- Email: [email protected]
Photon correlation spectroscopy applied to tear analysis
Colloids and Surfaces B: Biointerfaces 157 (2017) 26–30
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Protocols
Photon correlation spectroscopy applied to tear analysis S. Picarazzi a , M. Lecchi b,c , V. Pastori b,c , M. D’Arienzo a,c , R. Scotti a,c , S. Tavazzi a,c,∗ a
University of Milano Bicocca, Materials Science Department, Via R. Cozzi 55, I-20125 Milan, Italy University of Milano Bicocca, Department of Biotechnology and Bioscience, Piazza Della Scienza 2, I-20126 Milan, Italy c University of Milano Bicocca, COMiB, Via R. Cozzi 55, I-20125 Milan, Italy b
a r t i c l e
i n f o
Article history: Received 21 February 2017 Received in revised form 19 May 2017 Accepted 22 May 2017 Available online 24 May 2017 Keywords: Tear Photon correlation spectroscopy Electron spin resonance Contact lens
a b s t r a c t This study aims to deepen the knowledge on tear film properties by the development of a protocol for analyses of Photon Correlation Spectroscopy (PCS) on human tears and by the comparison between PCS results obtained on tears of contact lens wearers and non-wearers. Tears (5 L) were collected by a glass capillary. The analyses provide the hydrodynamic diameter of tear components by analyzing intensity fluctuations in time of scattered light. PCS appears a promising technique for studying tear features and for shedding light on specific eye conditions, such as on the clinical effects of CL wear. In fact, statistical difference (p < 0.001) was found between the measured mean hydrodynamic diameter of tear components of wearers and non-wearers, the resulting value significantly higher for CL wearers. The scenario does not substantially change after (25 ± 5) min from the CL removal. The difference is attributed to changes in the interactions between tear constituents due to CL wear. In order to get deeper insights on the influence of CL wear on aggregation and structure of tear components, a preliminary Electron Spin Resonance (ESR) investigation was performed, monitoring Fe3+ species. ESR spectra on tears of both CL wearers and non-wearers showed the presence of intense signals, probably associated to iron (III) centers in proteins such as lactoferrin, and a weaker resonance attributable to Fe3+ species interacting with S–S bridges of lysozyme. Differences in ESR spectra between CL wearers and non-wearers were detected and tentatively ascribed to changes in coordination or in local environment of Fe3+ centers connected to aggregation phenomena induced by CL wear, which promote their interaction with other neighboring iron species. © 2017 Published by Elsevier B.V.
1. Introduction The tear film is a highly specialized structure that performs several functions such as optical, mechanical, and defensive functions. Its volume is typically (6 ± 2) L [1]. The tear production was estimated to be 1–2 L/min and increases up to 4 L/min after stimulation of the ocular surface [2]. Under normal physiological conditions, the thickness of tear film is a few micrometers [3]. After blinking the thickness is reduced by about 20% after 5 s and it is reduced by 50% after 30 s [4]. The average refractive index is 1.337 [5]. The pH is slightly alkaline, with an average value between 7.4 and 7.5 [6]. The osmolarity, under normal conditions, is expected to be lower than 320 mOsm/kg [7]. The properties of the tear film change during the day [8,9]; some of these variations can depend on the method used to extract tears from the eye [10,11]. From the
∗ Corresponding author at: University of Milano Bicocca, COMiB & Materials Science Department, Via R. Cozzi 55, I-20125 Milan, Italy. E-mail address: [email protected] (S. Tavazzi). http://dx.doi.org/10.1016/j.colsurfb.2017.05.057 0927-7765/© 2017 Published by Elsevier B.V.
