Skip to main content

Factors affecting thermal stability of collagen from the aspects of extraction, processing and modification

A Correction to this article was published on 06 May 2023

This article has been updated

Abstract

Collagen, as a thermal-sensitive protein, is the most abundant structural protein in animals. Native collagen has been widely applied in various fields due to its specific physicochemical and biological properties. The beneficial properties would disappear with the collapse of the unique triple helical structure during heating. Understanding thermal stability of collagen is of great significance for practical applications. Previous studies have shown the thermal stability would be affected by the different sources, extraction methods, solvent systems in vitro and modified methods. Accordingly, the factors affecting thermal stability of collagen are discussed in detail in this review.

Graphical abstract

1 Introduction

Leather manufactured from animal skins is extensively used for industrial commodity. The quality of leather is established on the stability of collagen which is an important structural protein in animal skins and accounts for 80–85% in dermal proteins [1]. Collagen as the most abundant protein in many tissues is widely presented not only in the skins, bones, tendons, ligaments, cartilage and blood vessels of vertebrates but also in the corneum of invertebrates and some simple multicellular organisms [2,3,4]. Today, 29 members of collagen have been found and could be classified as fibril-forming collagens (I, II, III, V, XI, XXIV and XXVII) and non-fibril-forming collagens [5].

The feature of collagen is defined as containing at least one helix domain, namely triple helix. Three parallel left-hand polypeptide chains (α chains) with polyproline II-type (PPII) conformation coil about each other to form a right-handed triple helix (Fig. 1a). Collagen could be a homotrimer of three identical α chains as well as a heterotrimer of two or three different α chains. The amino acid sequence shows every third residue is glycine (Gly), which results in a Gly-X-Y repeating sequence where X is often proline (Pro) and Y is often hydroxyproline (Hyp) [6]. Type I collagen is the most major and abundant member in the collagen family. At present, the knowledge of collagen mostly comes from the studies of type I collagen [4]. As a heterotrimer of two different α chains ([α1(Ι)]2α2(Ι)), Type I collagen contains 96% triple-helical domain and less non-helical domain, as shown in Fig. 2b. Non-helical domain with loose folded conformation is attributed to the absence of the Gly-X-Y repeating consequence in the N- and C-telopeptides [9, 10]. The hydrogen bonds, including N–H (Gly) ••• O=C (X) hydrogen bond, O-H (Hyp in Y) ••• O=C (peptide backbone) hydrogen bond and water mediated hydrogen bond, among α chains stabilize collagen triple helix (Fig. 1b) [7]. Other intermolecular non-covalent bonds such as van der Waals forces, electrostatic attractions and hydrophobic bonds also play an important role in stabilizing collagen [8]. In vivo, the biosynthetic route of type I collagen is shown in Fig. 2. N- and C-propeptides of type I procollagen are cleaved by corresponding proteases to transform into tropocollagen. The tropocollagens laterally and longitudinally aggregate to microfibers by self-assembly. The microfibers are further cross-linked and eventually form macroscopic fibers and network which could be observed in tissues. As weaving into a three-dimensional architecture, the collagen fibers contribute a desirable mechanical function to skin [11].

Fig. 1
figure 1

(a) The triple helix of type I collagen [6]. (b) Hydrogen bonds of the collagen model [7]

Fig. 2
figure 2

Biosynthetic route of type I collagen in vivo [8]

Triple helix establishes excellent characteristics for collagen, such as fiber-forming property, biocompatibility and bioactivity, which make collagen widely used in leather industry, tissue engineering [12], pharmaceuticals [13], foods [14] and cosmetics [15]. Native collagen has been molded into different forms to adequate to the specific demands: fibers for skin in leather industry, members for sausage casing in foods, aqueous solution for a standard raw material in cosmetic formulations and gels, pellets, nanoparticles, scaffolds, sponges for biomedicines [16].

Collagen is premised on maintaining the stability of the triple helix for all biological applications. The hydrogen bonds would be easily destroyed after absorbing excessive heat, which lead to the conformation transition of collagen. The thermal denaturation would take place above a certain temperature. In general, the thermal stability of collagen is expressed by denaturation temperature (Td) as well as shrinkage temperature (Ts). The higher Td or Ts, the better thermal stability [17]. Comparing with native collagen, the thermal denatured products of native collagen (gelatin and collagen hydrolysate) exhibit quite another properties such as relative lower molecular weight, absence of conformation translation, loss of ability of fibril formation, faintly promoting effect on the adsorption and growth of keratinized cells [15, 18]. The occurrence of thermal denaturation should be prevented during processing, storage and application. The thermal stability of collagen has always been an issue of concern. This review mainly discusses the effects of extraction methods, sources, solvent systems and modification methods on thermal stability of type I collagen.

2 The effect of extraction methods on thermal stability

As an insoluble macromolecular structure in vivo, collagen has usually been extracted into solution in order to be further used. Native collagen with intact triple helix would be obtained by the following brief processes: pre-treating for removing the non-collagenous substances, extracting and purifying at a low temperature. The effect of the extraction methods which are commonly used or are new on thermal stability of collagen are summarized in this review.

2.1 Acid extraction and enzyme extraction

Acid and enzyme have been wildly applied to the extraction of native collagen. Dilute acid could destroy the intermolecular salt bonds and Schiff bases, enhancing the repulsive charges on the triple helix to swell collagen fibers. Collagen extracted in acid solution still retains triple helix with amino telopeptide. Acetic acid, citric acid, lactic acid and hydrochloric acid could be used to extract collagen from tissues. Acetic acid is the best ideal acid solvent due to the high yield of extraction and bacteriostatic action [19, 20]. Acid-soluble collagen (ASC) is commonly extracted using 0.5M acetic acid solution. Enzyme could break the highly crosslinked bonds that cannot be broken by dilute acid in the telopeptide non-helical domains of collagen to obtain the atelocollagen. Pepsin-soluble collagen (PSC), which is generally obtained using 0.5M acetic acid solution containing pepsin, has intact triple helix and shows low antigenicity thanks to the removal of telopeptides [21, 22]. The comparison of physicochemical properties between ASC and PSC has been studied by many researchers. The difference in denaturation temperature (ΔTd) between ASC and PSC was more than 1°C in some literatures, as listed in Table 1. Similar thermal stability shown between ASC and PSC has also been reported: ASC (35.2°C) and PSC (34.5°C) of Nile tilapia skin [30], ASC (41.58°C) and PSC (41.01°C) of Spotted golden goatfish scale [31], ASC (32.1°C) and PSC (31.6°C) of Largefin longbarbel catfish skin [32], ASC (16.1°C) and PSC (15.7°C) of Deep-Sea Redfish skin [33], ASC (31.5°C) and PSC (31.5°C) of Frog skin [34], ASC (26.6°C) and PSC (26.7°C) of Flatfish skin [35], ASC (34.23°C) and PSC (34.37°C) of Blacktip shark skin [36], ASC (39.3°C) and PSC (39.6°C) of Striped catfish skin [37].

Table 1 The difference in denaturation temperature/ΔTd > 1°C between ASC and PSC

The effect of those two methods on the thermal stability would be distinct owing to the different collagen sources, sample states and determination methods, as reported by Portier et al. [26] and Kozlowska et al. [38]. ΔTd between ASC and PSC extracted from Black drum skin was 1.6°C, whilst ΔTd between ASC and PSC of Sheepshead skin was merely 0.3°C [26]. Td of ASC was 1.5°C higher than that of PSC determined by viscosity measurement but was 4.7°C lower than that of PSC in the form of film determined by differential scanning calorimetry (DSC) [38]. However, the intra- and inter-molecular crosslinks in ASC-s was evidently richer than that in PSC-s. Additionally, ASC-s after the removal of telopeptides using the specific enzyme exhibited a relative lower rate of fiber formation and a weaker collagen gel strength [39,40,41]. The telopeptides rich in hydrophobic residues are helpful for inter-microfibrillar interaction [42]. Accordingly, telopeptides might make a contribution to the thermal stability of collagen after fiber formation.

2.2 Alkali extraction

Back in early 20th century, several researchers investigated the effect of mild alkaline treatment on collagen, because liming is an essential process in the conversion of skins to leathers. Bowes et al. [43] suggested the hydrolysis of amide groups was the main reaction with a little breakage of some peptides when insoluble collagen was treated with alkali at pH 13.0. Courts [44] further proved the progressive hydrolysis of labile peptide bonds occurred in alkaline pretreatment of insoluble collagen. People focused on remaining collagen insoluble during alkaline treatment by keeping the pH value below 13 or using swelling restrainers [45]. Little work was described about the fraction soluble in alkali until 1971 by Kemp et al. [46]. A fraction of high molecular weight was isolated from cow bone ossein with 5% (w/v) NaOH and 0.1 M NaCl. The resulting material merited the description of ‘collagen’ owing to high content of the characteristic helix observed by optical-rotation studies. The Td values of the alkaline-soluble collagen was about 4°C lower than that of ASC extracted from cow skin [46]. Type I collagen was successfully extracted from bovine skin by Hattori et al. [47] using 3% hydroxide and 1.9% monomethylamine at 4°C for 1d. The thermal stability of the alkaline-soluble collagen (38.0°C) was lower than that of ASC (40.4°C). Similar results were reported by others [48, 49]. Td of the alkaline-soluble collagen was 5°C lower than that of ASC-s and PSC-s extracted from shark skin [48]. The thermal stability of the alkaline-soluble collagen was lower than that of PSC extracted from pig skin and the Td values were 34.5 and 37°C respectively [49].

Alkali dissolved collagen through removing the telopeptides involved in crosslinkage of the molecules and could lead to the deamination of Asn and Gln into Asp and Glu. For the loss of telopeptides and the generation of the stronger electrostatic repulsion of Asp and Glu, the native collagen extracted with alkali has slightly less thermal stability comparing with ASC-s and PSC-s.

2.3 Ultrasonic assisted and microwave assisted extraction

Ultrasonic irradiation, as a safe, reliable, reproducible and environmentally friendly technology, is widely used to demonstrate physicochemical properties in low-intensity and promote emulsification, cell destruction and chemical reactions in high-intensity [50]. Cavitation as the most usual property of ultrasonic produces intense physical forces to provide sufficient energy for an extraction process [51]. Akram et al. [52] extracted acid-soluble type II collagen from chicken sternal cartilage using ultrasonic assist (950W, 20–25kHz, amplitude ɸ10 for 24min). Td of the collagen was 5°C higher than that of the un-ultrasonic treated collagen. Qu et al. [53] extracted pepsin-soluble type I collagen from bovine hide via ultrasonic assist in the optimal condition (20kHz and 90W for 20min). Td of the collagen (40.09°C) was similar to that of PSC (39.87°C). With the help of ultrasonic, native collagen has been successfully obtained in a short time with high yield [54,55,56,57]. It should be noted that an evident degradation of collagen would take place after ultrasonic treatment for a longer duration time or under a higher degree of amplitude, as reported by Kim HK et al. [58, 59]. The effects of different ultrasonic amplitudes and duration time on ASC extracted from sea bass (Lateolabrax japonicus) skin in control of the same temperature and frequency (4°C, 20kHz) were investigated. The collagens treated below 60% amplitude for 6h still exhibited a typical type I collagen structure, including α1, α2 and β chains, which were observed in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) patterns. Gradual degradation occurred as the ultrasonic time lengthened and the amplitude increased, which was observed as the rapid increase of the components and the wide distribution of molecular weight [58]. Excessive physical forces generated from cavitation might break the hydrogen bonds among collagen molecules and weaken van der Waals interaction among the polypeptide chains, leading to collagen denaturation [59].

Microwave irradiation as heating energy source or transmission medium is still a new technology to be applied to the extraction of native collagen [60]. Li et al. [61] isolated collagen with papain from pigskin by microwave pretreatment at 25°C for 2 min. The yield of collagen observably increased from 51.24% to 76.72%. The collagen with intact structure was extracted from cattle hide by Cheng et al. [62] using microwave irradiation at 37°C for 7 h as the ratio of material to liquid (0.5 mol/ml acetic acid) was 1:45 (w/v). Td of the collagen treated with microwave irradiation was 38.8°C, which was similar to that of the collagen extracted under water bath heating (38.91°C). The yield of collagen extracted by microwave irradiation was 1.5 times higher than that of traditional water bath heating. The thermal denaturation behavior of the bovine tendon collagen in acetic acid solution (5.0 mg/ml) during microwave irradiation and oil-bath heating was examined by combination of CD and polarimetry [63]. The feature of native collagen is revealed in CD spectra at positive peak around 210–230 nm and at negative peak around 190–200nm [64]. The positive peak (223nm) of collagen treated microwave irradiation disappeared at 40°C and that of oil-bath heating disappeared at 50°C. The significant change of relative optical rotation variable of collagen under microwave irradiation started at 25°C, which was 9°C lower than that of oil-bath heating (34°C). The thermal denaturation of collagen in acetic acid solution was more easily accelerated under microwave irradiation than under conventional heating. When being applied to extraction of collagen with a lower denaturation temperature, the microwave irradiation should be used with caution.

2.4 Ionic liquids pretreatment extraction

Owing to the tight weave of collagen in tissues, some chemical reagents have been used to remove non-collagenous constituents and to loosen fiber contexture for extracting collagen easily [65, 66]. Instead of traditional reagents, Ionic liquids (ILs) composed of bulky organic cations and small inorganic anions have been applied to the pretreatment of skin [67]. Choline thioglycolate (LD50-Oral-rat: 3400 mg/kg) rather than sodium sulphide (LD50-Oral-rat: 246 mg/kg) was used to remove hair from goat skin. The tensile strength and tear strength of the skin treated with choline thioglycolate were better than that of sodium sulphide, which were 22.7 MPa, 56.4 N/mm and 15.8 MPa, 40.1 N/mm respectively [68]. 1-butyl-3-methylimidazolium chloride ([BMIM][Cl]) as a fiber opening reagent was used to treat wet salted goat skin. The Ts values of the pelts treated with 0.25–1.0% [BMIM][Cl] (108–116°C) were similar to that of conventional process (112°C) [69]. 1-butyl-3-methylimidazolium tetrafluoroborate ([BMIM][BF4]) was employed in both the unhairing and fiber-opening processes of wet salted goat skin. The Ts values of the leather treated with [BMIM][BF4] and that treated with calcium hydroxide and sodium sulfide were 111 and 116 °C respectively [70]. ILs could help to open collagen fibers and take little damage to the structure of collagen, which make ILs feasible to assist in extracting native collagen. Liu et al. [71] firstly used two ILs, namely 1-ethyl-methylimidazolium dicyanamide ([EMIM] [N (CN)2]) and 1-ethylmethylimidazolium tetrafluoroborate ([EMIM][BF4]), of different concentrations (30, 50 and 70%) to pretreat calfskin at 4°C for 24h as the ILs/skin ratio was 10:1(w/w). Then PSC-s were extracted from the skins pretreated with the ILs, which were observed to always consist of α1, α2 and β chains in SDS-PAGE patterns. The extraction yield of collagen increased by about 9% after pretreating skins with the 70% ILs. The Td values of collagen decreased less than 1°C as the concentration of the ILs increased from 0 to 70%. ILs might not destroy the triple helix but affect interaction among collagen molecules at low temperature.

The native collagen with distinct properties would be obtained by different extraction methods. Assistant means could help to raise the extraction yield but would destroy the triple helical structure of collagen under an inappropriate condition. The extraction methods and assistant means could be adjusted according to practical conditions and applications.

3 The effect of collagen extracted from different sources on thermal stability

In the collagen family, the fibril-forming collagens, especially type I–III collagens, have been the most wildly used because of the relatively abundant sources. Type I collagen is the major component in skin, bone, tendon, ligament and cornea. Type II collagen mainly exists in cartilage, vitreous body and nucleus pulposus. Type III collagen which often co-distributes with type I collagen resides in the skin, blood vessels and intestines [72]. The distribution of collagen members in organisms is quite different. The choice of source rich in desired collagen is of importance. Skin of porcine, bovine and sheep with low cost, alternatives such as deer, rabbit, chick, makes up the majority of collagen sources. Tendon of the rat-tail and bovine is the most commonly sources of type I collagen used by researchers. Cartilage is mainly used for isolating type II collagen [73]. Land mammals are always the major sources of collagen owing to the high abundance sources and low cost. Recently, collagen derived from marine sources has caused more investigator concern. Collagens of marine vertebrates and invertebrates such as fish (skin, scales and frame), jellyfish, sea cucumber, starfish and sponges have been exploited. Frog skin as amphibians source is also alternative [74, 75]. Physicochemical properties of collagen would be affected by varying amino acid compositions of different collagen types and sources. Herein, hydroxyproline (Hyp) content and denaturation temperature (Td) of collagen extracted from different species are listed in Table 2 and those of different tissues from the same sources are shown in Fig. 3.

Table 2 Hydroxyproline content (residues/1000 total residues) and denaturation temperature (Td) of collagen from various species
Fig. 3
figure 3

Denaturation Temperature (Td/°C) and Hydroxyproline Content (residues/1000 total residues) of collagen from varying tissues. Sheep [28, 78], Calf [66, 76], Leather Jacket fish [96] and Pacific whiting fish [97]

Thermal stability of collagen significantly varies with different sources. A slight difference in thermal stability is presented in different tissues from the same sources. According to previous studies, the thermal stability of collagen from different living environments might be distinct. Gaill et al. [98] compared Td of the collagens extracted from annelids living in distinct habitats. Alvinella species were living directly at the vent walls with a fluctuating habitat temperature of 20 to 60°C. Paralvinella was living in the lower temperatures of 15 to 20°C. Shallow sea-water annelids inhabited constantly a cold environment (10 to 20°C). The Td values of collagens extracted from those three annelids were 45°C, 35 and 28°C respectively. The thermal stability of carp scales collagen caught in summer was 1.8°C higher than that caught in winter [99]. Td of collagen extracted from orbicular batfish in the deep-see water was 1.7°C lower than that extracted from shallow-sea water [100].

