyte effect and is consistent with a release of 4.4 monovalent cations from DNA upon complex formation or the formation of 5 ion pairs at the protein-DNA interface. Extrapolation of salt back-titration data to 1 M KC1 indicates a -2.2 kcal/mol nonelectrostatic contribution to the binding free energy. A van't Hoff analysis of poly(dGdC) x poly(dGdC) binding shows that the binding enthalpy is approximately zero and the process is entropically driven. The affinity decreases slightly between pH 5.4 and 8.0. There is no significant difference between the binding parameters of recombinant Sac7d and native Sac7 proteins, indicating that methylation of the native protein has no effect on the DNA binding function. The binding of Sac7d to various DNAs leads to a significant increase in the DNA long-wavelength circular dichroism (CD) band, the intensity of which shows a sigmoidal dependence on Sac7d concentration. The sigmoidal CD binding isotherm can be quantitatively modeled by a conformational transition in the DNA that is cooperatively induced when protein monomers are bound within a given number of base pairs, ranging from zero for poly(dIdC) x poly(dIdC) to 8 or less for poly(dAdG) x poly(dCdT).

Animals↗

Pharmacokinetic analysis of 5-[18F]fluorouracil tissue concentrations measured with positron emission tomography in patients with liver metastases from colorectal adenocarcinoma.

The purpose of our study was to develop a pharmacokinetic model to quantify the intracellular 5-fluorouracil (5-FU) concentration in liver metastases, which is expected to be closely correlated to therapy response. In addition, the influence of the biomodulator folinic acid on the action of 5-FU in the metastases was investigated. After i.v. application of 5-FU labeled with the positron emitter fluorine-18 (5-[18F]FU), the kinetics of the regional 5-[18F]FU/uptake was measured dynamically with positron emission tomography over 120 min in 14 patients with a total of 27 liver metastases from colorectal adenocarcinoma. Activity-time curves were evaluated in the metastases, the normal liver tissue, as well as in the aorta and analyzed by a six-compartment model. The catabolic breakdown of 5-FU to alpha-fluoro-beta-alanine (FBAL) in the normal liver tissue was modeled to separate the catabolites from the cytostatic agent 5-[18F]FU and the active 5-[18F]fluorodeoxyuridine nucleotides. With our model, all measured activity-time courses could be described adequately with only small interindividual variations in parameters connected with liver and blood. Extrahepatic clearance of 5-FU was estimated as 0.66 +/- 0.33 liters/min, whereas the hepatic clearance was 0.52 +/- 0.25 liters/min. The Michaelis-Menten parameters describing the nonlinear conversion of 5-FU to FBAL were Km = 11.3 +/- 6.4 micromol and Vmax = 147.1 +/- 130.7 micromol/min. The maximum FBAL concentration in the liver was reached between 35 and 65 min after i.v. 5-FU infusion. The most sensitive parameters for therapy monitoring were k(in) and k(out), which characterize the transport in and out of the intracellular volume of the metastases, respectively. Tumor response can only be expected if k(in) is high and k(out) is low ("trapping"). These criteria were met by 6 of the 27 metastases, which were identical to those with high values for the area under the intracellular 5-FU concentration curve (AUC[meta,IC]5-FU). The parameters k(in) and k(out) were also used to investigate the influence of the biomodulating agent folinic acid on drug effect. Five of the six metastases that showed trapping belonged to patients who received folinic acid. With the exception of one patient, however, all patients who received folinic acid had multiple metastases, of which only one was able to trap 5-FU. Because patient response can only be expected when all metastases trap 5-FU, folinic acid showed no effect on the overall clinical response. With the quantitative modeling approach used, trapping of 5-FU can be assessed noninvasively and on an individual basis. This makes it possible to adjust the dose for each individual patient to optimize the treatment schedule.

Adult↗

Ascribed meaning: a critical factor in coping and pain attenuation in patients with cancer-related pain.