biochemical point of view, the main components of tear film are lipids, proteins, mucins and electrolytes [12]. The application of a contact lens (CL) can change the structure, the composition, the physical and chemical properties and the behavior of the normal tear film. Many studies have been reported in the literature concerning the properties of new and worn CLs [13–20]. As far as the tear properties of CL wearers are concerned, Bahgat showed that tear break up times and tear production are affected by CL wear [21]. Guillon et al. [22] demonstrated that wearing CLs produces a greater evaporation than that experienced by non-wearers. This behavior may be associated to an intrinsic change of the tear film, the effect being partially present also 24 h after CL wear. Another variation observed in the tear film was an increase of the ocular surface temperature immediately following CL wear, the temperature being significantly higher compared to non-wearers (37.1 ± 1.7 ◦ C versus 35.0 ± 1.1 ◦ C), predominantly in the continuous wear group (38.6 ± 1.0 ◦ C) [23]. Similar trends were also observed in a study reported many years ago by Martin et al. [24]. Remarkably, Reim and Schrage [25] showed that wearing CLs causes an increase of glucose and lactate levels in the cornea
S. Picarazzi et al. / Colloids and Surfaces B: Biointerfaces 157 (2017) 26–30
and in tears, while ATP and glycogen in the cornea were found to decrease. The presence of CLs can also affect the properties of lacrimal proteins, which can undergo structural modifications such as unfolding, as recently discussed by Mann and Tighe [26]. This process also causes the loss of their biological functions and the formation of protein aggregates. The same authors also discussed the possible variation both in type and concentration of the tear film components, the immobilization of lipid on the CL surface and their possible oxidative degradation, the deposition on the CL and denaturation of proteins, the stimulation of cascade processes leading either to the generation of additional proteins and peptides or to an increase in fraction of components. In general, there are significant gaps in the understanding to what extent CLs induce tear changes and, in turn, to what extent these changes are responsible for discomfort and other negative effects. Prompted by this background and aiming to better understand tear properties and the consequences of CL wear, we have developed a new methodological tool to reveal tear film changes between CL wearers and non-wearers based on Photon Correlation Spectroscopy (PCS), also known as Dynamic Light Scattering (DLS). Azharuddin et al. [27] recently reported the results of DLS analyses on tears collected by Schirmer’s strips, extracted and centrifugated. They found differences between patients with dry eye and healthy subjects and attributed these differences to a higher abundance of aggregated proteins in dry eye patients. Apart from these preliminary outcomes, many aspects deserve to be investigated. In this work, a novel method for tear analyses by PCS has been established. The method was applied for measuring the tear properties of CLs wearers and non-wearers. In detail, the hydrodynamic diameter of the main tear components was determined and a comprehensive comparison between the results obtained for CLs wearers and non-wearers was achieved. In order to get deeper insights on the influence of CLs wear on the aggregation and structure of tear constituents, in particular proteins, a preliminary Electron Spin Resonance (ESR) investigation was performed, to our knowledge for the first time, on both tears of CL wearers and non-wearers, monitoring the Fe3+ species. The results, besides revealing the presence of different iron centers associated to proteins and enzymes, supported the PCS data and highlighted the validity of ESR for characterizing metal binding sites in tears.
2. Materials and methods 2.1. Tear collection Tears of 63 subjects were analyzed, 24 wearers of CLs (14 females and 10 males, age interval: 21–34 years) and 39 nonwearers (23 females and 16 males, age interval: 20–45 years). CL wearers wore CLs of different materials. The hydrogel materials belong to II FDA group (nelfilcon A, omafilcon A, hilafilcon B) and IV FDA group (etafilcon A, methafilcon A). Silicone-hydrogels are comfilcon A, lotrafilcon B, narafilcon A, galyfilcon A, balafilcon A, delefilcon A, senofilcon A. The procedure used for the collection of tears was to place, parallel to the lower meniscus tear, a glass capillary of 5 L. The capillary was filled by capillarity in a few minutes. For CL wearers, tears were collected with CLs in-situ and after (25 ± 5) min from their removal. The measurements were performed on samples of 5 L diluted with 45 mg (45 L) of deionized water (tear concentration (10 ± 2)%V /W ). The diluted tears were placed in 50 L cells. On each tear sample, three PCS acquisitions were carried out three times (nine acquisitions). For each triple acquisition, the Malvern software provided an average result and a quality report obtained by a number of quality tests on the three measurements. The data reported in the paper were taken from one triple acquisition, but