Generally, the thermal stability of collagen extracted from terrestrial animals is higher than that of aquatic animals. Imino acids, especially Hyp, were speculated to improve the thermal stability of collagen because of their high content in the collagen of terrestrial animals. The Td values of native collagens isolated from chicken tendon and embryos were demonstrated to be 15°C higher comparing with that of the procollagens isolated without hydroxyproline [101, 102]. Hyp which benefited from the ability of making hydrogen bonds was proposed to have an important and unique role in the maintenance of the native structure of collagen [103]. Miles et al. [104, 105] further stressed the importance of hydroxyproline in stabilizing the triple helix. The major thermally labile domains were identified through the analysis of the triple helix fragments in collagen. The denaturation process initiated from the hydroxyproline deficient sequences in fibril-forming type I–III collagens as well as in non-fibril-forming type IV and IX collagens. Burjanadze and his colleagues [76] analyzed the relationship between 4-Hyp content (NHyp) and Td of the collagens extracted from various species such as tissues from fish, land animals and invertebrate by using curve-fitting analytical program (Table-Curve-Jandel). The simple Eq. (Td = 69.9-3105/ NHyp) was linearized from all the data except for that of the ice fish collagens. For most collagens, the thermal stability is in the positive correlation with the content of Hyp. A few collagens of invertebrate deviate considerably from the trend. The collagen of Lumbricus terrestris (earthworm) had lower Td (22°C) but higher amount of Hyp (~17%) [106]. The collagen of Riftia pachyptila cuticle exhibited higher Td (37°C) but contained lower levels of imimo acids (~5%). 18% Thr rather than Hyp was fond in the Y position by analyzing the amino acid sequences of the collagen isolated from Riftia pachyptila cuticle. Thr was proposed to replace Hyp in the triple helix for stabilizing function [107]. The effect of the individual amino acid and various amino acid sequences on stability of collagen is difficult to achieved by direct determination of native collagen with complex structure. Crystallization of native collagen and collagen-like domains is necessary. Collagen-like peptides have been crystalized to further confirm the basic triple helix model and elucidate the hydrogen-bonding patterns in detail [108,109,110,111].

The relationship between amino acids and the thermal stability of collagen was evaluated via a “host-guest” peptide which had the following sequence -(Gly-Pro-Hyp)3-Gly-X-Y-(Gly-Pro-Hyp)4-. The Gly-Pro-Hyp repeating sequence in both ends provided a stability triple-helical environment for the guest Gly-X-Y (GXY) triplet. All of the host-guest peptides had characteristic of the collagen triple helix in CD spectrum. The melting temperature (Tm) of all different GXY was determined by CD [112, 113]. Initially, the most frequent 5 non-polar residues in collagens were evaluated by host-guest studies. The peptide where GXY = Gly-Pro-Hyp was the most stable (Td = 44.5°C). The replacement of a single Pro or Hyp dropped the thermal stability (20.7°C–39.9°C) [112]. All 20 possible amino acids in the X and Y positions were further studied. (Gly-Pro-Hyp)8 had the highest Tm value of 47.3°C. The change of identity in the X position showed a small effect on thermal stability (31.9°C–47.3°C). The replacing of Hyp in the Y position leaded to large spread of thermal stability (26.0°C–47.3°C). Arg in the Y position like Hyp is also one of the most stable residues, because Arg is involved in hydrogen bonds and hydrophobic interactions [113]. The importance of imino acids was confirmed. The effect of Hyp in the Y position on stabilization was greater than that of Pro in the X position.

For vertebrate collagen, Hyp as the form of 4R-hydroxy-L-proline (4R-Hyp) is found exclusively in the Y position to confer extra stability [114, 115]. Two controversial mechanisms of Hyp stabilizing the triple helix were explained by crystallographic evidences. One mechanism was pointed out by Brodsky and her colleagues [116, 117], that was Hyp stabilized triple helical structure by water mediated hydrogen bonds. Water bridge molecules linked the hydroxyl group of Hyp residues to a carbonyl group on the backbone of an adjacent chain via hydrogen bonds. The triple helical structure of NH2-Gly-(Gly-Pro-Hyp)4-Gly-Pro-Flp-(Gly-Pro-Hyp)3-GlyAc was less stable than that of NH2Gly-(Gly-Pro-Hyp)8-GlyAc. The replacement of Hyp with Flp was suggested to disrupt the hydration induced by Hyp, because Flp could not form strong hydrogen bonds [118]. The other mechanism was proposed by Raines et al. [119, 120]. Td of (Gly-Pro-Flp)10 was much higher than that of (Gly-Pro-Hyp)10, which were 91 and 69°C respectively. Because Flp is more electronegative comparing with Hyp, the stereo electronic inductive effect rather than water bridges was suggested to play a major role in triple helix. Hyp was proposed to stabilize the trans conformation of the imide peptide bond through stereoelectronic inductive effect, because all peptide bonds are trans. In some invertebrate and basement membranes collagens, Hyp occurred in both X and Y positions [121, 122]. Hyp in the X position was demonstrated to exhibit effects of stabilizing or destabilizing on triple helix. Td of (Gly-Hyp-Hyp)10 was 4°C higher than that of (Gly-Pro-Hyp)10. Hyp-induced stabilization in the X position attributed to the interchain dipole-dipole interactions between proximal hydroxy bonds of adjacent Hyp residues [123, 124]. Comparing with (Gly-Pro-Pro)10, the failure of (Gly-Hyp-Pro)10 to form a triple helical structure demonstrated Hyp in the X position could destabilize the triple helix [120, 125]. Since Hyp preferred the exo ring pucker, Hyp was proposed to be improper for the X position where a Pro residue adopted the endo ring pucker to stabilize a triple helix [125, 126]. Hyp-induced stabilization in the X position is supposed to occur only when the residue in the Y position is able to provide extra-stabilizing interactions. Hyp-induced stabilization in the Y position is independent of the residue type in the X position [127].

Thr replaced Hyp in the Y position to stabilize the triple helix in the cuticle collagen of vestimentiferan Riftia pachyptila. The triple-helix structure was demonstrated to achieve only after glycosylation by analyzing a scale of synthetic Gly-Pro-Thr peptides [128, 129]. Glycosylated threonine was speculated to stabilize the triple helix in a way similar to proline ring pucker, inductive effects or hydrogen bonding [130].

Imino acids as a major role in most collagens make a contribution to thermal stability. Other mechanisms of stabilization, such as Arg in the Y position and glycosylation of Thr, might work in a collagen of lacking imino acids. The effects of amino acid compositions on the thermal stability of collagen mainly came from the studies of idealized sequences. Whereas, amino acid sequences are much more complex in native collagens than that in ideal model peptides. Collagen of various sources shows a different expression of molecules to satisfy the biological function.

For the first time, collagen model fragments with natural amino acid sequences were constructed by Li and her group [131] using computer simulation. To further study the effect of Hyp content on thermal stability of collagen molecules, the fragments of grass carp collagen were selected according to the distribution of Hyp content in diverse regions of amino acid sequences. The model fragments were optimized to fit the typical triple helix. A mathematical model was carried out to estimate the hydrogen bond energy of intact collagen chains according to the model fragments. The hydrogen bond energy between the collagen chains (17.74 J/g) calculated by the mathematical model was consistent with the DSC results (17.98 J/g). The lowest energy reduction (1.247 kcal/mol) was shown in the simulated fragment of the highest Hyp content (15.6%) after heating simulation at 35°C. The collapse of the triple helix began in the regions of less Hyp content and then in the regions of high Hyp content, as shown in the simulation of structural change of model fragments during heating process (Fig. 4). The physical and chemical properties of native collagen could be further explored by constructing collagen model fragments with natural amino acid sequences.

Fig. 4
figure 4

(a) The optimal molecular model. The yellow-marked amino acid is Hyp. (b) The schematic diagram of structural changes of a collagen model during degeneration. The yellow dots represent the distribution of Hyp. Adapted with permission from [131]. Copyright © 2019, American Chemical Society

4 The effect of different solvent systems on thermal stability

Collagen is always used in form of solution or is developed with aqueous preparations. The triple helix of collagen is maintained by several interactions such as hydrogen bonds, electrostatic attractions and hydrophobic interactions. These interactions might be disturbed by the change of solvent systems, thus affecting the thermal stability of collagen. This review mainly discusses the thermal stability of collagen in different solvent systems from four aspects.

4.1 The concentration of collagen

Td of the collagen extracted from bovine skin in acid solution was observed to decrease by about 0.8°C as the collagen concentration increased from 5 to 20 mg/ml in Liu’s study, which indicated collagen molecular state might change in different collagen concentrations [132]. Three collagen concentrations were chosen according to the critical aggregation concentration (CAC) of collagen (bovine skin) in acid solution [133]. 0.25 mg/ml expressed non-aggregation state of collagen molecules. 0.5 and 1.0 mg/ml exhibited critical aggregation and complicated aggregation state respectively. Transition temperature of collagen slightly decreased by 0.3°C as collagen concentration increased from 0.25 to 1.0 mg/ml. The morphology of collagen solution changed from homogenous to inhomogeneous and the collagen fibers gradually became denser with the increase of collagen concentration. The aggregation behavior was induced by increasing collagen concentration in acid solution, which was proposed to be largely mediated through the interactions of aromatic residues. The interaction between the aromatic residues in one monomer and imino acids residues within another monomer caused the aggregation of collagen [134]. The triple-helical conformation was proposed to be buttress though ordered water networks between adjacent monomers [135]. In a higher collagen concentration, more hydrophobic microdomains were formed during the process of aggregation. Ordered water networks might be destroyed, which resulted in a slightly decrease of thermal stability. In the other hand, a higher rate of heat transfer occurred during heating process because of the shorter distance among collagen molecules in an aggregation state [133]. From a thermodynamic point of view, the aggregation is an entropy-driven process described as randomness reduction in the spatial arrangement of the collagen monomers in solution [136]. The thermal denaturation behavior of collagen could also be described as a thermodynamic process [132, 136]. Lumry-Eyring model was the most probable mechanism for estimating the thermal denaturation of collagen and showed a process of gradual unfolding of triple-helical structure during heating. Collagen existed three states, namely native triple helix state, the partially unfolded state and the denatured state, during thermal denaturation. The quantity of those three states would be changed with the varying temperature and heating rates [137].

4.2 The concentration of acetic acid

Acetic acid (AA) is regarded as the most ideal solvent in extracting and dissolving collagen. Collagen could be efficiently extracted in and homogenously dispersed into AA solution due to the electrostatic interaction or lyotropic hydration of AA. Yang et al. [138] dissolved lyophilized collagen of calf skin into 0.1~2.0 M AA solution respectively and kept collagen concentration at 0.5 mg/ml. The denaturation temperatures (Tm2) of the collagen soluble in 0.1M AA solution and in 2.0M AA solution were 42.56 and 34.76°C respectively. Tm2 was a clear negative liner correlation with the AA concentration (CAA), as described in the formula Tm2 = 44.30-1.93 × CAA. The characteristic CD spectra of native collagen exhibits a positive peak around 210–230 nm and a negative peak around 190–200nm [64]. All samples exhibited the typical CD spectra of native collagen conformation. The ratio of positive to negative peaks (Rpn) could be used to indicate the integrity of triple helical structure. The Rpn values of collagen in 0.1 M and 0.5 M AA solution almost the same, which were 0.132 and 0.133 respectively. Further increasing AA concentration, Rpn first increased to 0.145 and then decreased to 0.129. Native collagen structure remained in the AA solution of different concentrations but the conformation of collagen was affected. Additionally, the morphology of collagen in low AA concentrations (0.1–1.0 M) was observed as inhomogeneous aggregation state. Collagen morphology gradually became homogenous and the fibers get thinner as the AA concentration was above 1.0 M. Disaggregation would take place in a higher AA concentration. The repulsion of NH2+ groups would be enhanced with the increase of free H+ concentration released by increasing AA concentration. The water mediated hydrogen bonds among collagen molecules would be gradually disrupted as AA concentration increased. Intra-and inter-molecular hydrogen bonds were weakened with the increase in AA concentration, thus leading to the decrease of collagen thermal stability.

4.3 Different inorganic ion concentrations and species

In aqueous solution, ions might interact with collagen and/or water molecules to control the state of collagen. The effect of inorganic salt on protein as “salting-in” and “salting-out” have been well known. Specific ion effects were first be proposed by Hofmeister [139], which was now known as the Hofmeister series. The Hofmeister series is the ordering of ions in terms of their ability to salt out or salt in proteins as follows: SO42-> HPO42-> CH3COO-> F-> Cl-> Br-> NO3-> I-> ClO4-> SCN- and Mg2+> Li+> Na+> K+> NH4+> (CH3)4N+. Generally, kosmotropes on the left of Cl-and K+ were called “water structure makers” which make the bulk water more ordered and collagen preferentially hydrated to keep the original state of collagen. Chaotropes on the right were called “water structure breakers” which reduce the strength of intrahelical hydration and promote the denaturation of collagen via increasing the translational and vibrational frequencies of the water molecules [140]. The effect is dominated by anions which are more polarizable and strongly hydratable [141, 142]. Komsa-Penkova et al. [143] chose 12 inorganic salts to study the effect of different ion concentrations and species on thermal stability of collagen. The salts included sodium salts with Cl-, SCN-, H2PO4-, HPO42-, SO42-anions and chloride salts with Li+, K+, Na+, NH4+, Ca2+ cations. Li2SO4 and NaN3 were also measured. The salts were dissolved in 50 mM acetic acid to get different concentrations. Then the salt solutions and collagen solutions (in 50 mM acetic acid) were mixed at the ratio of 1:1 (v/v) and the final collagen concentration was kept at 0.5 mg/ml. The change of the pH values of the samples, except Na2HPO4, between 3 and 4.5 was proposed not to affect Td. At low salt concentrations of below 20 mM, Td of the all collagen solutions decreased by about 0.2°C for every 1 mM increase in concentration. At intermediate salt concentrations of roughly 20–500 mM, the anions dominated the change of thermal stability of collagen and followed the order H2PO4-≥ SO42-> Cl- > SCN-. The Td values of collagen in NaH2PO4 (20–500 mM) and in Na2SO4 (50–100 mM) slightly fluctuated as concentration increased, which were 37.6 ± 0.3 and 35.5 ± 0.1°C respectively. NaSCN and all the chloride salts decreased the thermal stability of collagen and the tendency induced by the former was more obvious. Td of collagen decreased by about 6°C as the concentration of NaSCN increased to150 mM and the same effect could be obtained in 450 mM NaCl. The increase of the Td values in Na2HPO4 might be affected by the pH value because of the significant increase of pH from 6.0 at 10 mM up to 9.6 at 500 mM Na2HPO4. SO42- anions interacted specifically with collagen. Td of collagen existed both a lower value and a higher value in Na2SO4 (100–150mM) and Li2SO4 (300–750mM) solutions, for example, the Td values were 32.2°C and 45.4°C in 150 mM Na2SO4 and were 36.3°C and 43.7°C in 300 mM Li2SO4. The reason was inferred to be the presence of a complex two-step denaturation process or two fractions of dissolved and salted-out collagens. At higher concentrations (above 500 mM), a precipitous increase of thermal stability took place due to the observable decrease of free water in collagen induced by salting-out.

4.4 Ionic liquids (ILs)

Ionic liquids (ILs) are commonly defined as consisting of asymmetrical bulky cation and symmetrical small anion. ILs generally display low melting points and some are liquid at room temperature due to a larger volume distribution of the charge of the cation and anion [144, 145]. Organic cations in ILs mainly include imidazolium, ammonium, phosphonium, cholinium and pyridinium and anions can be various organic or inorganic ions. ILs of different concentrations and types would show distinct effect on the thermal stability of collagen, as reported by Tarannum and her colleagues [146, 147]. Collagen of rat tail tendon (RTT) was treated with diethyl methyl ammonium methane sulfonate (AMS) of different concentrations. The concentration of collagen was kept at 2.7 μM (pH 4.0) after being mixed with AMS at collagen/AMS ratios from 1:0.05% to 1:10% (v/v). RTT fibers were treated by 0.05–10% AMS at 25°C for 24h. Td of the collagen solutions slightly decreased by 1°C with the increase of AMS. The Td values of collagen fibers gradually reduced from 63°C (0% AMS) to 55°C (5% AMS) and then obviously decreased to 48°C as AMS was 10%. AMS was shown not to affect the secondary structure of collagen and just to decrease thermal stability at inter-fibrillary level [146]. RTT collagen solution was added with the other two ILs, namely bis-choline sulphate (BCS) and 1-butyl-3-methyl imidazolium dimethyl phosphate (IDP), of same concentration (10%) respectively to make final collagen concentration of 1.33 μM (pH 4.0) at 4°C. The small pieces of collagen fiber were prepared after incubation at 25°C for 24h to further be determined by DCS and CD. Td of the collagen treated with BCS was 5°C higher and that of the collagen treated with IDP was 17°C lower than that of the native collagen fibers. The collagen added with BCS showed the typical CD spectrum of native collagen. However, both positive peak (222 nm) and negative peak (197 nm) completely disappeared in the CD spectrum of the collagen added with IDP [147]. Mehta et al. [148] found collagen could be dramatically stabilized by choline dihydrogen (cDHP). Collagen of RTT was added with cDHP of different concentrations from 0.05 to 10% (w/v) to keep the final collagen concentration at 0.15 mg/ml (pH 4.0). The collagen added with 0.05% (w/v) genipin was considered as positive control (CG). The solutions were incubated at 4°C overnight and then air dried on polyethylene plates to form membranes. Td of the membranes were measured by microshinkage tester. Meanwhile, the degree of crosslinking (crosslinking efficiency) of the solutions were determined using TNBS assay. The Td values of collagen treated with cDHP showed a linearly upward with the increase of cDHP concentration, which increased by 20°C as cDHP reached to 10%. Crosslinking efficiency of collagen added with 0.05% cDHP was similar to CG (about 30%). The crosslinking efficiency of collagen was remarkably increased as the concentration of cDHP above 0.05%, which reached to 89% for 10% cDHP. As a potential crosslinker, phosphate anion of cDHP was postulated to attached to cationic functional groups of collagen via electrostatic interaction to make collagen chains closer, resulting in the formation of crosslinking among collagen molecules. The aggregation behaviors of collagen molecules (calf skin) in a novel acetic acid/1-ethyl-3-methylimidazolium acetate (AA/[EMIM][Ac]) biphasic solvent was studied by Yang et al. [149]. The blends of collagen and AA/[EMIM][Ac] were kept the collagen concentration constant using the varying AA/[EMIM][Ac] ratios of 10:0 to 5:5 (v/v) where AA was 0.1 M. The denaturation temperature of collagen decreased from 42.523 to 38.138°C and the average size of collagen fibers gradually decreased from 1947.66 to 806.06 nm as AA/[EMIM][Ac] rations increased from 10:0 to 5:5. The samples exhibited the typical SDS-PAGE patterns of native collagen, including α1, α2 and β chains. The hydrogen bonds formed by the amino hydroxyl and ester oxygen groups among adjacent collagen monomers were broken in the AA/[EMIM][Ac] solutions, because two stronger hydrogen bonds were formed between Ac- and the amino hydroxyl groups and between EMIM+ and the ester oxygen groups. The thermal stability of collagen solution was weakened as [EMIM][Ac] increased.