Advanced cancer is frequently associated with severe pain, and some patients suffer markedly more than others. The experience of pain is a multifactorial phenomenon involving not only sensory but also cognitive, affective, motivational, and behavioral dimensions. The latter variables were investigated within the framework of Melzack and Wall's gate control model of pain and Lazarus' coping psychology model. Quantitative measures of the McGill Pain Questionnaire, a self-report depression scale, and a coping strategy questionnaire were employed, and a face-to-face interview was conducted in which patients ascribed meaning to their pain. One hundred terminal cancer patients in the Winnipeg, Manitoba area were studied, all of whom received care at home, with one half having access to support from a hospital palliative care unit (PCU) and one half not. Analgesic medication levels were recorded and converted to a common standard. Statistical results showed that the ability to cope with the pain experience was little influenced by access to the PCU; nor was medication level a confounding variable. Variance analysis showed that the strongest impact on pain, depression, and coping scores was made by the meaning ascribed to pain by patients (p = 0.0001). The most frequently ascribed meanings were challenge (n = 30), punishment (n = 26), and enemy (n = 20). Successful pain attenuation thus depends in part on understanding the cognitive processing of pain by individual patients.

Adaptation, Psychological↗

Single-site mutations in the C-terminal domain of bacteriophage lambda cI repressor alter cooperative interactions between dimers adjacently bound to OR.

Wild-type cI repressor dimers bind with 2.5-3 kcal/mol of cooperative free energy to the tripartite right operator region (OR) of bacteriophage lambda [Johnson, A. D., et al. (1981) Nature 294, 217-223; Brenowitz, M., et al. (1986) Methods Enzymol. 130, 132-181]. Quantitative modeling has suggested that cooperativity is required for maintenance of the lysogenic state and for the efficient switch from lysogenic to lytic growth [Ackers, G. K., et al. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 1129-1133; Shea, M. A., & Ackers, G. K. (1985) J. Mol. Biol. 181, 211-230]. Cooperativity and self-association are thought to involve protein-protein contacts between C-terminal domains of the repressor molecule [Pabo, C. O., et al. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 1608-1612]. To address the importance of the C-terminal domain in mediating the cooperativity exhibited by lambda cI repressor, a number of single-site mutant candidates were screened for possible deficiencies in cooperative interactions [Beckett, D., et al. (1993) Biochemistry 32, 9073-9079; Burz, D. S., et al. (1994) Biochemistry 33, 8399-8405]. Since repressor dimerization and binding to operator sites are coupled processes, elucidation of the energetic basis of regulation in this system requires that the equilibrium dimerization constants and the intrinsic and cooperative free energies of binding be measured. In this work we evaluate the interaction of eight mutant repressors with OR DNA: Gly147-->Asp (GD147), Pro158-->Thr (PT158), Glu188-->Lys (EK188), Lys192-->Asn (KN192), Tyr210-->His (YH210), Ser228-->Arg (SR228), and Ser228-->Asn (SN228), each with an amino acid substitution in the C-terminal domain, and Glu102-->Lys (EK102) where the substitution lies in the "linker sequence" between domains. Self-assembly properties of six of these mutant repressors are presented in the preceding paper (Burz et al., 1994). In this work, the binding of mutant cI repressors to OR was examined using quantitative DNAse I footprinting. This technique monitors individual site occupancy concurrent with binding at the other sites within a multisite complex.(ABSTRACT TRUNCATED AT 400 WORDS)

Bacteriophage lambda↗

Thermodynamic and structural analysis of the folding/unfolding transitions of the Escherichia coli molecular chaperone DnaK.