The dissolution and thermal stability of collagen in ILs of different polarity and concentrations were investigated by Liu et al. [150]. Tow EMIM-based ILs with anions of tetrafluoroborate (BF4) and acetate (Ac) were chosen. [EMIM][BF4] had low polarity and [EMIM][Ac] had high polarity. 150 mg lyophilized collagen was added to 30g those two ILs respectively. The concentration of soluble collagen in 70% [EMIM][BF4] was just 0.02 mg/ml and that of collagen in [EMIM][Ac] increased from 1.94 for 10% [EMIM][Ac] to 3.57 mg/ml for 70% [EMIM][Ac]. Td of the insoluble collagen fibers treated with 70% [EMIM][BF4] was 61.2°C similar to the native collagen fibers (61.9°C) and that of fibers treated with 70% [EMIM][Ac] decreased to 50.6°C. The collagen fibers treated with [EMIM][BF4] were looser than the native ones, while the fibers derived from [EMIM][Ac] were much more sparsely distribution, as shown in Fig. 5. In the high polarity IL, the collagen fibers would become looser and easier mobile, resulting in the reducing of thermal stability. The thermal stability and fibril-formation of collagen (grass carp skin) in three EMIM-based ILs with anions of bromide (Br), chloride (Cl) and acetate (Ac) were reported by Zhai et al. [151]. The collagen solution was mixed with equal volume of phosphate buffer (pH 7.4) containing one of the three ILs (0.05 M) and was kept the final collagen concentration at 1 mg/ml. The native collagen solution and the solutions treated with the ILs were incubated at 30°C for 2h to be further determined by DSC. The solutions showed a typical native collagen CD spectrum. The Td values of native collagen fibers and the fibers treated with [EMIM][Br], [EMIM][Cl] and [EMIM][Ac] were 38.86, 39.47, 40.38 and 50.9°C respectively. The diameters of collagen fibers showed a gradually upward trend in the order of CH3COO-> Cl-> Br-> native collagen, which were 109.4 ± 32.6, 96.5 ± 20.8, 91.8 ± 18.8 and 75.8 ± 14.3 nm respectively. The three ILs could promote the fiber-formation and then increased the thermal stability of collagen fibers.

Fig. 5
figure 5

AFM images of the native collagen and the insoluble collagen fibers derived from 70% [EMIM][BF4] (a) & [EMIM][Ac] (b) respectively. Adapted with permission from [150]. Copyright © 2019, Springer Nature

Collagen is an amphoteric charged protein. Meanwhile, various types of ILs would exhibit a number of distinct properties such as polarity, hydrophobicity and hydrogen-bonding capability. The interaction between ILs and collagen might be highly specific and quite complex in nature. The structure and thermal stability of collagen would be vast affected by the different combination of the cations and anions in ILs.

4.5 Alcohols

In aqueous environment, collagen could fold into the most stable conformation through hydrophobic interactions among nonpolar amino acid residues. Additionally, the repulsive forces between water and nonpolar side chains of collagen could favor the ordering of water structure to stabilize the collagen [152]. The thermal stability of collagen was greatly affected by adding alcohols which included glycerol, monohydric alcohols and dihydric alcohols.

The effect of glycerol (Gly) on collagen was studied by Penkova et al. [153, 154]. 0.5 mg/ml PSC-s of calf skin (CSC), sheep skin (SSC), RTT (RTC) and human placenta (HPC) were chosen to be mixed with glycerol of different concentrations (0–4M). Td of the samples was positive linear dependent with the glycerol content. The equations were given as Td = 40.45+1.15×[Gly] (CSC), Td = 40.68+0.84×[Gly] (SSC), Td = 40.46+0.71×[Gly] (RTC), Td = 40.74+0.95×[Gly] (HPC). The collagen was proposed to preferred to interact with Gly rather than water. Every individual polypeptide α-chain of collagen was stabilized through hydrogen bonding which was formed between the hydroxyl groups of Gly and Hyp residues within two neighboring triplets [153]. The effect of Gly on collagen during denaturation was further studied. 0–3 M Urea and 0–0.15 M NaCl were chosen as the reagents of destabilizing collagen. The effect of those two reagents destabilizing collagen were reduced owing to the presence of the Gly. The dependence of Td on the urea and sodium chloride concentration in the presence of 0–3M Gly showed a set of parallel lines which were shifted upward by about 1°C per 1M glycerol [154]. The solvation shell surrounding collagen molecules were stabilized, because Gly with greater ability of forming hydrogen bonds was preferential binding with water [155]. The collagen was stabilized by Gly.

The thermal stability and aggregation behavior of collagen in Gly solution and 2-propanol solution were studied by Li et al. [156, 157]. Lyophilized collagen of calf skin was dissolved into 0.1 M acetic acid containing 0.5 –3 M Gly or 2-proanol. The final collagen concentration was kept at 1mg/ml. The transition temperature (Tm) of the collagen was positive linear dependent with the concentration of Gly but was negative linear dependent with the concentration of 2-proanol. The equations were given as Tm = 41.45+0.929×[Gly], Tm = 41.44-1.638×[2-proanol]. Additionally, the mean diameter of collagen aggregates in 2-propanol showed higher and that of in Gly exhibited lower than that of the pure collagen aggregates, as shown in Fig. 6. The intermolecular aggregation was induced by the increase of hydrophobic residues in 2-propanol solution. The aggregation behavior of collagen was hampered in Gly due to the interaction between Gly and collagen. Aliphatic alcohols could promote the solubility of hydrophobic residues of collagen into the solvent medium to weaken hydrophobic bonds, thus destabilizing the triple helix [158, 159]. The effect of monohydric alcohols with different specific dielectric constant (Ɛr) on the collagen was reported by Nezu et al. [160]. Water has the Ɛr value of above 80 and many organic compounds have smaller Ɛr values. The lower Ɛr value of alcohols, the lower affinity between water and alcohols. Methanol (n-C1OH), ethanol (n-C2OH), 1-propanol (nC3OH) and1-hexanol (n-C6OH) were chosen, whose Ɛr values were 33.0, 25.3, 20.8 and 13.0 respectively. Calfskin collagen was added with the alcohols of different concentrations (wt%) to obtain a constant collagen concentration at pH 3.0. Precipitation of collagen from solution would occur in a higher concentration of solvents [161], thus the thermal behavior of collagen was observed mainly in low concentrations of monohydric alcohols (<40%). The Td values of collagen solutions decreased with the increase of monohydric alcohols concentration. The tendency was more evident in the alcohols of a lower specific dielectric constant (Ɛr). Td of the collagen in 20% n-C1OH, n-C2OH and n-C3OH decreased by about 1, 4 and 7°C respectively. The hydrophobic interactions among collagen molecules were weakened as a more hydrophobic alcohol was added, which leaded to a lower Td. Gopinath et al. [162] investigated the effect of ethanol on the triple helix of collagen. 4 mg/ml collagen (RTT) stock solution was prepared by dissolving lyophilized collagen into 50 mM acetic acid. The final collagen solution (0.4 mg/ml) was prepared through adding appropriate volumes of ethanol of different concentrations and 50 mM acetic acid into collagen stock solution (4 mg/ml). The collagen solution gradually turned into a gel in the ethanol with concentration above 50% due to the dehydration effect of ethanol. Therefore, collagen solutions in 0–40% ethanol were further studied. Melting temperature (Tm) of the solutions was determined by CD. The Tm values of collagen gradually decreased from 40 (0% ethanol) to 34°C (40% ethanol). The characteristic CD spectrum of native collagen could be observed from the solutions. The molar ellipticity at 220nm in the presence of ethanol (10%–40%) was higher than that in the absence of ethanol. The collagen molecules became closer owing to the dehydration effect of ethanol, which could further disrupt the water network surrounding collagen monomers and water-mediated hydrogen bonds among collagen molecules. The change of molar ellipticity of collagen at 220nm in various alcohols was studied by Usha et al. [163]. RTT collagen was mixed with 0.05–0.2 M methanol and ethanol respectively. 0.2 M Gly, n-propanol and propane 1,2-diol were added into collagen solution respectively as well. The collagen of constant concentration (2 ×10-6 M) was measured by CD. The molar ellipticity at 220 nm of collagen increased with the increase in concentration of methanol and ethanol and that of the collagen in 0.2M Gly, n-propanol and 1,2-propanediol was lower than that of the pure collagen. The reason is still not clear. The collagen molecules would become closer and the water-mediated hydrogen bonds among collagen molecules might be destroyed in methanol and ethanol due to dehydrating effect. The hydrophobic residues of collagen would be more easily exposed in solvent after adding an alcohol of a lower polarity, which causes the secondary structure of collagen to become loose. Both mechanisms could make the triple helix require less energy to unfold during heating. Collagen shows a lower thermal stability in monohydric alcohols.

Fig. 6
figure 6

The mean diameter of collagen aggregates in collagen (Col), collagen-glycerol (Col-G) and collagen-2-propanol (Col-P) [157]

The effect of a number of substituted diols on the thermal stability of collagen was investigated by Hart et al. [164]. A set of diols with different hydrocarbon chains and hydroxyl position included ethylene glycol, propan-1,3-diol, propan-1,2-diol, butane-1,4-diol, butane-1,3-diol, pentane-1,5-diol and hexane-1,6-diol. Calfskin collagen solution was added with these diols respectively and kept a constant collagen concentration for further determining melting temperature (Tm) by optical rotation. Tm of the collagen in ethylene glycol and propan-1,3-diol increased and that of the collagen in the other diols decreased as the concentration of diols increased. The decreased tendency of Tm became more obviously as the hydrocarbon chains of diols lengthened. The collagen could be stabilized in the hydroxyl-terminated diols with short chain, probably because the diols similar to Gly have the ability of forming hydrogen bonds with Hyp residues of collagen. The stabilization of collagen could not be obtained in propan-1,2-diol, speculating the reason is that the two hydroxyls are too close to form “link” among collagen molecules. The thermal stability of collagen kept decreasing as the hydrocarbon chains lengthened, because the nonpolar groups which dominated in a longer chain diol promoted the exposure of hydrophobic residues within collagen chains into solvent.

5 The effect of blending with polymers on thermal stability

The disadvantages of collagen, such as low thermal stability, weak mechanical property and poor water & enzymatic degradation resistance, would limit its applications. The blending of collagen and other polymers is the easiest way to obtain collagen-based biomaterials of improved properties. Natural polymers and biocompatible synthetic polymers with desirable biological features and mechanical properties are usually used to blend with collagen. The blends are usually prepared in solution and then manufactured into hydrogels, scaffolds, sponges, nanofibers or films. The thermal stability and miscibility of the blends in solution are important for further processing. The miscibility is judged by the comparation of the specific viscosity ([ƞ] exp m) and the ideal intrinsic viscosity [ƞ] i m of the blends. The blends are compatible as [ƞ] exp m > [ƞ] i m and are incompatible as [ƞ] exp m < [ƞ] i m [165].

5.1 Natural polymers

As natural polymers with excellent biocompatible and biodegradable properties, Chondroitin sulfate (CS), Hyaluronic acid (HA), Hydroxypropyl methylcellulose (HPMC), Carboxymethyl cellulose (CMC), Chitosan (CH) and Alginic acid have been commonly used to blend with collagen. The chemical structure of those nature polymers is presented in Scheme 1. CS and HA as linear polyanions belong to the class of macromolecules known as glycosaminoglycans (GSG). Cellulose, CH and Alginic acid belong to polysaccharide with a highly ordered chemical structure. All those polymers contain abundant hydroxyl groups. Expect for HPMC and CH, the others contain negatively charged group such as sulfo and/or carboxyl groups, which easily form precipitate with positively charged collagen in acid solution due to the strong electrostatic interaction. The charge shielding is necessary to obtain a miscible binary blend. The addition of NaCl is an effective approach.

Scheme 1
scheme 1

Chemical structures of Chondroitin 4-sulfate, Chondroitin 6-sulfate, Hyaluronic acid, Heparin (set a) and Hydroxypropyl methylcellulose, Carboxymethyl cellulose, Chitosan, Alginic acid (set b)

The properties of the blends of collagen (Col) with CS or HA and the interaction between collagen and CS or HA were studied by Li ang her groups [166,167,168,169]. Lyophilized collagen and chondroitin 4-sulfate (C4S) were dissolved in the phosphate buffered saline (PBS including 10 mmol/L phosphate and 100 mmol/NaCl) at pH7.4 respectively to prepare the stock solutions. The Col/C4S blends were obtained after mixing those two stock solutions at varying Col/C4S weight ratios of 100/0–20/80. The final collagen concentration was kept at 1 mg/ml by adding suitable PBS. The solutions were incubated at 37°C for 60 min to form cofibrils. The thermal stability of cofibrils increased with the increase of the C4S content. The maximum transition temperature (Tm) of the cofibrils evidently increased from 42.7 to 53.9 at the Col/C4S ratios of 100/0 –91/9 and then reached to 61.9°C at ratio of 70/30. The increased tendency became slow at a ratio > 50/50 and Tm was 65.8°C at ratio of 20/80. The microfibrillar became large and order from small and disorder as C4S increased. Additionally, the D-periodicity of the cofibrils could be observed as ratios were 20/80 and 80/20. The self-assembly was carried out according to the original model and was accelerated after the addition of CS4, which promoted the thermal stability of cofibrils [166]. The properties of the blends of collagen and HA was also studied. Tris-HCl at pH 7.2 was used as the solvent. The blends were prepared at various Col/HA weight ratios of 8/2, 5/5 and 2/8 and then lyophilized. Td of the sponges at ratios of 8/2, 5/5 and 2/8 were 56, 61.3 and 69.5°C respectively. The morphology of the sponge at ratio of 5/5 contained the least sheet structure and exhibited the most homogeneous comparing with that of the sponges as ratios were 8/2 and 2/8 [167]. The hydrogen bonds between collagen and CS or HA was observed in the further studies using FTIR (Fourier transform-infrared) and 2D correlation FTIR analysis. 0.25 mol/L NaCl at pH 7.4 was used as solvent. The Col/CS blends were prepared at Col/CS weight ratios of 100/0–9/91. The blends were judged to be miscible, because [ƞ] exp m was always higher than [ƞ] i m. The evidence of interaction between collagen and CS came from the shift of the amide I band and the decrease of the intensity of amide II band of collagen as Col/CS ratios changed from 100/0 to 20/80. The hydrogen bond was formed between hydroxyl groups of CS and C=O group of collagen as the CS content was below 50 wt% because of the charge shielding effect of NaCl. When CS content was more than 50 wt%, the electrostatic interactions was formed between carboxyl or sulfate group of CS and amino group of lysine or guanidine group of arginine residues as well as hydrogen bonds between C=O group of CS and amino groups of collagen [168]. The similar measurement was used to analyze the interaction between collagen and HA. The blends of collagen and HA in 0.2 mol/L NaCl solution were prepared at the Col/HA weight ratios of 100/0–0/100 and then air-dried. The red-shift of amide I bond and blue-shift of amide II bond of collagen were observed from FTIR spectra, which demonstrated the existence of interaction between collagen and HA. The hydrogen bond between hydroxyl group of HA and C=O group of collagen would be formed as HA content increased from 0 to 50% and that between C=O group of HA and N-H group of collagen was generated as HA content was 50%–90% [169]. The physicochemical and biological properties of the blends might be related to the different Col/CS or Col/HA ratios at which the interactions between collagen and those two polymers were different.

Cellulose with highly hydrophilic property cannot be soluble in a common solvent owing to strong inter- and intra-molecular hydrogen bonds. As cellulose derivatives, HPMC and CMC of highly water-solubility are usually used to blend with collagen. HPMC is a typical nonionic polysaccharide and CMC is anionic. The properties of the collagen/HPMC blends were investigated by Li and her group [170,171,172]. The blend solutions were prepared by mixing 15mg/L collagen with 15mg/L HPMC solution (0.1M acetic acid as the solvent) at the Col/HPMC ration of 1/1 (w/w) and then air-dried at room temperature. The endothermic peak of the Col/HPMC film was 6°C higher than that of the collagen film. The tensile strength of the film increased by 7.1 Mpa comparing with the collagen film [170]. The thermal stability of Col/HPMC gels at different Col/HPMC weight ratios also be studied. PBS (pH 7.2) containing 100 mM NaCl was used as the solvent. The blends of collagen and HPMC were prepared at the ratios of 0/1–3/1 and the collagen concentration was kept at 1mg/ml. The blend solutions were then incubated at 37°C for 60 min. The thermal stability of hydrogels was determined by turbidity measurement at 46°C. The lower reduction of turbidity, the higher thermal stability. The values of reduction in turbidity of the hydrogels mixed with HPMC were lower than that of the native collagen (49.8%), which indicated the thermal stability of Col/HPMC was increased. The thermal stability of hydrogels exhibited the tendency of increase before decrease and reached to the highest at ratio of 0.25/1 where the value of reduction in turbidity was 24.5% [171]. The interaction between collagen and HPMC varied with the different Col/HPMC ratios. The Col/HPMC blends were prepared by mixing the two solutions (in 0.1 M acetic acid solution) at the ratios of 10/0–0/10 and kept the collagen concentration at 5 mg/ml. The blends showed incompatible as the HPMC content was above 50% under which [ƞ] exp m was lower than [ƞ] i m. The hydrogen bonds between HPMC and collagen were observed in 2D correlation FTIR spectrum and became weaker at a ratio above 7/3. The hydrogen bonds tended to be formed among HPMC molecules, which resulted in weakening of interaction between collagen and HPMC [172]. Collagen whose isoelectric point is near physiological pH could not be dissolved in neutral solution. The acylation of collagen could solve the problem through converting the amino groups of lysine residues to carboxy groups [173]. Zhang et al. [174] used the deionized water as the solvent to blend succinylated collagen (SC) and CMC at different ratios of 10/0–0/10 (v/v). The blends were incompatible ([ƞ] exp m < [ƞ] i m) as the ratios were 3/7 and 1/9. The denaturation temperature of blends increased by 0.9°C at ratio of 5/5 and then decreased by 0.2°C and 0.5°C as ratios were 3/7 and 1/9 respectively comparing with SC. The hydrogen bonds and electrostatic interactions between SC and CMC could improve the stability of the blends as the content of CMC was below 50%. The interaction between SC and CMC would gradually be weakened with the increase of CMC content due to preferential formation of hydrogen bonds among CMC molecules, which decreased the thermal stability of the blends.

CH containing free amino groups is the only alkaline natural polysaccharide. Fu et al. [175] studied the properties of collagen/CH membranes. The blends solution could be obtained by mixing 3 mg/ml collagen solution and 2% CH at virous volume ratios of 4/1–1/4 and then air-dried. Td of the membranes at Col/CH ratio of 4/1was 10°C higher and that of the membranes at other ratios was lower than that of the native collagen membrane. When CH content was above 40%, the collagen fibers became remarkably loose and CH evenly distributed around collagen fibers observed in AFM images. Sionkowska et al. [176] using X-ray diffraction detected the helix structure in the collagen/CH films gradually was lost as CH content increased. 0.5M acetic acid was used as the solvent in the study. The blends were prepared to contain different collagen contents of 0–100% and then dried in vacuum at room temperature. The solutions were miscibility, because [ƞ] exp m were always higher than [ƞ] i m. The characteristic peaks of native collagen in the solutions, which were observed by X-ray diffraction, gradually disappeared as the CH content increased. The hydrogen bonding was formed between -OH groups of CH and -NH2 groups of collagen as well as between the end groups (-COOH and -NH2) of collagen and -OH and -NH2 groups of CH. The appearance of new hydrogen boning networks was speculated to alter the collagen helical character, thus affecting the thermal stability.

Alginic acid is a linear anionic copolymer arranged as homopolymeric or heteropolymeric block. In Mitra’s study [177], alginic acid was regarded as a potential “cross-linker” for collagen. 0.5% collagen of bovine skin was added with alginic acid solution of different concentrations (PBS, pH 6.5) at the Col/Alginic acid ratio of 3/1. The blend solutions were incubated for overnight at 4°C and then air-dried at 37°C for 12h. The solution of Glutaraldehyde (1.5%) crosslinking collagen was carried out as a comparison. As alginic acid concentration was 1.5%, the melting temperature of the films gradually increased to 150°C, which was similar to the comparison (151°C) and much higher than that of the native collagen film (96.98°C). The crosslinking degree kept constant after adding alginic acid at a concentration above 1.5%. The biologically active lactone ester was proposed to be formed in alginic acid due to loss of water during film formation and then reacted with free-NH2 groups of lysine residues in collagen chains, which formed chemical crosslinks among collagen molecules to improve the thermal stability. The thermal stability decreased as alginic acid concentration was further increased. Melting temperature shifted towards left might be caused by unreacted alginic acid whose melting temperature was about 90°C.

5.2 Biocompatible synthetic polymers

Water soluble synthetic polymers with desirable biocompatibility, physiological inertia and plasticity, such as Poly-vinyl-alcohol (PVA), poly-vinyl-pyrrolidone (PVP), poly-ethylene-glycol (PEG) and poly-D, L-lactide-co-glycolide (PLGA), have broad range of applications. The thermal stability and properties of collagen blending with those polymers have been investigated. Collagen are commonly blended with PVA and PVP which have been wildly applied in various fields. The properties of Col/PVA blends were studied by Lai et al. [178]. 2.1wt% blends were obtained after mixing the collagen and PVA solution at different Col/PVA weight ratios and then lyophilized. Dynamic denaturation temperature of the blends increased from 33.0°C for 100/0 to 35.6°C for 20/80. The interaction between collagen and PVA showed the strongest as the contents of PVA were 20–30%, which was demonstrated by analysis of FTIR. The formation of hydrogen bonds between collagen and PVA would be weakened at a higher content of collagen or PVA. The miscibility of collagen and PVA was reported by Sarti and Scandola [179]. The Col/PVA complexes with different PVA contents were air-dried at room temperature to obtained the films. Td of the native collagen film was 117°C and the glass transition temperature (Tg) of PVA was 50°C. The Td values of the films showed an upward tendency with the increase of PVA content and increased by 14°C at PVA content of 70%. Two distinct peaks of the Col/PVA films could be evidently observed in DSC curves, especially when the contents of PVA were 70% and 50%, which accorded with Tg and Td of individual compounds. The blends of collagen and PVA was proposed to be partly miscible. The properties of Collagen/PVP were studied by Sionkowska [180]. 0.4 M acetic acid was used to dissolve collagen and PVP. The blends at Col/PVP weight ratios of 20/80–80/20 were demonstrated to be miscible ([ƞ] exp m > [ƞ] i m]). The first peak (T1) of DSC for collagen/PVP film (87°C) at ratio of 50/50 was higher than that of the single components (collagen was 80°C and PVP was 82°C), where T1 attributed to unfolding of triple helix for collagen and the single glass transition temperature for PVP respectively. The strong hydrogen bonds were formed between the carbonyl groups of the pyrrolidone rings in PVP and the hydroxyl and/or amino groups of collagen, which caused the collagen and PVP to be miscible. Additionally, the diameter of collagen fibrils in PVA film was larger than that in PVP at the same condition, as shown in Fig. 7. The weak interaction between collagen and PVA resulted in partially miscible, which did not disturb in collagen fibril formation. The strong interaction between collagen and PVP hampered the collagen molecules to form fibrils [181]. The thermal stability of Col/PEG sponge was studied by Xue et al. [182]. 1% collagen solution (5 mol/L acetic acid) was added with PEG-20000 of different concentrations and then lyophilized. Td of the native collagen was 68.5°C and Tg of PEG was 67.5°C. The Td values of Col/PEG films were first decreased to 59.7°C (10% PEG) and then increased to 65.1°C (50% PEG). The decrease of thermal stability was also observed in the blends of collagen and PLGA [183]. Collagen and PLGA were dissolved in 1,1,1,3,3,3-hexafluoro-2-propanol (HFP). The blends were obtained with different Col/PLGA ratios of 50/50, 35/65 and 20/80 and then used for spinning. Td of the native collagen was about 80°C and Tg of PLGA was 42°C. The Td values of complex decreased with the increase of the PLGA content. The blends were suggested to be immiscible due to the appearance of two distinct peaks of DSC curves.

Fig. 7
figure 7

Diameter of collagen fibrils in PVA (Col-PVA) and PVP (Col-PVP) films containing 1%, 3% and 5% of collagen respectively [181]

Other biocompatibility polymers have also been used to blend with collagen for biomedical applications. The complex materials are usually manufactured into the form of hydrogel, film and scaffold which have much higher thermal stability due to loss of free water. Mechanical and biological properties of the blends have always been taken a major concern. The ratio of collagen to polymer would affect the physicochemical properties of the blends and should adapt to practical demand.

6 The effect of crosslinking on thermal stability

Crosslinking as the best effective method is generally used to promote collagen’s thermal stability, mechanical property and resistance to enzymatic degradation. A crosslinking could be defined as the induction by physical or chemical method to connect the functional groups within one collagen chain to that in another chain. Herein, on the view of thermal stability, the physical and chemical crosslinking of collagen are summarized.

6.1 Physical crosslinking

Physical crosslinking is accepted as a safe, cheap and simple modification owing to no additional chemical reagents. Dehydrathermal (DHT) treatment, ultraviolet (UV) irradiation and ionizing radiation (gamma-ray and electron beam) are generally used.

DHT treatment is a commonly recommended approach of physical crosslinking, which is performed on heating dry collagen matrix under vacuum to around 100°C for 24–72 h. During DHT process, the formation of the amide bonds between amine and carboxyl groups of collagen as well as the formation of the lysino-alanine bonds between the dehydro-alanine (produced by β-elimination of serine residues) and lysine residues in collagen chains cause intra- and inter-molecular crosslinks, as shown in Scheme 2 [184]. An appropriate DHT treatment is positive to thermal stability of collagen. The spinning collagen fibers were dried under tension at room temperature for 16 h and then heated in a vacuum oven at 110°C for 72 h. The shrinkage temperature of bovine skin collagen fibers after DHT treatment increased by 12°C comparing with the uncrosslinked fibers, as reported by Weadock et al. [185]. The thermal stability of the porcine acellular dermal matrix scaffolds was also improved after DHT treatment [186]. The scaffold was obtained after lyophilizing the porcine skin which was removed the non-collagenous substances. The residual water continued to be removed at 40°C, 0.05 bar for 6 h. Crosslinking was carried out at 110°C for12 h. The Td values of the crosslinked scaffold was 6°C higher than that of uncross-linked scaffold. The thermal stability of fish skin collagen film was not insignificantly improved after DHT treatment. 0.5% silver carp collagen solution was air-dried under a laminar air-flow at room temperature. Td of the crosslinked film after heating at 105°C for 24 h under vacuum (0.05 Bar) was merely 3°C higher than that of non-crosslinked film [187]. Gorham and Light [184] found thermal degradation would take place as the heating temperature increased. 0.3% collagen solution (1–2 years old bovine skin) was lyophilized and then heated at different temperatures for 24 h under a vacuum of 11 mbar. Td of the collagen without heat treatment was 66.3°C. The Td values of crosslinked collagen increased to 66.2°C as the heating temperature was 60°C and then decreased as the heating temperature further increased. The characteristic α1, α2 and β chains of native collagen were observed in SDS-PAGE to be degraded into low molecular peptides as the heating temperature was 120°C. However, the lysino-alanine bonds (crosslink bonds) increased as the heating temperature rose. The crosslinking and denaturation occur simultaneously during DHT treatment. The parameters of DHT treatment should be adjusted according to the different sources of collagen and the state of samples. Dehydration and drying of collagen samples are necessary prior to DHT treatment to prevent degradation under the high temperature. The thermal stability of collagen would significantly increase with the decrease of intra-fibrillar water content in collagen. Schroepfer and Meyer [188] determined the denaturation temperature of bovine hide at varying humidity by DSC. The humidity was adjusted by placing a small piece of hide under saturated water vapor or in a desiccator or by directly adding water. The Td values of hide decreased from about 180°C at water content of 3.7% to about 60°C at water content of 44%. Td would not decrease as a water content was above 44% at which the collagen fibers were fully hydrated. The thermal stability of collagen could be improved by dehydration drying. Shen et al. [189] prepared a high thermal stability fish (southern catfish) collagen gel through the dehydration of ethanol. Collagen solution (5 mg/ml) was incubated at 30°C for 5 h as a control. The improved gels were obtained after being immersed in different concentrations of ethanol solutions (20–100%) for 24 h at collagen/ethanol ratio of 1/10 (w/v). Td of the gels slowly increased from 43.1°C (control) to 49.2°C (60% ethanol) and then evidently increased to 60°C (80% ethanol). Td of the gel could reach to 75.8°C in 100% ethanol solution.

Scheme 2
scheme 2

The formation of amide bond and lysino-alanine during DHT treatment [184]

UV irradiation as a non-ionizing irradiation could induce the amino acids residues (such as Trp, Phe and Tyr) of collagen to generate free radical to cause intermolecular crosslinks. Sinokowska and his group [190, 191] investigated the effect of sample state and UV irradiation time on the thermal stability of collagen. Three forms of collagen were prepared using RTT collagen. Tendon pieces were obtained from rat tail and then dried at 35°C. Collagen solution was prepared by dissolving tendon fibers into 0.04 M acetic acid. Collagen film (0.015 mm) were obtained by air-drying the collagen solution at 35°C. The samples were irradiated at room temperature for different duration time. Td of the solution was determined by viscosity measurements and Td of pieces and film were determined by DSC. Td of the native solution, tendon piece and film were 40.2, 101.9 and 110.2°C respectively. The Td values of samples decreased as the irradiation time lengthened. After irradiating 2 h, Td of the solution could not be determined due to completely denaturation. The Td values of piece and film decreased by 16.3°C and 4.3°C respectively [190]. The effect of irradiation on the thermal stability of RTT fibers in aqueous was also studied, which was carried out through immersing the RTT fibers in water and then irradiating at room temperature under air atmosphere for varying time. Td increased from 64.1°C to 67°C after irradiating for 0–3h and then deceased from 57°C to 39.3°C after irradiating time was lengthened from 20 to 66 h [191]. The thermal stability of collagen crosslinked using UV irradiation was depended on the hydration degree of sample and the dosage of incident radiation (= intensity × time).

Exposure of collagen to ionizing irradiation such as gamma ray and electron beams is the other way to crosslink collagen through free radical mediation. Usually, hydroxyl radicals generated by the amino acid residues such as Phe, His and Met residues in collagen after absorbing irradiation form intermolecular crosslinks [192]. Tyrosine residues is also proposed as a crosslinking point. Crosslinking is carried out by bonding two tyrosyl radicals to form dityrosine bonds [193]. Inoue et al. [194] prepared collagen gels by directly irradiating collagen solution in acidic condition using gamma-rays. Type I collagen of porcine was dissolved into solution at pH 3.0 to obtain acid collagen solution (0.3%). The acidic gels were obtained after irradiating with gamma-rays of different doses of 0–31 kGy (dose rate 15–16kGy/h) at room temperature in air. Neutral collagen solution was obtained by adjusting the pH value with PBS (pH 7.4) and then incubated at 37°C for 30 min. The neutral gels were irradiated at same condition as acidic gels. Acidic gels were transparent and highly shrank while neutral gels kept original state as the irradiation dose increased. Crosslinking rather than degradation was proposed to occur in acidic gels under the low dose of gamma-rays irradiation (1.4 and 2.1 kGy), because the α1 and α2 chains of the acidic gels molecules shifted to a higher molecular weight and there were not any fragmented collagen with lower molecular weight observed in SDS-PAGE. Acidic gels were gradually degraded as the irradiation dose further increased. Collagen gels prepared in aqueous solution by gamma-irradiation were studied by Zhang et al. [195]. 1.54% type I collagen solution (RTT, Beijing Kelaode, pH6.2) was directly irradiated with different dose (dose rate was 20 kGy/min) at room temperature to form collagen gels. The collagen gels were transparent and kept almost same size as irradiation doses were 10–20 kGy. The Td values of collagen gels irradiated with gamma-rays were higher than that of un-irradiated gel (66.4°C), which were 83.5 (5 kGy), 73.6 (10 kGy), 71.7 (15 kGy), 81.1 (20 kGy) and 89.8°C (25 kGy) respectively. Jiang et al. [196] used electron beams to modified calfskin collagen membrane which was obtained by drying self-assemble collagen fibers. The membranes were irradiated by electron beams at low temperature in nitrogen atmosphere with different irradiation dose or without. The hydrothermal and xerothermic shrinkage temperature of the samples were determined by micro-shrinkage tester. The shrinkage temperature of the membranes increased with the increase of the irradiation dose and were about 15°C (100 kGy) higher than that of the un-irradiated membrane.

Obviously, physical crosslinking is affected by the state of sample, temperature, radiation dose, pH value and atmosphere condition. Crosslinking induced by physical method is always together with degradation which should be prevented by fine tuning of physical processes.

6.2 Chemical crosslinking

Chrome, plant tannin and aldehyde have been wildly used in tannage to obtain leathers of various desirable properties such as high hydrothermal stability, favourable biological inertness and excellent mechanical property. Nowadays, more chemical regents have been used to crosslink native collagen or collagenous materials for versatile applications. Chemical crosslinking is carried out via introducing functional groups into collagen molecules. Metal complex compounds, which are mainly applied to tannage for leather industry, stabilize collagen by forming coordinative bonds with the carboxylate groups of collagen chains. Plant polyphenols, as one of the commonly tanning agents as well as a crosslinker, crosslink collagen mainly through forming multipoint hydrogen bonds. Most of other chemical reagents crosslink collagen through different crosslinked mechanisms, which are mainly involved in the Ɛ-amino groups of collagen.

6.2.1 Metal complex compounds

Metal complex compounds crosslink collagen through coordinating with the carboxylate groups of collagen. Chromium (III) as the most commonly used tanning agent could remarkably improve the hydrothermal stability and chemical properties of leather. Ts of chrome-tanned leather could easily achieve to 110°C or higher. At least two carboxylate groups of collagen replace ligands (mainly water) of the polynuclear chrome complex to form stable coordination bonds, as shown in Scheme 3a [197]. At the same condition, the Ts values of the leathers tanned with chrome, zirconium and aluminum were102, 83 and 72°C respectively [199]. Chromium is not suitable to crosslink collagen for other applications, because chromium is cytotoxic and might turn into hexavalent chromium of highly toxic. Zirconium, which is only second to chromium to promote the hydrothermal stability of leather, has good biocompatibility and low toxicity [200]. Therefore, zirconium is a candidate crosslinker to improve the properties of collagenous materials used in biomedicine. Zirconium is easily hydrolyzed and only stable at a low pH (< 3). Masking agents are usually added to modified the properties of zirconium complex. Liu et al. [201] used zirconium complex modified by sodium citrate to crosslink calfskin collagen sponge which was obtained after lyophilizing collagen solution. Taking the denaturation temperature (by DSC) as the index, the optimal crosslinked condition was carried out: the molar ratio of sodium citrate/ zirconium (IV) was 0.18, crosslinked time was 2 h, the concentration of sponge in complex solution was 0.01 mol/L, temperature was 37°C. Td of the crosslinked sponge was 36.5°C higher than that of the uncross-linked sponge.

Scheme 3
scheme 3

Schematic illustration showing the reaction of chromium (a) [197] and plant polyphenols (b) with collagen respectively [198]

6.2.2 Plant polyphenols

Plant polyphenols, also known as vegetable tannins, were firstly applied in leather industry. As a biomaterial, plant polyphenols have been used to modify collagen for more extensive applications. Multipoint hydrogen bonds could be formed between phenolic hydroxyl groups of polyphenols and hydroxyl, carboxyl, amino or amide groups of collagen chains, as shown in Scheme 3b [198]. He et al. [202] used procyanidin to crosslink collagen in solution. Collagen solution and procyanidin solution were mixed at different procyanidin/collagen ratios of 0–8% (w/w) at pH 3.0. The collagen concentration was kept at 2.5 mg/ml. Then the crosslinked mixtures were dried at room temperature for 1 week. Td of the native collagen films was 65.5°C and melting temperature of procyanidin was 70.0°C. The Td values of crosslinked films first increased and then decreased. Td of the film at procyanidin content of 4% was the highest (83.3°C) and then reduced to 74.3°C as procyanidin content was 8%. Madhan et al. [203] crosslinked RTT fibers through soaking the RTT fibers obtained after the removal of non-collagenous substances into the catechin solutions of different concentrations at 27°C for 27 h. The hydrothermal stability of the samples was determined by micro-shrinkage tester. The Ts values of the crosslinked fibers increased with the increase of catechin concentration and increased by 9°C as 20 mM catechin was added. Catechin with crosslinking effect was also used as natural dyestuff to dye collagen membrane [204]. Catechin contains a large number of benzene rings and phenolic hydroxyl groups which are easily oxidized to form quinonoid chromophoric groups and also easily form conjugated fused rings to make polyphenols appear color. The best properties of membrane could be obtained after agitating in 2% catechin for 3h at 40°C (pH 4). Td of the membrane was 18°C higher than that of the native collagen membrane and the tensile strength of the membrane increased by 1.26 MPa after being crosslinked.