The thermal unfolding of the Escherichia coli 70 kDa heat shock protein, DnaK, exhibits three well defined transitions. At pH 7.6, these transitions are centered at 45.2, 58.0 and 73.3 degrees C. High sensitivity calorimetric scans as a function of pH indicate that the folding/unfolding behavior is well described by a four-state model which includes a delta H, tm and delta Cp for each state. Calorimetric scans of a 44 kDa N-terminal proteolytic fragment show a major transition centered at 47.5 degrees C (N1) and a minor transition at 79.4 degrees C (N2). A calorimetric scan of a 23 kDa C-terminal proteolytic fragment exhibits a low temperature peak at 58.5 degrees C (C1) and a high temperature peak at 70.6 degrees C (C2). Deconvolution analysis of the low temperature peak reveals that it is actually composed of two transitions of roughly equal delta H centered at 50.4 degrees C (C1a) and 58.2 degrees C(C1b). These experiments have allowed us to assign the transitions of the intact protein as follows. The low temperature transition of DnaK can be assigned to the N-terminal region on the basis of the similarity between the delta H and tm values for the low temperature transition and those obtained for the N1 transition of the isolated N-terminal fragment. This assignment is also supported by measurements of the intrinsic fluorescence emission as a function of temperature. DnaK contains a single tryptophan localized at residue 102 in the N-terminal domain of the protein. Additionally, calorimetric scans show that the tm of the low temperature transition increases by 9.2 degrees C in the presence of excess ADP, which is known to bind to the N-terminal domain. The middle transition can be assigned to the C1a and C1b transitions of the C-terminal fragment on the basis of the similarity of delta H and tm. In the intact protein C1a and C1b form a single cooperative unit; however, the cooperative interactions between these folding/unfolding domains are disrupted in the isolated fragment. The high temperature transition of the intact protein is composed of contributions from both the N-terminal and C-terminal regions of the protein. These studies have allowed us to develop a quantitative model of the folding/unfolding behavior of DnaK.

Bacterial Proteins↗

Alpha 1-adrenoceptor subtype selectivity: molecular modelling and theoretical quantitative structure-affinity relationships.

This study constitutes a preliminary rationalization, at the molecular level, of antagonist selectivity towards the three cloned alpha 1-adrenergic receptor (alpha 1-AR) subtypes. Molecular dynamics simulations allowed a structural/dynamics analysis of the seven alpha-helix-bundle models of the bovine alpha 1a-, hamster alpha 1b-, and rat alpha 1d-AR subtypes. The results showed that the transmembrane domains of these subtypes have different dynamic behaviours and different topographies of the binding sites, which are mainly constituted by conserved residues. In particular, the alpha 1a-AR binding site is more flexible and topographically different with respect to the other two subtypes. The results of the theoretical structural/dynamics analysis of the isolated receptors are consistent with the binding affinities of the 16 antagonists tested towards the three cloned alpha 1-AR subtypes. Moreover, the theoretical quantitative structure-affinity relationships obtained from the antagonist-receptor interaction models further corroborates the hypothesis that selectivity towards one preferential subtype is mainly modulated by receptor and/or ligand distortion energies. In other words, subtype selectivity seems to be mainly guided by the dynamic complementarity (induced fit) between ligand and receptor. On the basis of the quantitative models presented it is possible to predict both affinities and selectivities of putative alpha 1-AR ligands as well as to estimate the theoretical alpha 1-AR subtype affinities and selectivities of existing antagonists.

Adrenergic alpha-1 Receptor Antagonists↗

[Possibility of using the larvae of anuran amphibians as a test model for the quantitative determination of hypophyseal hormones].

A possibility to use the larvae of ecaudate amphibia as a test model for the quantitative determination of certain hypophyseal hormones has been shown. The physiological reactions of the pigmentary cells in tadpoles of Rana temporaria can be used for a quick and exact testing of the adrenocorticotropic hormone preparations (ACTH): corticotropin and Zn-corticotropin. Evidently the method elaborated for the biological testing of hypophyseal hormones may be used in the production of the endocrinous preparations as a quick assessment of melanocyte-stimulating hormones in integral raw and sublimated hypophyses.

Adrenocorticotropic Hormone↗

Cross-bridge model of muscle contraction. Quantitative analysis.

We recently presented, in a qualitative manner, a cross-bridge model of muscle contraction which was based on a biochemical kinetic cycle for the actomyosin ATPase activity. This cross-bridge model consisted of two cross-bridge states detached from actin and two cross-bridge states attached to actin. In the present paper, we attempt to fit this model quantitatively to both biochemical and physiological data. We find that the resulting complete cross-bridge model is able to account reasonably well for both the isometric transient data observed when a muscle is subjected to a sudden change in length and for the relationship between the velocity of muscle contraction in vivo and the actomyosin ATPase activity in vitro. This model also illustrates the interrelationship between biochemical and physiological data necessary for the development of a complete cross-bridge model of muscle contraction.