6.2.3 Other crosslinkers

The functional groups of most crosslinkers such as aldehydes, genipin, multifunctional epoxy compounds and N-hydroxysuccinimide (NHS)/Polycarboxylic acids could directly react with free amino groups within different collagen chains to form crosslinking. The crosslinking induced by 1-ethyl-3(3-dimethylaminopropyl)-carbodiimide (EDC) and diphenylphosphorylazide (DPPA) begins with the activation of the carboxyl groups in collagen chains to promote the formation of amido bonds between adjacent carboxyl and amine groups of collagen. Transglutaminase (TGase) catalyzes an acyl-transfer reaction mainly involved in glutamine and lysine residues of collagen.

Aldehydes crosslink collagen via forming Schiff bases between the Ɛ-amino groups of lysine or hydroxylysine residues and the aldehyde groups of aldehydes, as shown in Scheme 4a. Glutaraldehyde (GTA) as a typical crosslinker could remarkably enhance the thermal stability of collagen. Different states of collagen crosslinked using GTA have been investigated. 1g dermal layer, which was obtained after removing non-collagenous substances of sheepskin, was crosslinked in 100 ml 0.5% GTA solution for 1h at room temperature (pH 7.4). The Ts values of the crosslinked and the uncross-linked dermal layers were 78 and 56°C respectively [212]. Calfskin collagen solutions were crosslinked using GTA at different GTA/collagen ratios (w/w) for 24h at room temperature (pH 4.0). The Td values of the uncross-linked collagen were 39.7°C and that of crosslinked solution using 1% GTA was 40.3°C. The thermal stability of the collagens and the degree of crosslinking tended to flatten out as GTA content further increased [213]. Collagen gels were prepared by incubating the collagen solutions (PBS, 100 mmol/L NaCl, pH 7.4) at 37°C for 3 h, and then crosslinked using GTA at different molar ratios of [CHO]/[NH2] where CHO was aldehyde groups of GTA and NH2 was Ɛ- amino groups of lysine or hydroxylysine residues. Td of the uncross-linked gel was 47.1°C. The Td values observably increased to 63.9°C as [CHO]/[NH2] was 0.6 and then levelled off as [CHO]/[NH2] was above 3 [214]. The improvement of thermal stability was significantly obtained when low dosage of GTA was added. The majority of reactive amino groups in collagen could form intra- and inter-molecular multipoint crosslinks with aldehyde groups of GTA. The thermal stability of collagen levels off due to the formation of single-point bonding as the dosage of GTA is over high [212]. Despite GTA-treated materials such as bioprosthesis valves and grafts have been applied to clinical, an undesirable side-effect of toxicity is an issue. Recently, dialdehyde compounds derived from natural biomass with good biodegradability and low toxicity have attracted increasing attention. The thermal stability of collagen have been demonstrated to be improved by dialdehyde compounds such as oxidized glycogen, oxidized chitosan oligosaccharide, dialdehyde cellulose, oxidized sodium alginate and dialdehyde carboxymethyl cellulose (DCMC) [215,216,217,218]. A collagen cryogel was prepared by Tan et al. [219] using dialdehyde carboxymethyl cellulose to crosslink collagen in solution at different DCMC/Col weight ratios of 0.0–0.2% with constant collagen concentration of 5 mg/ml. Td of the cryogels increased by 22°C at ratio of 0.1%, and then decreased slightly at ratio of 0.2% where insoluble complexes were found. The complex solution of collagen and the dialdehydes might be not stable enough due to large molecular weight as well as the strong electrostatic attraction between collagen and the dialdehydes. A more stable hydrogel of collagen and DCMC with a higher concentration was prepared by Yang et al. [205] using the biphasic acetic acid/1-ethyl-3methylimidazolim acetate solvent system. The Col/DCMC solution with constant DCMC content contained different collagen concentrations of 5–25 mg/ml. The Td values exhibited a linearly upward tendency with the increase of collagen content. The gels of all the collagen concentrations were transparent and homogenous.

Scheme 4
scheme 4

Crosslinking mechanisms of collagen with Aldehydes (a) [205], Genipin (b) [206], Epoxy compounds (c) [207], NHS-activated carboxylic acid esters (d) [208], EDC/NHS (e) [209], DPPA (f) [210] and Transglutaminase (g) [211] respectively

Genipin is an iridoid derivative. The C-atom of genipin is firstly under a nucleophilic attack from primary amine groups of amino acids residues (Lys, Hyl, Arg), and then tertiary nitrogen of collagen embeds into the six-membered ring of genipin to replace an oxygen atom, finally a nitrogen-iridoid could be formed. Subsequently, two iridoids are linked via a radical reaction to form intra- and inter-molecular crosslinks (Scheme 4b) [206]. The report of fixing bovine pericardium using genipin came from Sung et al. [220]. The bovine pericardium was soaked in 0.625% genipin solution for 3 days at room temperature (pH 7.4). Ts of the fixed pericardium increased by about 13°C comparing with the unfixed pericardium (63.9°C). Storage stability of genipin-crosslinked and GTA-crosslinked collagen was reported by Sung et al. [221]. Porcine pericardia prepared by removing non-collagenous substances was fixed using 0.625% genipin or GTA in buffer solution (pH 7.4) at 37°C for 3 days. The samples were stored in a sterilized water after sterilizing. The Td values of GTA-fixed pericardia was 86.5°C and then began to decrease at 3 months after storage. Td of genipin-fixed pericardia was 80.5°C and kept constant until 6 months after storage. Additionally, no apparent cytotoxicity was observed for both the genipin-fixed tissue and storage solution during storage process when compared to GTA-fixed tissue.

Multifunctional epoxy compounds have strong reactivity due to three-membered cyclic ether groups. Epoxy groups of epoxy compounds could form multipoint crosslinking among collagen molecules with the Ɛ-amino groups of lysyl residues in collagen at alkali condition (Scheme 4c) [207, 222]. The crosslinking between collagen and epoxy compounds have been demonstrated to be remarkably affected by pH [222]. 1 g sheep dermal collagen prepared by removing non-collagenous substances was immersed in 100 mL buffered solution containing 4 wt% 1,4-butanediol diglycidyl ether at the different pH values of 4–12 (20°C). At pH 7, Ts of the crosslinked collagen showed the least increase comparing with the crosslinked collagens under other pH values. Under acid condition (pH < 7), amine group content of collagen kept constant and Ts slightly increased with the decrease of the pH values. At pH = 8–10, Ts exhibited a lineal upward tendency accompanying with the evidently decease of amine group contents. At a higher pH (>10), both Ts and amin group content were kept constant. Additionally, Ts of the crosslinked collagen in alkali condition were always much higher than that in acid condition. Shen et al. [223] pointed out the glycerol polyglycidyl ether exhibited good uniform distribution of crosslink density throughout the porcine aortic valves. Porcine aortic valves were immerged in 4% glycerol polyglycidyl ether solution for 6 days at room temperature. Hydrothermal stability of distinct layers (outer, middle, inner and entire wall as well as leaflet) in aortic valves was almost same and was 20°C higher than that of the uncross-linked aortic valve.

The carboxyl groups of polycarboxylic acids are turned into an active ester groups by NHS under the action of EDC or 1,3-dicyclohexylcarbodiimide catalyst. Crosslinking would be formed between the active ester groups and the amino groups of collagen molecules, thus enhancing the thermal stability of collagen (Scheme 4d). Citric acid, adipic acid, gallic acid, poly (γ-glutamic acid) and poly (α,β-malic acid) have been chosen into system of NHS-activated carboxylic acid esters [208, 224,225,226,227,228,229,230]. The properties of collagen crosslinked using NHS activated adipic acid (NHS-AA) were studied by Li and her group [208, 224,225,226,227]. Alkali-solubilized collagen of bovine hide and PSC of fish skin were dissolved in phosphate solution buffer (pH 6.3) to make collagen concentration at 0.3% (w/w). Two collagen solutions were mixed with NHS-AA of different concentrations. Td of the alkali-solubilized collagen reached to the highest (40.8°C) at NHS-AA concentration of 1.5 mM, which was 4.2°C higher than that of the uncross-linked collagen. Td of PSC increased by 13.3°C at NHS-AA concentration of 1.05 mM comparing with the native collagen [208, 224]. The decrease of amino groups in alkali-solubilized collagen due to the deamination of Asn and Gln into Asp and Glu caused less crosslinks than in PSC. The collagen self-aggregation was observed to be affected with the increase of NHS-AA. Calfskin collagen solution was crosslinked using NHS-AA of different concentrations at pH 7.2 and then heated from 4°C to 37, 39 and 41°C within 1 min to form fibrils. Concentrations of NHS-AA were converted to molar rations of [NHS-AA]/ [NH2] where [NHS-AA] was active ester groups and [NH2] was Ɛ-amino of lysine and hydroxylysine residues. The turbidity and the rate of fibril formation decreased as the NHS-AA increased, especially at a [NHS-AA]/ [NH2] > 0.6/1. The self-aggregation was still carried out at a [NHS-AA]/ [NH2] ration < 0.3/1 when temperature was below 39°C [225]. The properties of collagen gels (fish skin) prepared by incubating the complex solutions (pH 7.4) of collagen and NHS-AA at 30°C for 5h were further studied. The gels became transparent as [NHS-AA]/[NH2] was above 0.2. The degree of crosslink observably increased with the increase of NHS-AA and then leveled off at a [NHS-AA]/[NH2] ratio > 0.8. Td of the gel increased by 7.9°C at ratio of 1.6 comparing with the uncross-linked gel [226]. Different thermal stability of collagen gels was revealed by changing the order of adding NHS-AA. The collagen gels (calfskin) were crosslinked using NHS-AA of the same dosage before the self-aggregation (CBF), simultaneously with and after the self-aggregation (CSF and CAF) respectively. The thermal stability of the gels exhibited in order of CAF > CSF > CBF. The appearance of the gels was non-transparent for CAF and became transparent for CSF and CBF at a [NHS-AA]/ [NH2] ratio > 0.6/1 [227]. Additionally, the isoelectric point of crosslinked collagen could be decreased via introducing extra carboxyl groups by NHS-activated carboxylic acid esters to made it possible to dissolve in the neutral water [228,229,230].

As “zero-length” crosslinker, EDC and DPPA promote the formation of amido bond between adjacent carboxylic and amino groups of collagen and don’t stay in the systems [209, 210]. The activated mechanisms of carboxylic groups of Glu and Asp residues induced by EDC and DPPA are different. EDC converts carboxyl groups of collagen into active O-acylisourea groups, subsequently crosslinking takes place after nucleophilic attacking O-acylisourea groups by free amine groups of lysine or hydroxylysine residues. Usually, during the EDC crosslinking procedure, NHS is added as a catalyst to significantly enhance the crosslinking efficiency owing to the formation of more stable active intermediates (Scheme 4e). The thermal stability of the collagen crosslinked using EDC and EDC/NHS respectively was compared in Old Damink’s study [231]. 1g sheep dermal layers obtained after removing non-collagenous substances were immersed into 100 ml aqueous solution containing 1.15g EDC or 1.15g EDC and 0.69g NHS at room temperature for 4h. The pH value was kept at 5.5 during the reaction. Ts of the dermal layer treated with EDC increased by 17°C and that treated by EDC/NHS increased by 30°C comparing with the native dermal layer (56°C). DPPA converts carboxyl groups into acylazide groups which further react with the amino groups of adjacent collagen chains (Scheme 4f). The thermal stability of the collagen treated with DPPA was significantly affected by the solvent used for DPPA [210]. Herbage and his group immersed 10 mg dry pericardial samples into 5 ml 1% DPPA solution in different solvents at 4°Cfor 24 h. Td of the untreated sample was 67.3°C. The Td values of the samples treated with DPPA in dimethylformamide, acetone and dichloromethane were 81.9, 72.4 and 67.4°C respectively. The thermal stability of the collagen treated with DPPA (according to the above method) and GTA respectively was further compared. Collagen gel was obtained after complete swelling of calf dermis powder in acetic acid. Collagen sponge was obtained by lyophilizing the 0.7% collagen gel and film were prepared by drying 1.5% gel at room temperature for 3 days. The sponge and film were crosslinked using 1% GTA at 20°C and pH 6.2 for 24 h. The thermal stability of the DPPA-treated sponge and film increased by 20°C comparing with the untreated sponge and film. The Td values of GTA-treated sponge and film were 27 and 22°C higher than that of the uncross-linked sample, respectively [232]. At the optimal condition, the thermal stability of collagen could be significantly improved by DPPA.

TGase-mediated crosslinking is a nontoxic method for stabilizing collagen and collagenous materials. As an enzyme in various organisms, TGases are involved in essential biological processes [211]. The γ-carboxamide groups of glutamine residues catalyzed by TGases are acyl donor and the Ɛ-amino groups of lysine residues or primary amino groups act as acyl acceptor. Consequently, the intra-or inter- molecular crosslinks can be achieved via the formation of Ɛ-(γ-glutamyl) lysine bonds (Scheme 4g) [233]. The thermal stability of the collagen treated with TGase was affected by the different states of collagen. An evident improvement of thermal stability of the TGase-treated collagen solution was reported by Chen et al. [234]. Porcine skin collagen solution (10 mg/ml) was added with TGase (60 U/g collagen) or without and then incubated at 25°C and pH 4 for 12 h. The sponges were obtained after lyophilizing the solutions. The Td values of the untreated and TGase-treated sponges were 114 and 177°C respectively. TGase showed not contribution to thermal stability of collagen fibers in Cheng’s study [235]. 10g/kg collagen fiber suspensions prepared by removing non-collagenous substances from limed bovine skin were added with TGase (20 U/g collagen) and then reacted at room temperature for 4h. Td of the TGase-treated collagen fibers was 82.95°C which was same as the native fibers (82.58°C).

The thermal stability of collagen or collagenous materials could be significantly improved by crosslinking. Usually, the denaturation temperature of collagen as insoluble forms such as fibers, films and sponges are always much higher than that in solution. Therefore, physical crosslinking, which introduce strong energy into collagen, is more suitable for modification of dried collagen matrixes. The degree of chemical crosslinking would be affected by different states of collagen. The collagen molecules in solution are more scattered, which is not conducive to the formation of intermolecular crosslinks. For leather industry, the hydrothermal stability of leather is very important. The desirable properties of leather could be easily obtained by the use of chromium and GTA in spite of the toxic issue. Besides of thermal stability, the suitable physicochemical property and acceptable biocompatibility are also taken major concern when collagen is applied in biomedicine fields. The selection and combination of crosslinkers and crosslinking methods are necessary during practical processing.

7 Conclusion

Intact triple helical structure endows collagen excellent characteristics which is the premise for diversified applications. The native structure of collagen would collapse during heating process. The thermal stability is a macroscopic expression of collagen structure stability. The studies of thermal denaturation mechanism have been developed from traditional statistics to model simulation which could more accurately reveal the thermal denaturation behavior of collagen. Usually, collagen is extracted into aqueous solution for direct use or to be further manufactured into gel, fiber, membrane, sponge, scaffold and so on according to various applications. The stability of triple helix would be easily affected by the use of solvent and physical assisted means. The monitoring of thermal stability is necessary for helping to adjust the processing technology. Modification of collagen is carried out to promote thermal stability, mechanical property and resistance to enzymatic degradation. The demand of thermal stability varies with the different uses of collagenous materials. Hydrothermal stability is the most important standard for leather. However, for biomedical materials, good biocompatibility and low toxicity are also taken major concern. The choice of reagents and assisted means during processing and the use of modified methods should adapt to the demand of practical applications.

Availability of data and materials

The datasets analyzed during the review are available in the references which have been specified in the article.

Change history

Abbreviations

ASC:

Acid-soluble collagen

PSC:

Pepsin-soluble collagen

Col:

Collagen

AA:

Acetic acid

Gly:

Glycerol

RTT:

Rat tail tendon

ILs:

Ionic liquids

AMS:

Diethyl methyl ammonium methane sulfonate

BCS:

Bis-choline sulphate

IDP:

1-butyl-3-methyl imidazolium dimethyl phosphate

cDHP:

Choline dihydrogen

[EMIM][Cl]:

1-ethyl-methylimidazolium chloride

[EMIM][Br]:

1-ethyl-methylimidazolium bromide

[EMIM][Ac]:

1-ethyl-3-methylimidazolium acetate

[EMIM][BF4]:

1-ethylmethylimidazolium tetrafluoroborate

[EMIM] [N (CN)2]:

1-ethyl-methylimidazolium dicyanamide

[BMIM][Cl]:

1-butyl-3-methylimidazolium chloride

[BMIM][BF4]:

1-butyl-3-methylimidazolium tetrafluoroborate

AA/[EMIM][Ac]:

Acetic acid/1-ethyl-3-methylimidazolium acetate

CS:

Chondroitin sulfate

C4S:

Chondroitin 4-sulfate

HA:

Hyaluronic acid

HPMC:

Hydroxypropyl methylcellulose

CMC:

Carboxymethyl cellulose

CH:

Chitosan

PVA:

Poly-vinyl-alcohol

PVP:

Poly-vinyl-pyrrolidone

PEG:

Poly-ethylene-glycol

PLGA:

Poly-D, L-lactide-co-glycolide

HFP:

1,1,1,3,3,3-hexafluoro-2-propanol

GTA:

Glutaraldehyde

DCMC:

Dialdehyde carboxymethyl cellulose

NHS:

N-hydroxysuccinimide

EDC:

1-ethyl-3(3-dimethylaminopropyl)-carbodiimide

NHS-AA:

NHS activated adipic acid

DPPA:

Diphenylphosphorylazide

TGase:

Transglutaminase

Ɛr:

Specific dielectric constant

DHT:

Dehydrathermal

UV:

Ultraviolet

[ƞ] exp m :

Specific viscosity

[ƞ] i m :

Ideal intrinsic viscosity

PBS:

Phosphate buffered saline

CD:

Circular dichroism measurements

DSC:

Differential scanning calorimetry

SDS-PAGE:

Sodium dodecyl sulfate polyacrylamide gel electrophoresis

FTIR:

Fourier transform-infrared

References

  1. Li ZQ, Liao LL. The chemistry and histology of animal skins. China: China Leather Industry Press; 2010.

    Google Scholar 

  2. Avery NC, Bailey AJ. The effects of the Maillard reaction on the physical properties and cell interactions of collagen. Pathol Biol. 2006;54:387–95.

    Article  CAS  Google Scholar 

  3. Gelse K. Collagens-structure, function, and biosynthesis. Adv Drug Deliv Rev. 2003;55:1531–46.

    Article  CAS  Google Scholar 

  4. Jiang TD. Collagen and collagen protein. China: Chemical Industry Press; 2006.

    Google Scholar 

  5. Ricard-Blum S. The collagen family. Cold Spring Harb Perspect Biol. 2010;3:1.

    Google Scholar 

  6. Engel J, Bächinger HP. Structure, stability and folding of the collagen triple helix. Top Curr Chem. 2005;247:7–33.

    Article  CAS  Google Scholar 

  7. Brodsky B, Persikov AV. Molecular structure of the collagen triple helix. Adv in Protein Chem. 2005;70:301–39.

    Article  CAS  Google Scholar 

  8. Usha R, Ramasami T. Influence of hydrogen bond, hydrophobic and electrovalent salt linkages on the transition temperature, enthalpy and activation energy in rat tail tendon (RTT) collagen fibre. Thermochim Acta. 1999;338:17–25.

    Article  CAS  Google Scholar 

  9. Brodsky B, Ramshaw JAM. The collagen triple-helix structure. Matrix Biol. 1997;15:545–54.

    Article  CAS  Google Scholar 

  10. Lee CH, Singla A, Lee Y. Biomedical applications of collagen. Int J Pharm. 2001;221:1–22.

    Article  CAS  Google Scholar 

  11. Kielty CM, Grant ME. In: Royce PM, Steinmann B, editors. The collagen family: structure, assembly, and organization in the extracellular matrix. Connective tissue and its heritable disorders: molecular, genetic, and medical aspects, second edition. Hoboken: Wiley; 2003.

    Google Scholar 

  12. Yoon D, Cho YS, Joo SY, Seo C. A clinical trial with a novel collagen dermal substitute for wound healing in burn patients. Biomater Sci. 2020. https://doi.org/10.1039/c9bm01209e.

  13. Chen LJ, Bao B, Wang NP, Xie J, Wu WH. Oral administration of shark Type II collagen suppresses Complete Freund’s Adjuvant-induced rheumatoid arthritis in rats. Pharmaceuticals. 2012;5:339–52.

    Article  CAS  Google Scholar 

  14. Chen XF, Zhou LL, Xu HZ, Yamamoto M, Shinoda M, Tada I, Minami S, Urayama K, Yamane H. The structure and properties of natural sheep casing and artificial films prepared from natural collagen with various crosslinking treatments. Int J Biol Macromol. 2019;135:959–68.

    Article  CAS  Google Scholar 

  15. Li GY, Fukunaga S, Takenouchi K, Nakamura F. Comparative study of the physiological properties of collagen, gelatin and collagen hydrolysate as cosmetic materials. Int J Cosmet Sci. 2005;27:101–6.

    Article  Google Scholar 

  16. Li Y, Liu WT, Du ZL, Li GY. Comparison of dynamic denaturation temperature of collagen with its static denaturation temperature and the configuration characteristics in collagen denaturation processes. Thermochim Acta. 2008;469(1−2):71–6.

    Article  CAS  Google Scholar 

  17. Burjanadze TV, Veis A. A thermodynamic analysis of the contribution of hydroxyproline to the structural stability of the collagen triple helix. Connect Tissue Res. 1997;36(4):347–65.

    Article  CAS  Google Scholar 

  18. Zhang ZK, Li GY, Shi B. Physicochemical properties of collagen, gelatin and collagen hydrolysate derived from bovine limed split wastes. J Soc Leather Tech Chem. 2006;90:23–8.

    Google Scholar 

  19. Sadowska M, Koiodziejska I, Niecikowska C. Isolation of collagen from the skins of Baltic cod (Gadus morhua). Food Chem. 2003;81(2):257–62.

    Article  CAS  Google Scholar 

  20. Skierka E, Sadowska M. The influence of different acids and pepsin on the extractability of collagen from the skin of Baltic cod (Gadus morhua). Food Chem. 2007;105(3):1302–6.

    Article  CAS  Google Scholar 

  21. Takahashi S, Geenen D, Nieves E. Collagenase degrades collagen in vivo in the ischemic heart. Biochim Biophys Acta. 1999;1428:251–9.

    Article  CAS  Google Scholar 

  22. Zhang ZQ, Zhang YH, An Q, Xu RR, Li H, Wang EB. Study on immunogenicity of type I atelocollagen. Chin J Tissue Eng Res. 2019;23(22):3542–8.

    Google Scholar 

  23. Veeruraj A, Arumugam M, Balasubramanian T. Isolation and characterization of thermostable collagen from the marine eel-fish (Evenchelys macrura). Process Biochem. 2013;48(10):1592–602.

    Article  CAS  Google Scholar 

  24. Yu D, Chi CF, Wang B, Ding GF, Li ZR. Characterization of acid-and pepsin-soluble collagens from spines and skulls of skipjack tuna (Katsuwonus pelamis). Chin J Nat Med. 2014;12(9):712–20.

    CAS  Google Scholar 

  25. Li ZR, Wang B, Chi CF, Zhang QH, Gong YD, Tang JJ, Luo HY, Ding GF. Isolation and characterization of acid soluble collagens and pepsin soluble collagens from the skin and bone of Spanish mackerel (Scomberomorous niphonius). Food Hydrocoll. 2013;31(1):103–13.

    Article  CAS  Google Scholar 

  26. Portier RJ, Moody MW, Ogawa M. Biochemical properties of black drum and sheepshead seabream skin collagen. J Agric Food Chem. 2003;51(27):8088–92.

    Article  Google Scholar 

  27. Wang J, Pei XL, Liu HY, Zhou D. Extraction and characterization of acid-soluble and pepsin-soluble collagen from skin of loach ( Misgurnus anguillicaudatus ). Int J Biol Macromol. 2017. https://doi.org/10.1016/j.ijbiomac.2017.08.046.

  28. Gao LL, Wang ZY, Li Z, Zhang CX, Zhang DQ. The characterization of acid and pepsin soluble collagen from ovine bones (Ujumuqin sheep). J Integr Agric. 2018;17(3):704–11.

    Article  CAS  Google Scholar 

  29. Chen Y, Ye R, Wang Y. Aci-soluble and pepsin-soluble collagens from grass carp (Ctenopharyngodon idella) skin: a comparative study on physicochemical properties. Int J Food Sci Technol. 2015;50(1):186–93.

    Article  CAS  Google Scholar 

  30. Sun LL, Hou H, Li BF, Zhang Y. Characterization of acid- and pepsin-soluble collagen extracted from the skin of Nile tilapia (Oreochromis niloticus). Int J Biol Macromol. 2017;99:8–14.

    Article  CAS  Google Scholar 

  31. Matmaroh K, Benjakul S, Prodpran T, Encarnacion AB, Kishimura H. Characteristics of acid soluble collagen and pepsin soluble collagen from scale of spotted golden goatfish (Parupeneus heptacanthus). Food Chem. 2011;129(3):1179–86.

    Article  CAS  Google Scholar 

  32. Zhang M, Liu WT, Li GY. Isolation and characterisation of collagens from the skin of largefin longbarbel catfish (Mystus macropterus). Food Chem. 2009;115(3):826–31.

    Article  CAS  Google Scholar 

  33. Wang L, An X, Xin Z, Zhao L, Hu Q. Isolation and characterization of collagen from the skin of deep-sea redfish (Sebastes mentella). J Food Sci. 2007;72(8):E450–5.

    Article  CAS  Google Scholar 

  34. Zhang JJ, Duan R. Characterisation of acid-soluble and pepsin-solubilised collagen from frog (Rana nigromaculata) skin. Int J Biol Macromol. 2017;101:638–42.

    Article  CAS  Google Scholar 

  35. Heu MS, Lee JH, Kim HJ, Jee SJ, Lee JS, Jeon YJ, Shahidi F, Kim JS. Characterization of acid- and pepsin-soluble collagens from flatfish skin. Food Sci Biotechnol. 2010;19(1):27–33.

    Article  CAS  Google Scholar 

  36. Kittiphattanabawon P, Benjakul S, Visessanguan W, Shahidi F. Isolation and properties of acid- and pepsin-soluble collagen from the skin of blacktip shark (Carcharhinus limbatus). Eur Food Res Technol. 2010;230(3):475–83.

    Article  CAS  Google Scholar 

  37. Singh P, Benjakul S, Maqsood S, Kishimura H. Isolation and characterisation of collagen extracted from the skin of striped catfish (Pangasianodon hypophthalmus). Food Chem. 2011;124(1):97–105.

    Article  CAS  Google Scholar 

  38. Kozlowska J, Sionkowska A, Skopinska-Wisniewska J, Piechowicz K. Northern pike (Esox lucius) collagen: extraction, characterization and potential application. Int J Biol Macromol. 2015;81:220–7.

    Article  CAS  Google Scholar 

  39. Sato K, Ebihara T, Adachi E, Kawashima S, Hattori S, Irie S. Possible involvement of aminotelopeptide in self-assembly and thermal stability of collagen I as revealed by its removal with proteases. J Biol Chem. 2000;275(33):25870–5.

    Article  CAS  Google Scholar 

  40. Walton RS, Brand DD, Czernuszka JT. Influence of telopeptides, fibrils and crosslinking on physicochemical properties of type I collagen films. J Mater Sci Mater Med. 2010;21(2):451–61.

    Article  CAS  Google Scholar 

  41. Woodley DT, Yamauchi M, Wynn KC, Mechanic G, Briggaman RA. Collagen telopeptides (cross-linking sites) play a role in collagen gel lattice contraction. J Invest Dermatol. 1991;97(3):580–5.

    Article  CAS  Google Scholar 

  42. Piez KA, Trus BL. Microfibrillar structure and packing of collagen: hydrophobic interactions. J Mol Biol. 1977;110(4):701–4.

    Article  CAS  Google Scholar 

  43. Bowes JH, Kenten RH. The effect of alkalis on collagen. Biochem J. 1948;43(3):365–72.

    Article  CAS  Google Scholar 

  44. Courts A. Structural changes in collagen. The action of alkalis and acids in the conversion of collagen into eucollagen. Biochem J. 1960;74(2):238–47.

    Article  CAS  Google Scholar 

  45. Kuehn K, Zimmer E, Waykole P, Fietzek P. The effect of alkali on collagen. The change in the charge pattern of the tropocollagen molecule and splitting of intra- and intermoleculer compounds. Hoppe Seylers Z Physiol Chem. 1963;333:209–17.

    CAS  Google Scholar 

  46. Kemp GD, Tristram GR. The preparation of an alkali-soluble collagen from demineralized bone. Biochem J. 1971;124(5):915–9.

    Article  CAS  Google Scholar 

  47. Hattori S, Adachi E, Ebihara T, Shirai T, Someki I, Irie S. Alkali-treated collagen retained the triple helical conformation and the ligand activity for the cell adhesion via α2β1 integrin. J Biochem. 1999;125:676–84.

    Article  CAS  Google Scholar 

  48. Suzuki Y, Someki I, Adachi E, Irie S, Hattori S. Interaction of collagen molecules from the aspect of fibril formation: acid-soluble, alkali-treated, and MMP1-digested fragments of type I collagen. Jpn Biochem Soc. 1999;128:54–67.

    Google Scholar 

  49. Zhao S, Zhang M, Li GY, Liu WT. Preparation and characterization of alkali-soluble collagen from pigskin shavings. J Am Leather Chem Assoc. 2009;104:344–51.

    CAS  Google Scholar 

  50. Jayasooriya SD, Bhandari BR, Torley P, D’Arcy BR. Effect of high power ultrasonic waves on properties of meat: a review. Int J Food Prop. 2004;7(2):301–19.

    Article  CAS  Google Scholar 

  51. Kruus P, O’Neill M, Robertson D. Ultrasonic initiation of polymerization. Ultrasonics. 1990;28:304–9.

    Article  CAS  Google Scholar 

  52. Akram A, Zhang CH. Effect of ultrasonication on the yield, functional and physicochemical characteristics of collagen-II from chicken sternal cartilage. Food Chem. 2019. https://doi.org/10.1016/j.foodchem.2019.125544.

  53. Qu Y, Su HY, Liu WT, Li GY. Extraction of collagen from bovine hide assisted by ultrasonic. Leather Sci Eng. 2018;28(2):05–9.

    Google Scholar 

  54. Ran XG, Wang LY. Use of ultrasonic and pepsin treatment in tandem for collagen extraction from meat industry by-products. J Sci Food Agric. 2014;94(3):585–90.

    Article  CAS  Google Scholar 

  55. Li DF, Mu CD, Cai SM, Lin W. Ultrasonic irradiation in the enzymatic extraction of collagen. Ultrason Sonochem. 2009;16(5):605–9.

    Article  CAS  Google Scholar 

  56. Zou Y, Xu PP, Li PP, Cai PP, Zhang MH, Sun ZL, Sun C, Xu WM, Wang DY. Effect of ultrasonic pre-treatment on the characterization and properties of collagen extracted from soft-shelled turtle (Pelodiscus sinensis). Food Sci Technol. 2017;82:72–81.

    CAS  Google Scholar 

  57. Jung SK, Kim YH, Kim YE, Song KM, Lee NH. Development of industrial ultrasonic system for mass production of collagen and biochemical characteristics of extracted collagen. Food Bioprod Process. 2018. https://doi.org/10.1016/j.fbp.2018.04.001.

  58. Kim HK, Kim YH, Kim YJ, Park HJ, Lee NH. Effects of ultrasonic treatment on collagen extraction from skins of the sea bass Lateolabrax japonicus. Fish Sci. 2012;78(2):485–90.

    Article  Google Scholar 

  59. Kim HK, Kim YH, Park HJ, Lee NH. Application of ultrasonic treatment to extraction of collagen from the skins of sea bass Lateolabrax japonicus. Fish Sci. 2013;79(5):849–56.

    Article  CAS  Google Scholar 

  60. Huang KM, Yang XQ. New progress in the study of non-thermal effects in microwave accelerated chemical reactions. Prog Nat Sci. 2006;16(3):273–9.

    Google Scholar 

  61. Li YW, Li HJ. Optimization of microwave-assisted extraction process for collagen from pigskin. Food Sci. 2012;33(6):11–4.

    Google Scholar 

  62. Cheng YL, Wu JC, Zhang JW, Chen WY. A novel collagen extraction method ased on microwave irradiation. Dresden: XXXV IULTCS Congress; 2019.

  63. Li DF, Mu CD, Zhang Q, Zhou YL, Lin W. Effects of microwave irradiation on collagen denaturation. J Food Biochem. 2010;34(6):1319–31.

    Article  CAS  Google Scholar 

  64. Jenness DD, Sprecher C, Johnson WC. Circular dichroism of collagen, gelatin, and poly (proline) II in the vacuum ultraviole. Biopolymers. 1976;15(3):513–21.

    Article  CAS  Google Scholar 

  65. Li D, Yang W, Li GY. Extraction of native collagen from limed bovine split wastes through improved pretreatment methods. J Chem Technol Biotechnol. 2008;83:1041–8.

    Article  CAS  Google Scholar 

  66. Xu SC, Yang H, Shen LR, Li GY. Purity and yield of collagen extracted from Southern catfish (Silurus meridionalis Chen) skin through improved pretreatment methods. Int J Food Prop. 2017;20:141–53.

    Article  Google Scholar 

  67. Earle M, Seddon K. Ionic liquids, green solvents for the future. Pure Appl Chem. 2000;72(7):1391–8.

    Article  CAS  Google Scholar 

  68. Vijayaraghavan R, Vedaraman N, Muralidharan C, Mandal AB, MacFarlane DR. Aqueous ionic liquid solutions as alternatives for sulphide-free leather processing. Green Chem. 2015;17:1001–7.

    Article  CAS  Google Scholar 

  69. Jayakumar GC, Mehta A, Rao JR, Fathima NN. Ionic liquids: new age materials for eco-friendly leather processing. RSC Adv. 2015;5:31998–2005.

    Article  CAS  Google Scholar 

  70. Alla JP, Rao JR, Fathima NN. Integrated depilation and fiber opening using aqueous solution of ionic liquid for leather processing. ACS Sustain Chem Eng. 2017;5:8610–8.

    Article  CAS  Google Scholar 

  71. Liu SC, Liu WT, Li GY. Effect of ionic liquids pretreatmen on the extaction of collagen from calf skin. J Am Leather Chem. 2019;114:391–9.

    CAS  Google Scholar 

  72. Rodríguez MIA, Barroso LGR, Sánchez ML. Collagen: a review on its sources and potential cosmetic applications. J Cosmet Dermatol. 2017;10:1–7.

    Google Scholar 

  73. Li GY, Liu WT. Collagen chemistry. 1st ed. China: China Light Industry Press; 2013.

    Google Scholar 

  74. Joy DP, Kumar KK, Kumar SKS. Collagen from squid and its bioliogical activity. Int J Curr Pharm Res. 2017;9(3):24–6.

    CAS  Google Scholar 

  75. Kotler ED, Marshall WS, Gareta EG. Sources of collagen for biomaterials in skin wound healing. Bioengineering. 2019. https://doi.org/10.3390/bioengineering6030056.

  76. Burianadze TV. New analysis of the phylogenetic change of collagen thermostability. Biopolymers. 2000;53:523–8.

    Article  Google Scholar 

  77. Ikoma T, Kobayashi H, Tanaka J, Walsh D, Mann S. Physical properties of type I collagen extracted from fish scales of Pagrus major and Oreochromis niloticas. Biol Macromol. 2003;32:199–204.

    Article  CAS  Google Scholar 

  78. Burjanadze TV. Hydroxyproline content and location in relation to collagen thermal stability. Pept Sci. 1979;18(4):931–8.

    Article  CAS  Google Scholar 

  79. Lodhi G, Kim YS, Kim EK, Hwan JW, Won HS, Kim W, Moon SH, Jeon BT, Park PJ. Isolation and characterisation of acid- and pepsin-soluble collagen from the skin of Cervus korean TEMMINCK var. mantchuricus Swinhoe. Anim Prod Sci. 2018;58(3):585–94.

    Article  CAS  Google Scholar 

  80. Zhang FX, Wang AN, Li ZH, He SW, Shao LJ. Preparation and characterisation of collagen from Freshwater fish scales. Food Nutr Sci. 2011;2:818–23.

    CAS  Google Scholar 

  81. Liu WT, Li GY, Miao YQ, Wu XH. Peparation and characterization of pepsin-solubilized type I collagen from the scales of snakhead (ophiocephalus argus). Food Biochem. 2009;33(1):20–37.

    Article  CAS  Google Scholar 

  82. Li H, Liu BL, Gao LZ, Chen HL. Studies on bullfrog skin collagen. Food Chem. 2004;84(1):65–9.

    Article  CAS  Google Scholar 

  83. Ahmad M, Benjakul S. Extraction and characterisation of pepsin-solubilised collagen from the skin of unicorn leatherjacket (Aluterus monocerous). Food Chem. 2010;120(3):817–24.