Actins↗

Macroscopic growth of filamentous fungi on solid substrate explained by a microscopic approach.

A quantitative model predicting biomass growth on solid media has been developed. The model takes into account steric interactions between hyphae and tips at the microscopic level (competition for substrate and tip-hypha collisions). These interactions effect a slowing down of the hyphal, population-averaged extension rate and are responsible, at the microscopic level, for the distribution of tip orientations observed at the colony border. At the macroscopic level, a limiting value of the colony radial extension rate is attained. A mathematical model that combines hyphal branching, tip diffusion, and biomass growth was proposed to explain such behavior. Experiments using Gibberella fujikuroi were performed to validate the model; good agreement between experiments and simulations was achieved. Most parameters can be measured by simple image analysis on the peripheral growth zone, and they have clear physical meaning; that is, they correspond to properties of single, leading hyphae. The model can be used to describe two-dimensional (2D) solid media fermentation experiments under varying culture conditions; the model can also be extended to consider growth in three-dimensional (3D), complex geometry substrates.

Biomass↗

Hazard assessment of chemical carcinogens: the impact of hormesis.

The recent report of reductions in the number and area of preneoplastic hepatic lesions in response to low doses of the tumor promoter phenobarbital provides important new support for the existence of hormetic responses to carcinogens. The presence of hormetic responses to carcinogenic agents and the corollary that beneficial doses of these compounds can be determined have several implications for the bioassay and hazard assessment of carcinogens as well as for public policy regulating exposure to these agents. To be adequately sensitive to detect and quantify hormetic or other non-linear dose-response functions, current study designs must be modified to include lower doses and sufficiently large numbers of animals. Short- or medium-term animal studies are a cost-effective means of addressing these needs and have been used recently to describe a classical hormetic response to the non-genotoxic carcinogen phenobarbital. These basic changes should be supported by a continuing emphasis on mechanistic research and the development of biologically based quantitative models of toxicant action. Linking these models with physiologically based pharmacokinetic model descriptions of target dose holds the greatest promise for improving the description of the dose-response curve at low doses. These approaches are generally encouraged by the USEPA in the form of The 1996 Proposed Carcinogen Risk Assessment Guidelines. However, there remain substantial questions regarding integration of the concept of hormesis into hazard testing and public policy that require careful consideration. Herein, we explore the issues that surround testing for hormetic responses and the implications for public policy.

Animals↗

The ion channel switch biosensor.

A biosensor technology is described which provides a direct measurement for functional molecular interactions, at the surface of a tethered bilayer membrane, through the electrical transduction of chemically modified ion-channels. High sensitivity of analyte detection is achieved due to the large flux of ions transmitted through the ion channel. The biomimetic sensor surface allows the molecular recognition to be measured in complex biological matrices (such as blood and sera) without compromising sensitivity. We have used the sensor for activity and concentration measurements for a range of analytes, which include bacteria, DNA, proteins and drugs. We have a quantitative model for the biosensor performance which is described by three-dimensional molecular interactions with the membrane surface and two-dimensional molecular interactions within the tethered bilayer.

Biosensing Techniques↗

Altered peptide ligands and wild-type peptide induce indistinguishable responses of a human Th0 clone.

The response of the human influenza hemagglutinin (HA)-specific T helper clone HA1.7 to its wild-type (wt) HA307-319 peptide ligand and related altered peptide ligands (APL) was examined over a wide range of antigen concentrations. The time course of cytokine production and surface expression of CD40 ligand and CD3 was followed at the single-cell level by flow cytometry and compared with the induction of proliferation. We observed that the APL induced responses that were indistinguishable from those induced by the wt HA ligand, albeit at higher antigen densities. Moreover, the activation parameters were induced with identical kinetics. Blocking of CD4 co-ligation inhibited the recognition of the weakly stimulatory APL, but not of the wt HA307-319 peptide. Finally, in all cases the response of the T cell clone correlated with down-regulation of surface TCR/CD3 complexes. Together, these observations support a quantitative model of T cell activation.

Amino Acid Sequence↗

Dynamic imaging of perfusion in human skeletal muscle during exercise with arterial spin labeling.