    Article  CAS  Google Scholar 

  84. Zelechowska E, Sadowska M, Turk M. Isolation and some properties of collagen from the backbone of Baltic cod (Gadus morhua). Food Hydrocoll. 2010;24(4):325–9.

    Article  CAS  Google Scholar 

  85. Wang SS, Sun XQ, Zhou DQ. Physicochemical characteristics and fibril-forming properties of collagen from paddlefish ( polyodon spathula ) and globefish (Fugu flavidus) skin byproducts. Food Sci Technol. 2017;37(2):176–83.

    Article  Google Scholar 

  86. Barzideh Z, Latiff AA, Gan CY, Benjakul S, Karim AA. Isolation and characterisation of collagen from the ribbon Jellyfish (Chrysaora sp.). Food Sci Technol. 2013. https://doi.org/10.1111/ijfs.12464.

  87. Wu JL, Guo XB, Liu H, Chen L. Isolation and comparative study on the characterization of guanidine hydrochloride soluble collagen and pepsin soluble collagen from the body of surf clam shell (Coelomactra antiquata). Foods. 2019. https://doi.org/10.3390/foods8010011.

  88. Song WK, Liu D, Li BF, Hou H. Physicochemical and biocompatibility properties of Type I collagen from the skin of Nile Tilapia (Oreochromis niloticus) for biomedical applications. Mar Drugs. 2019. https://doi.org/10.3390/md17030137.

  89. Zhong M, Chen T, Hu CQ, Ren CH. Isolation and characterization of collagen from the body wall of sea cucumber stichopus monotuberculatus. Food Chem. 2015. https://doi.org/10.1111/1750-3841.12826.

  90. Nagai T, Araki Y, Suzuki N. Collagen of the skin of ocellate puffer fish (Takifugu rubripes). Food Chem. 2002;78(2):173–7.

    Article  CAS  Google Scholar 

  91. Tziveleka LA, Ioannou E, Tsiourvas D, Berillis P, Foufa E, Roussis V. Collagen from the marine sponges Axinella cannabina and Suberites carnosus: isolation and morphological, biochemical, and biophysical bharacterization. Mar Drugs. 2017. https://doi.org/10.3390/md15060152.

  92. Zhang Y, Liu WT, Li GY, Shi B, Miao YQ, Wu XH. Isolation and partial characterization of pepsin-soluble collagen from the skin of grass carp (Ctenopharyngodon idella). Food Chem. 2007;103:906–12.

    Article  CAS  Google Scholar 

  93. Tan CC, Karim AA, Latiff AA, Gan CY, Ghazali FC. Extraction and characterization of pepsin-solubilized collagen from the body wall of crown-of-thorns Starfish (Acanthaster planci). Int Food Res. 2013;20(6):3013–20.

    Google Scholar 

  94. Rigby BJ. Amino-acid composition and thermal stability of the skin collagen of the Antarctic Ice-fish. Nature. 1968;219:166–7.

    Article  CAS  Google Scholar 

  95. Veeruraj A, Arumugam M, Ajithkumar T, Balasubramanian T. Isolation and characterization of collagen from the outer skin of squid (Doryteuthis singhalensis). Food Hydrocoll. 2015;43:708–16.

    Article  CAS  Google Scholar 

  96. Muralidharan N, Shakia RJ, Sukumar D, Jeyasekaran G. Skin, bone and muscle collagen extraction from the trash fish, leather jacket (Odonus niger) and their characterization. Food Sci Technol. 2013;50(6):1106–13.

    CAS  Google Scholar 

  97. Kim JS, Park JW. Characterization of acid-soluble collagen from pacific whiting surimi processing byproducts. Food Chem Technol. 2004;69(8):637–42.

    Google Scholar 

  98. Gaill F, Mann K, Wiedemann H, Engel J, Timpl R. Structural comparison of cuticle and interstitial collagens from annelids living in shallow sea-water and at deep-sea hydrothermal vents. J Mol Biol. 1995;246(2):284–94.

    Article  CAS  Google Scholar 

  99. Duan R, Konno K, Wang SJ. Different thermostability of collagens from scale of carp (cyprius carpio) in winter and summer. Food Biochem. 2010;34:1275–87.

    Article  CAS  Google Scholar 

  100. Pan BS, Chen HE, Sung WC. Molecular and thermal characteristics of acid-soluble collagen from orbicular batfish: effects of deep-sea water culturing. Food Prop. 2018;21(1):1080–90.

    Article  CAS  Google Scholar 

  101. Berg RA, Prockop DJ. The thermal transition of a non-hydroxylated from of collagen, evidence for a role for hydroxyproline in stabilizing the triple-helix of collagen. Biochem Biophys Res Commun. 1973;52(1):116–20.

    Article  Google Scholar 

  102. Rosenbloom J, Harsch M. Hydroxyproline content determines the denaturation on temperature of chkick tendon collagen. Biochem Biophys. 1973;158:478–84.

    Article  CAS  Google Scholar 

  103. Gustavson KH. The function of hydroxyproline in collagens. Nature. 1955;175(4445):70–4.

    Article  CAS  Google Scholar 

  104. Miles CA, Knott L, Sumner IG, Bailey AJ. Differences between the thermal stabilities of the three triple-helical domains of type IX collagen. J Mol Biol. 1998;277(1):135–44.

    Article  CAS  Google Scholar 

  105. Miles CA, Bailey AJ. Thermally labile domains in the collagen molecule. Micron. 2001;32(3):325–32.

    Article  CAS  Google Scholar 

  106. Josse J, Harrington WF. Role of pyrrolidine residues in the structure and stabilization of collagen. J Mol Biol. 1964;9(2):269–87.

    Article  CAS  Google Scholar 

  107. Gaill F. Aspects of life development at deep sea hydrothermal vents. FASEB J. 1993;7(6):558–65.

    Article  CAS  Google Scholar 

  108. Engel J, Kurtz J, Katchalski E, Berger A. Polymers of tripeptides as collagen models: II. Conformational changes of poly (L-prolyl-glycyl-L-prolyl) in solution. J Mol Biol. 1966;17(1):225–72.

    Article  Google Scholar 

  109. Rao NV, Harrington WF. Relationship between the positioning of pyrrolidine residues and the stability of collagen. J Mol Biol. 1966;21:577–81.

    Article  CAS  Google Scholar 

  110. Nagarajan V, Kamitori S, Okuyama K. Structure analysis of a collagen-model peptide with a (Pro-Hyp-Gly) sequence repeat. J Biochem. 1999;125(2):310–8.

    Article  CAS  Google Scholar 

  111. Kramer RZ, Bella J, Mayville P, Brodsky B, Berman H. Sequence dependent conformational variations of collagen triplehelical structure. Nat Struct Biol. 1999;6(5):454–7.

    Article  CAS  Google Scholar 

  112. Shah NK, Ramshaw JAM, Kirkpatrick A, Shah C, Brodsky B. A host-guest set of triple-helical peptides: stability of Gly-X-Y triplets containing common nonpolar residues. Biochemistry. 1996;35(32):10262–8.

    Article  CAS  Google Scholar 

  113. Persikov AV, Ramshaw JAM, Brodsky B. Collagen model peptides: Sequence dependence of triple-helix stability. Biopolymers. 2000;55(6):436–50.

    Article  CAS  Google Scholar 

  114. Inouye K, Sakakibara S, Prockop DJ. Effects of the stereo-configuration of the hydroxyl group in 4-hydroxyproline on the triple-helical structures formed by homogenous peptides resembling collagen. Biochim Biophys Acta. 1976;420(1):133–41.

    Article  CAS  Google Scholar 

  115. Friedman L, Higgin JJ, Moulder G, Barstead R, Raines RT, Kimble J. Prolyl 4-hydroxylase is required for viability and morphogenesis in Caenorhabditis elegans. Proc Natl Acad Sci U S A. 2000;97(9):4736–41.

    Article  CAS  Google Scholar 

  116. Bella J, Eaton M, Brodsky B, Berman HM. Crystal and molecular structure of a collagen-like peptide at 1.9Å resolution. Science. 1994;266:75–81.

    Article  CAS  Google Scholar 

  117. Bella J, Brodsky B, Berman HM. Hydration structure of a collagen peptide. Structure. 1995;3:893–906.

    Article  CAS  Google Scholar 

  118. Persikov AV, Ramshaw AM, Kirkpatrick A, Brodsky B. Triple-helix propensity of hydroxyproline and fluoroproline: comparison of host-guest and repeating tripeptide models. J Am Chem Soc. 2003;125:11500–1.

    Article  CAS  Google Scholar 

  119. Holmgren SK, Taylor KM, Bretscher LE, Raines RT. Code for collagen’s stability deciphered. Nature. 1998;392(6677):666–7.

    Article  CAS  Google Scholar 

  120. Holmgren SK, Bretscher LE, Taylor KM, Raines RT. A hyperstable collagen mimic. Chem Biol (London). 1999;6(2):63–70.

    CAS  Google Scholar 

  121. Goldstein A, Adams E. Glycylhydroxyprolyl sequences in earthworm cuticle collagen: glycylhydroxyprolylserine. J Biol Chem. 1970;245(20):5478–83.

    Article  CAS  Google Scholar 

  122. Kefalides NA. Structure and biosynthesis of basement membranes. Int Rev Connect Tissue Res. 1973;6:63–104.

    Article  CAS  Google Scholar 

  123. Mizuno K, Hayashi T, Peyton DH, Bächinger HP. Hydroxylation-induced stabilization of the collagen triple helix. J Biol Chem. 2004;279(36):38072–8.

    Article  CAS  Google Scholar 

  124. Improta R, Berisio R, Vitagliano L. Contribution of dipole-dipole interactions to the stability of the collagen triple helix. Protein Sci. 2008;17(5):955–61.

    Article  CAS  Google Scholar 

  125. Inouye K, Kobayashi Y, Kyogoku Y, Kishida Y, Sakakibara S, Prockop DJ. Synthesis and physical properties of (hydroxyproline-proline-glycine)10. Hydroxyproline in the X-position decreases the melting temperature of the collagen triple helix. Arch Biochem Biophys. 1982;219:198–203.

    Article  CAS  Google Scholar 

  126. Vitagliano L, Berisio R, Mazzarella L, Zagari A. Structural bases of collagen stabilization induced by proline hydroxylation. Biopolymers. 2001;58(5):459–64.

    Article  CAS  Google Scholar 

  127. Berisio R, Granata V, Vitagliano L, Zagari A. Imino acids and collagen triple helix stability: characterization of collagen-like polypeptides containing Hyp-Hyp-Gly sequence repeats. J Am Chem Soc. 2004;126(37):11402–3.

    Article  CAS  Google Scholar 

  128. Mann K, Mechling DE, Bächiger HP, Eckerskorn C, Gaill F, Timpl R. Glycosylated threonine but not 4-hydroxyproline dominates the triple helix stabilizing positions in the sequence of a hydrothermal vent worm cuticle collagen. J Mol Biol. 1996;261(2):255–66.

    Article  CAS  Google Scholar 

  129. Bann JG, Peyton DH, Bächinger HP. Sweet is stable: glycosylation stabilizes collagen. FEBS Lett. 2000;473(2):237–40.

    Article  CAS  Google Scholar 

  130. Bann JG, Bächinger HP, Peyton DH. Role of carbohydrate in stabilizing the triple-helix in a model for a deep-sea hydrothermal vent worm collagen. Biochemistry. 2003;42(14):4042–8.

    Article  CAS  Google Scholar 

  131. Xu SC, Gu M, Wu K, Li GY. Unraveling the role of hydroxyproline in maintaining the thermal stability of the collagen triple helix structure using simulation. Phys Chem. 2019;123:7754–63.

    CAS  Google Scholar 

  132. Liu WT, Li GY. Non-isothermal kinetic analysis of the thermal denaturation of type I collagen in solution using isoconversional and multivariate non-linear regression methods. Polym Degrad Stabil. 2010;95:2233–40.

    Article  CAS  Google Scholar 

  133. Ding CC, Zhang M, Wu K, Li GY. The response of collagen molecules in acid solution to temperature. Polymer. 2014;55:5751–9.

    Article  CAS  Google Scholar 

  134. Wu K, Liu WT, Li GY. The aggregation behavior of native collagen in dilute solution studied by intrinsic fluorescence and external probing. Spectrochim Acta A Mol Biomol Spectrosc. 2013;102:186–93.

    Article  CAS  Google Scholar 

  135. Bella J. Collagen structure: new tricks from a very old dog. Biochemistry. 2016;473:1001–250.

    Article  CAS  Google Scholar 

  136. Cassel JM. Collagen aggregation phenomena. Biopolymers. 1966;4(9):989–97.

    Article  CAS  Google Scholar 

  137. Liu WT, Tian ZH, Li CH, Li GY. Thermal denaturation of fish collagen in solution: a calorimetric and kinetic analysis. Thermochim Acta. 2014;581(6):32–40.

    Article  CAS  Google Scholar 

  138. Yang H, Xu SC, Shen LR, Liu WT, Li GY. Changes in aggregation behavior of collagen molecules in solutionwith varying concentrations of acetic acid. Biol Macromol. 2016;92:581–6.

    Article  CAS  Google Scholar 

  139. Hofmeister F. On the understanding of the effects of salts. Arch Exp Pathol Pharmakol. 1888;24:247–60.

    Article  Google Scholar 

  140. Luescher M. Effect of hydration upon the thermal stablility of tropocollagen and its sependence on the presence of neutral salts. Biopolymers. 1974;13:2489–530.

    Article  CAS  Google Scholar 

  141. Combariza JE, Kestner NR, Jortner J. Energy-structure relationships for microscopic solvation of anions inwater clusters. J Chem Phys. 1994;100:2851–64.

    Article  Google Scholar 

  142. Grossfield A, Ren P, Ponder JW. Ion solvation thermodynamics from simulation with a polarizable force field. J Am Chem Soc. 2003;125:15671–82.

    Article  CAS  Google Scholar 

  143. Komsa-Penkova R, Koynova R, Tenchov BG, et al. Thermal stability of calf skin collagen type I in salt solutions. Biochim Biophys Acta. 1996;1297:171–81.

    Article  Google Scholar 

  144. Goindi S, Kaur R, Kaur R. An ionic liquid-in-water microemulsion as a potential carrier for topical delivery of poorly water soluble drug: development, ex-vivo, and in-vivo, evaluation. Int J Pharm. 2015;495(2):913–23.

    Article  CAS  Google Scholar 

  145. Hayes R, Warr GG, Atkin R. Structure and nanostructure in ionic liquids. Chem Rev. 2015;115(13):6357–642.

    Article  CAS  Google Scholar 

  146. Tarannum A, Muvva C, Mehta A, Rao RJ, Fathima NN. Role of preferential ions of ammonium ionic liquid in destabilization of collagen. J Phys Chem B. 2016. https://doi.org/10.1021/acs.jpcb.6b02723.

  147. Tarannum A, Adams A, Blumich B, Fathima NN. Impact of ionic liquids on the structure and dynamics of collagen. J Phys Chem B. 2017;122(3):1060–5.

    Article  Google Scholar 

  148. Mehta A, Raghava RJ, Fathima NN. Electrostatic forces mediated by choline dihydrogen phosphate stabilize collagen. J Phys Chem B. 2015;119(40):12816–27.

    Article  CAS  Google Scholar 

  149. Yang H, Li Q, Liu SC, Li GY. Acetic acid/1-ethyl-3-methylimidazolium acetate as a biphasic solvent system for altering the aggregation behavior of collagen molecules. J Mol Liquids. 2018;262:78–85.

    Article  CAS  Google Scholar 

  150. Liu SC, Li Q, Li GY. Investigation of the solubility and dispersion degree of calf skin collagen in ionic liquids. J Leather Sci Eng. 2019. https://doi.org/10.1186/s42825-019-0013-9.

  151. Zhai ZW, Wang HB, Wei BM, Yu PW, Xu CZ, He L, Zhang JT, Xu YL. Effect of ionic liquids on the fibril-formation and gel properties of grass carp (Ctenopharyngodon idellus) skin collagen. Macromol Res. 2018;26(7):609–15.

    Article  CAS  Google Scholar 

  152. Tanford C. The hydrophobic effect: formation of micelles and biological membranes 2nd ed. New York: Wiley; 1980.

    Google Scholar 

  153. Penkova R, Goshev I, Gorinstein S, Nedkov P. Stability of collagen during denaturation. J Protein Chem. 1999;18(4):397–401.

    Article  CAS  Google Scholar 

  154. Penkova R, Goshev I, Gorinstein S, Nedkov P. Stabilizing effect of glycerol on collagen type I isolated from different species. Food Chem. 1999;66(4):483–7.

    Article  CAS  Google Scholar 

  155. Na GC. Interaction of calf skin collagen with glycerol: linked function analysis. Biochemistry. 1986;25(5):967–73.

    Article  CAS  Google Scholar 

  156. Li JH, Li GY. The thermal behavior of collagen in solution: effect of glycerol and 2-propanol. Int J Biol Macromol. 2011;48(2):364–8.

    Article  CAS  Google Scholar 

  157. Li JH, Liu WT, Li GY. The effect of glycerol and 2-propanol on the molecular aggregation of collagen in solution. Int J Biol Macromol. 2015;72:1097–103.

    Article  CAS  Google Scholar 

  158. Usha R, Ramasami T. Role of aliphatic alcohols on the stability of rat-tail tendon (RTT) collagen fiber. J Polym Sci B Polym Phys. 1999;37(13):1397–405.

    Article  CAS  Google Scholar 

  159. Usha R, Ramasami T. The effects of urea and n-propanol on collagen denaturation: using DSC, circular dicroism and viscosity. Thermochim Acta. 2004;409(2):201–6.

    Article  CAS  Google Scholar 

  160. Nezu T, Morikawa T, Sasaki K, Saitoh S, Taira M, Terada Y, Araki Y. New index for the stability of a type I collagen affected by hydrophobic environment. Dent Mater J. 2007;26(3):373–81.

    Article  CAS  Google Scholar 

  161. Shanmugam G, Reddy SMM, Madhan B, Rao JR. Method of addition of acetonitrile influences the structure and stability of collagen. Process Biochem. 2014;49(2):210–6.

    Article  CAS  Google Scholar 

  162. Gopinath A, Reddy SM, Madhan B, Shanmguam G, Rao JR. Effect of aqueous ethanol on the triple helical structure of collagen. Eur Biophys J. 2014;43(12):643–52.