MR images acquired by using an arterial spin-labeling technique showed spatial and temporal variations of perfusion in the skeletal muscle of exercising humans. Perfusion measurements made during plantar flexion exercise in normal volunteers were consistent with those obtained by traditional techniques reported in the literature. Spatial heterogeneity of perfusion values clearly delineated the various muscle groups within the lower leg. These results are interpreted in terms of a quantitative model for the perfusion signal in muscle. This method can provide a useful tool in the study of muscle physiology. Magn Reson Med 42:258-267, 1999. Published 1999 Wiley-Liss, Inc.

Arteries↗

The role of cell density in the survival of cultured cerebellar granule neurons.

The dependence for survival of cerebellar granule neurons on the cell density was examined both experimentally and theoretically. The results of batch experiments revealed that the cell survival index (CSI) was inappreciable, if cell density was below a critical level. If cell density exceeded this critical value, CSI increased with the increase in cell density. In addition, CSI was significantly increased by using a conditioned medium from the dense cultures. This suggests that not only cell density promotes survival of neurons, but also an increased concentration of growth factors produced by neurons has a direct effect on the survival of the neurons. A quantitative model describing the distribution of the growth factor at different cell densities was proposed to investigate the role of cell density in the survival of the neurons. We showed the existence of a critical level for cell density, and good agreement in the improvement of CSI was found between the theoretical prediction and the experimental result. Finally, the average concentration of growth factor necessary for cell survival based on our model was in a reasonable range compared to the practice of the addition of neurotrophic factors to the medium of cultured cerebellar granule neurons.

Animals↗

Energy isosbestic points in third-row transition metal alloys.

The total electronic energies of the six electrons per atom (e per atom) alloys W, TaRe, HfOs, and YIr and the seven electrons per atom alloys Re, WOs, TaIr, HfPt, and YAu have been calculated in the local density approximation of density functional theory. When one considers common alloy structures such as atomically ordered variants of the body-centered cubic, face-centered cubic, or hexagonally closest packed structures and plots the total electronic energy as a function of the unit cell parameter, one finds for both the six and seven electrons per atom series energetic isosbestic points. An energetic isosbestic point corresponds to a critical value of the size parameter for which all members of the 6 or 7 e per atom series of compounds have nearly identical total electronic energy. Just as in spectroscopy, where the existence of such isosbestic points is the hallmark of two compounds present in the mixture, an energy isosbestic point implies there are just two separate energy curves. For both series it is found that the total electronic energy can be viewed as the weighted sum of a purely covalent term and a purely ionic term. Two semi-quantitative models are proposed to account for these two separate energies. In the first model the total energy is viewed as the sum of the elemental structural energy plus an ionic energy based on the Born-Mayer ionic model. In the second model one considers within the confines of mu2-Hückel theory the evolution of the total electronic energy as the Coulombic Hii integrals change in value.

Journal Article↗

The role of extrahelical peptides in stabilization of collagen fibrils.

A quantitative model for fibril assembly of type I collagen was extended to include the explicit effect of extrahelical peptides. The collagen molecule was simulated by rod-like sequences to which short, rigid tails were connected by "nondimensional" flexible joints. Three collagen structures were studied: (1) intact collagen, simulated by a rod of axial ratio 200 (The axial ratio x was taken as a segment length) with two tails of length x = 1 and x = 2, respectively, appended to each end; (2) pepsin-digested collagen, simulated by one rigid segment of length 200 and one tail of length 1; and (3) pronase-digested collagen, by a single rigid segment of length x = 200. Phase equilibria of such structures were calculated, using a lattice theory of Matheson and Flory, and the relation of the polymer-solvent interaction parameter chi to the equilibrium solubility was determined. The chi for each collagen species was then related to temperature (T) and ionic strength (I), based on the approximation that local (per segment) stabilization of collagen fibrils was due to hydrophobic and electrostatic forces only. Solubility vs temperature curves for all three collagen species were computed and compared to published experimental data. From the chi factors for each species, the composite chi was resolved into components representing energetic contributions of the extrahelical peptides relative to the helix, which were interpreted in terms of hydrophobic or electrostatic interactions stabilizing the collagen fibril.

Collagen↗