    Article  CAS  Google Scholar 

  163. Usha R, Maheshwari R, Dhathathreyan A, Ramasami T. Structural influence of mono and polyhydric alcohols on the stabilization of collagen. Colloids Surf B Biointerfaces. 2006;48(2):101–5.

    Article  CAS  Google Scholar 

  164. Hart GJ, Russell AE, Cooper DR. The effects of certain glycols, substituted glycols and related organic solvents on the thermal stability of soluble collagen. Biochem J. 1971;125(2):599–604.

    Article  CAS  Google Scholar 

  165. Lewandowska K, Sionkowska A, Grabska S, Kaczmarek B, Michalska M. The miscibility of collagen/hyaluronic acid/chitosan blends investigated in dilute solutions and solids. J Mol Liquids. 2016;220:726–30.

    Article  CAS  Google Scholar 

  166. Tian HL, Li CH, Liu WT, Li JH, Li GY. The influence of chondroitin 4-sulfate on the reconstitution of collagen fibrils in vitro. Colloids Surf B Biointerfaces. 2013;105:259–66.

    Article  CAS  Google Scholar 

  167. Zhao S, Lu W, Li GY. Reserch on properties of collagen/HA composite sponge. Leather Sci Eng. 2008;18(5):12–7.

    CAS  Google Scholar 

  168. Tian HL, Chen YH, Ding CC, Li GY. Interaction study in homogeneous collagen/chondroitin sulfate blends by two-dimensional infrared spectroscopy. Carbohydr Polym. 2012;89(2):542–50.

    Article  CAS  Google Scholar 

  169. Tan QT, Tian ZH, Li GY. Specific interaction study in collagen/hyaluronic acid blends by two-dimensional infrared correlation spectroscopy. Spectrosc Spect Anal. 2011;31(4):970–4.

    CAS  Google Scholar 

  170. Ding CC, Zhang M, Li GY. Preparation and characterization of collagen/hydroxypropyl methylcellulose (HPMC) blend film. Carbohydrate Polymers. 2015;119:194–201.

    Article  CAS  Google Scholar 

  171. Ding CC, Zhang M, Tian HL, Li GY. Effect of hydroxypropyl methylcellulose on collagen fibril formation in vitro. Int J Biol Macromol. 2013;52:319–26.

    Article  CAS  Google Scholar 

  172. Ding CC, Tian HL, Li GY. Specific interaction in collagen/hydroxypropyl methylcellulose blends studied by viscometry and two-dimensional infrared correlation spectroscopy. J Funct Mater. 2012;43(8):992–6.

    CAS  Google Scholar 

  173. Li CH, Tian HL, Duan L, Tian ZH, Li GY. Characterization of acylated pepsin-solubilized collagen with better surface activity. Int J Biol Macromol. 2013;57:92–8.

    Article  CAS  Google Scholar 

  174. Zhang M, Ding CC, Yaang JH, Lin S, Chen LH, Huang LL. Study of interaction between water-soluble collagen and carboxymethyl cellulose in neutral aqueous solution. Carbohydr Polym. 2016;137:410–7.

    Article  CAS  Google Scholar 

  175. Fu Q, Li GY. Study of improving properties of collagen membrane by blending chitosan and collagen. China Leather. 2007;36(5):22–5.

    CAS  Google Scholar 

  176. Sionkowska A, Wisniewski M, Skopinska S, Kennedy CJ, Wess TJ. Molecular interactions in collagen and chitosan blends. Biomaterials. 2004;25:795–801.

    Article  CAS  Google Scholar 

  177. Mitra T, Sailakshmi G, Gnanamani A, Raja STK, Thiruselvi T, Gowri VM, Selvaraj NV, Ramesh G, Mandal AB. Preparation and characterization of a thermostable and biodegradable biopolymers using natural cross-linker. Int J Biol Macromol. 2011;48(2):276–85.

    Article  CAS  Google Scholar 

  178. Lai GL, Du ZL, Li GY. The rheological behavior of collagen dispersion/poly (vinyl alcohol) blends. Korea Aust Rheol J. 2007;19(2):81–8.

    Google Scholar 

  179. Sarti B, Scandola M. Viscoelastic and thermal properties of collagen/poly (vinyl alcohol) blends. Biomaterials. 1995;16(10):785–92.

    Article  CAS  Google Scholar 

  180. Sionkowska A. Interaction of collagen and poly (vinyl pyrrolidone) in blends. Eur Polym J. 2003;39(11):2135–40.

    Article  CAS  Google Scholar 

  181. Sionkowska A, PɫAnecka A, KozɫOwska J, Skopińska-Wiśniewska J. Collagen fibril formation in poly (vinyl alcohol) and poly (vinyl pyrrolidone) films. J Mol Liquids. 2009;144(1):71–4.

    Article  CAS  Google Scholar 

  182. Xue Y, Liu XH, Ju HY, Dan NH, Dan WH. Effect on structure and performace of type I collagen after modified by polyethylene glycol. China Leather. 2013;42(33):10–4.

    CAS  Google Scholar 

  183. Jose MV, Thomas V, Dean DR, Nyairo E. Fabrication and characterization of aligned nanofibrous PLGA/Collagen blends as bone tissue scaffolds. Polymer. 2009;50(15):3778–85.

    Article  CAS  Google Scholar 

  184. Gorham SD, Light ND, Diamond AM, Willins MJ, Bailey AJ, Wess TJ, Leslie NJ. Effect of chemical modifications on the susceptibility of collagen to proteolysis. II. Dehydrothermal crosslinking. Int J Biol Macromol. 1992;14(3):129–38.

    Article  CAS  Google Scholar 

  185. Weadock KS, Miller EJ, Keuffel EL, Dunn MG. Effect of physical crosslinking methods on collagen-fiber durability in proteolytic solutions. J Biomed Mater Res. 1996;32(2):221–6.

    Article  CAS  Google Scholar 

  186. Hu Y, Liu L, Gu ZP, Dan WH, Dan NH, Yu XX. Synergistic effect of carbodiimide and dehydrothermal crosslinking on acellular dermal matrix. Int J Biol Macromol. 2013;55:221–30.

    Article  CAS  Google Scholar 

  187. Safandowska M, Pietrucha K. Effect of fish collagen modification on its thermal and rheological properties. Int J Biol Macromol. 2013;53:32–7.

    Article  CAS  Google Scholar 

  188. Schroepfer M, Meyer M. DSC investigation of bovine hide collagen at varying degrees of crosslinking and humiditie. Int J Biol Macromol. 2017;103:120–8.

    Article  CAS  Google Scholar 

  189. Shen LR, Xu SC, Wu K, Li GY. Novel method for preparing of fish collagen gels with excellent physicochemical propertiess via the dehhydration of ethanol. Dresden: XXXV IULTCS Congress; 2019.

    Google Scholar 

  190. Sionkowska A, Kamińska A. Thermal helix-coil transition in UV irradiated collagen from rat tail tendon. Int J Biol Macromol. 1999;24(4):337–40.

    Article  CAS  Google Scholar 

  191. Sionkowska A. Thermal denaturation of UV-irradiated wet rat tail tendon collagen. Int J Biol Macromol. 2005;35(3−4):145–9.

    Article  CAS  Google Scholar 

  192. Metreveli NO, Namicheishvili LO, Dzhariashvili KK, Chikvaidze E, Mrevlishvili GM. A microcalorimetric and electron spin resonance study of the influence of UV radiation on collagen. Biophysics. 2006;51(1):29–32.

    Article  Google Scholar 

  193. Hawkins CL, Davies MJ. Generation and propagation of radical reactions on proteins. Biochim Biophys Acta. 2001;1504(2):196–219.

    Article  CAS  Google Scholar 

  194. Inoue N, Bessho M, Furuta M, Kojima T, Okuda S, Hara M. A novel collagen hydrogel cross-linked by gamma-ray irradiation in acidic pH conditions. J Biomater Sci Polym Ed. 2006;17(8):837–58.

    Article  CAS  Google Scholar 

  195. Zhang XM, Xu L, Huang X, Wei S, Zhai M. Structural study and preliminary biological evaluation on the collagen hydrogel crosslinked by γ-irradiation. J Biomed Mater Res A. 2012;100A(11):2960–9.

    Article  CAS  Google Scholar 

  196. Jiang B, Wu ZH, Zhao HC, Taang FY, Lu J, Wei QR, Zhang XD. Electron beam irradiation modification of collagen membrane. Biomaterials. 2006;27(1):15–23.

    Article  Google Scholar 

  197. Usha R, Ramasami T. Effect of crosslinking agents (basic chromium sulfate and formaldehyde) on the thermal and thermomechanical stability of rat tail tendon collagen fibre. Thermochim Acta. 2000;356(1–2):59–66.

    Article  CAS  Google Scholar 

  198. Han B, Jaurequi J, Tang BW, Nimni ME. Proanthocyanidin: a natural crosslinking reagent for stabilizing collagen matrices. J Biomed Mater Res A. 2003;65(1):118–24.

    Article  Google Scholar 

  199. Onem E, Yorgancioglu A, Karavana HA, Yilmaz O. Comparison of different tanning agents on the stabilization of collagen via differential scanning calorimetry. J Therm Anal Calorim. 2017;129(1):615–22.

    Article  CAS  Google Scholar 

  200. Kuroda D, Niinoi M. Design and mechanical proerites of new β- type titanium alloys for implant materials. Mater Sci Eng. 1998;A243:244–9.

    Article  CAS  Google Scholar 

  201. Liu F, Shen LR, Li GY. Cross-linking of collagen sponges with zicronium complex. China Leather. 2012;41(21):24–7.

    Google Scholar 

  202. He LR, Mu CD, Shi JB, Zhang Q, Shi B, Lin W. Modification of collagen with a natural cross-linker, procyanidin. Int J Biol Macromol. 2011;48:354–9.

    Article  CAS  Google Scholar 

  203. Madban B, Subramanian V, Rao R, Nair BU, Ramasami T. Stabilization of collagen using plant polyphenol: role of catechin. Int J Biol Macromol. 2005;3:47–53.

    Google Scholar 

  204. Zhang M, Li CH, Li GY. Dyeing properties of catechin on collagen membrane. China Leather. 2014;43(15):9–12.

    Google Scholar 

  205. Yang H, Shen LR, Bu HH, Li GY. Stable and biocompatible hydrogel composites based on collagen and dialdehyde carboxymethyl cellulose in a biphasic solvent system. Carbohydr Polym. 2019;222:114974.

    Article  CAS  Google Scholar 

  206. Hwang YJ, Larsen J, Krasieva TB, Lyubovitsky JG. Effect of genipin crosslinking on the optical spectral properties and structures of collagen hydrogels. ACS Appl Mater Interfaces. 2011;3(7):2579–84.

    Article  CAS  Google Scholar 

  207. Sung HW, Hsu HL, Shih CC, Lin DS. Cross-linking characteristics of biological tissues fixed with monofunctional or multifunctional epoxy compounds. Biomaterials. 1996;17(14):1405–10.

    Article  CAS  Google Scholar 

  208. Chen YH, Zhang M, Liu WT, Li GY. Properties of alkali-solubilized collagen solution crosslinked by N-hydroxysuccinimide activated adipic acid. Korea Aust Rheol J. 2011;23(1):41–8.

    Article  Google Scholar 

  209. Grabarek Z, Gergely J. Zero-length crosslinking procedure with the use of active esters. Anal Biochem. 1990;185(1):131–5.

    Article  CAS  Google Scholar 

  210. Petite H, Frei V, Huc A, Herbage D. Use of diphenylphosphorylazide for cross-linking collagen-based biornaterials. J Biomed Mater Res. 1994;28:159–65.

    Article  CAS  Google Scholar 

  211. Griffin M, Casadio R, Bergamini CM. Transglutaminases: nature’s biological glues. Biochem J. 2002;368(2):377–96.

    Article  CAS  Google Scholar 

  212. Olde Damink LHH, Dijkstra PJ, Van Luyn MJA, Van Wachem PB, Nieuwenhuis P, Feijen J. Glutaraldehide as a cross-linking agentfor collagen-based biomaterials. J Mater Sci Mater Med. 1995;6(8):460–72.

    Article  CAS  Google Scholar 

  213. Tian ZH, Li CH, Duan L, Li GY. Physicochemical properties of collagen solutions cross-linked by glutaraldehyde. Connect Tissue Res. 2014;55(3):239–47.

    Article  CAS  Google Scholar 

  214. Tian ZH, Liu WT, Li GY. The microstructure and stability of collagen hydrogel cross-linked by glutaraldehyde. Polym Degrad Stabil. 2016;130:264–70.

    Article  CAS  Google Scholar 

  215. Rousseau CF, Gagnieu CH. In vitro cytocompatibility of porcine type I atelocollagen crosslinked by oxidized glycogen. Biomaterials. 2002;23:1503–10.

    Article  CAS  Google Scholar 

  216. Chen YN, Dan NH, Huang YP, Yang CK, Dan WH, Liang YX. Insights into the interactions between collagen and a naturally derived crosslinker, oxidized chitosan oligosaccharide. J Appl Polym Sci. 2019. https://doi.org/10.1002/app.48489.

  217. Hu Y, Liu L, Gu ZP, Dan WH, Dan NH, Yu XX. Modification of collagen with a natural derived cross-linker, alginate dialdehyde. Carbohydr Polym. 2014;102:324–32.

    Article  CAS  Google Scholar 

  218. Pietrucha K, Safandowska M. Dialdehyde cellulose-crosslinked collagen and its physicochemical properties. Process Biochem. 2015;50(12):2105–11.

    Article  CAS  Google Scholar 

  219. Tan H, Wu B, Li CP, Mu CD, Li HL, Lin W. Collagen cryogel cross-linked by naturally derived dialdehyde carboxymethyl cellulose. Carbohydr Polym. 2015;129:17–24.

    Article  CAS  Google Scholar 

  220. Sung HW, Chang Y, Chiu CT, Chen CN, Liang HC. Crosslinking characteristics and mechanical properties of a bovine pericardium fixed with a naturally occurring crosslinking agent. J Biomed Mat Res. 1999;47(2):116–26.

    Article  CAS  Google Scholar 

  221. Sung HW, Liang IL, Chen CN, Huang RN, Liang HF. Stability of a biological tissue fixed with a naturally occurring crosslinking agent (genipin). J Biomed Mater Res. 2001;55(4):538–46.

    Article  CAS  Google Scholar 

  222. Zeeman R, Dijkstra PJ, van Wachem PB, van Luyn MJA, Hendriks M, Cahalan PT, Feijien J. Crosslinking and modification of dermal sheep collagen using 1,4-butanediol diglycidyl ether. J Biomed Mater Res. 1999;46(3):424–33.

    Article  CAS  Google Scholar 

  223. Shen SH, Sung HW, Tu R, Hata C, Lin D, Noishiki Y, Quijano RC. Characterization of a polyepoxy compound fixed porcine heart valve bioprosthesis. J Appl Biomater. 1994;5(2):159–62.

    Article  CAS  Google Scholar 

  224. Zhang M, Li GY. Partial characterization of fish skin collagen cross-linked by N-hydroxysuccinimide activated adipic acid. J Aquat Food Prod Technol. 2014;23(1):44–58.

    Article  CAS  Google Scholar 

  225. Zhang M, Wu K, Li GY. Interactions of collagen molecules in the presence of N-hydroxysuccinimide activated adipic acid (NHS-AA) as a crosslinking agent. Int J Biol Macromol. 2011;49(4):847–54.

    Article  CAS  Google Scholar 

  226. Shen LR, Tian ZH, Liu WT, Li GY. Influence on the physicochemical properties of fish collagen gels using self-assembly and simultaneous cross-linking with the N-hydroxysuccinimide adipic acid derivative. Connect Tissue Res. 2015. https://doi.org/10.3109/03008207.2015.1020941.

  227. Duan L, Liu WT, Tian ZH, Li CH, Li GY. Properties of collagen gels cross-linked by N-hydroxysuccinimide activated adipic acid deriviate. Int J Biol Macromol. 2014;69:482–8.

    Article  CAS  Google Scholar 

  228. Yang JH, Ding CC, Huang LL, Zhang M, Chen LH. The preparation of poly(γ-glutamic acid)-NHS ester as a natural cross-linking agent of collagen. Int J Biol Macromol. 2017;97:1–7.

    Article  CAS  Google Scholar 

  229. Zhang M, Yang JH, Ding CC, Huang LL, Chen LH. A novel strategy to fabricate water-soluble collagen using poly(γ-glutamic acid)-derivatives as dual-functional modifier. React Funct Polym. 2018;122:131–9.

    Article  CAS  Google Scholar 

  230. Zhang M, Yang JH, Deng F, Guo CC, Yang QL, Wu H, Ni YH, Huang LL, Chen LH, Ding CC. Dual-functionalized hyaluronic acid as a facile modifier to prepare polyanionic collagen. Carbohydr Polym. 2019;215:358–65.

    Article  CAS  Google Scholar 

  231. Old Damink LHH, Dijkstra PJ, van Luyn MJA. Cross-linking of dermal sheep collagen using a water-soluble carbodiimide. Biomaterials. 1996;17(8):765–73.

    Article  Google Scholar 

  232. Rault I, Frei V, Herbage D, Abdul-Malak N, Huc A. Evaluation of different chemical methods for cros-linking collagen gel, films and sponges. J Mater Sci Mater Med. 1996;7(4):215–21.

    Article  CAS  Google Scholar 

  233. Folk JE. Transglutaminases. Ann Rev Biochem. 1980;49:517–31.

    Article  CAS  Google Scholar 

  234. Chen RN, Ho HO, Sheu MT. Characterization of collagen matrices crosslinked using microbial transglutaminase. Biomaterials. 2005;26(20):4229–35.

    Article  CAS  Google Scholar 

  235. Cheng S, Wang WH, Li Y, Gao GX, Zhang K, Zhou JY, Wu ZN. Cross-linking and film-forming properties of transglutaminase-modified collagen fibers tailored by denaturation temperature. Food Chem. 2019;271:527–35.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

The research is funded by the National Natural Science Foundation of China (Nos. 21776184 and 21606156).

Author information

Authors and Affiliations

Authors

Contributions

GYL made substantial contributions to conception and design of this review, and critical revised the manuscript. XXZ wrote and revised the manuscript. SCX participated in the design of this review. LRS revised the draft from the language and logicality aspects. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Guoying Li.

Ethics declarations

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, X., Xu, S., Shen, L. et al. Factors affecting thermal stability of collagen from the aspects of extraction, processing and modification. J Leather Sci Eng 2, 19 (2020). https://doi.org/10.1186/s42825-020-00033-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s42825-020-00033-0

Keywords