27
similar data were found in either two out of three or three out of three triple acquisitions. During the development of the PCS protocol, analyses were also performed on solutions with tear concentration (5 ± 1)%V /W and (3 ± 2)%V /W . PCS results were found to be independent on tear concentration. Lower tear concentrations could not be used because there was not enough scattered light to make measurements. Tear concentrations higher than (20 ± 2)%V /W were not taken into consideration due to the scarce repeatability of PCS results. 2.2. Photon correlation spectroscopy analyses The hydrodynamic diameter (dH ) of the main components of tear content was determined using a DLS Malvern Zetasizer ZS90 instrument. The Zetasizer system detects the Brownian motion of particles suspended in a solution by illuminating them with a laser and analyzing the intensity fluctuations of the scattered light as a function of time. The speed of movement is then used to determine dH , which is the diameter of hard spheres that would diffuse light at the same speed as the particles being measured. The relationship between dH and the particle speed is defined in the Stokes-Einstein kT equation dH = 3D , where k is the Boltzmann’s constant, T is the absolute temperature, is the viscosity, and D is the translational diffusion coefficient. The result is a particle size distribution calculated from the intensity of scattered light. The size distribution is displayed as relative intensity of light scattered by particles versus particle diameter. The most frequently used number to define dH is called z-average (dH,avg ). It is the intensity-weighted mean of the diameters calculated from the intensity particle size distribution. A second parameter is the polydispersity index (Ipd ), which describes the width of the intensity distribution. We also deduced dH,avg from (i) the fitting of the main peak of the intensity distribution and (ii) the particle size corresponding to maximum intensity of scattered light. These results are omitted here because a similar scenario was found as deduced from z-average. Although the fundamental size distribution generated by PCS is an intensity distribution, this can be converted, using Mie theory, to a volume distribution. This volume distribution can also be further converted to a number distribution. Here, volume and number particle size distributions are not taken into consideration because generated from the primary size distribution, which is the intensity one. Volume and number distributions are typically only recommended for comparison between relative amounts of different particles. The comparison of dH,avg for CL wearers and non-wearers was obtained keeping the same experimental conditions: the temperature was set to 25 ◦ C, the selected stabilization time was 60 s, the viscosity and the refractive index of solvent were assumed to be those of water (n = 1.330, v = 0.8872 cP). The algorithm used by the software for calculating the size distribution was the “General Purpose – NNLS (Non-negative least squares)” [28]. 2.3. Electron spin resonance investigation The electron spin resonance (ESR) investigation was performed by a Bruker EMX spectrometer operating at the X-band frequency and equipped with an Oxford cryostat working in the temperature range of 4–298 K. The capillary containing the tears samples were charged in quartz glass tubes and the spectra were recorded at 298 K. The spectra of pure lactoferrin were recorded at 298 K on a solution of the protein in a phosphate buffer (0.02 M at pH = 7.6). Three samples of lactoferrin in phosphate buffer solution were analyzed and the results were substantially equivalent. The g values were calculated by standardization with 2,2-diphenyl1-picrylhydrazyl (DPPH). Care was taken to always keep the most sensitive part of the ESR cavity (1 cm length) filled.
28
S. Picarazzi et al. / Colloids and Surfaces B: Biointerfaces 157 (2017) 26–30 Table 2 p-values of Student’s t statistics (t-test) to assess the statistical difference between mean values of dH,avg of paired (P) or unpaired (U) groups.
Non-wearers CL wearers (CL in-situ)
CL wearers (CL in-situ)
CL wearers (after 25 min from CL removal)
<0.0001 (U)
0.0009 (U) 0.7089 (P)
Table 3 p-values of Student’s t statistics (t-test) to assess the statistical difference between mean values of Ipd data of paired (P) or unpaired (U) groups.
Non-wearers CL wearers (CL in-situ)
CL wearers (CL in-situ)
CL wearers (after 25 min from CL removal)
0.5135 (U)
0.5972 (U) 0.9923 (P)
distributions for non-wearers, CL wearers with the CL in-situ, and the same CL wearers after (25 ± 5) min from CL removal. Mean and standard deviation deduced from these data are reported in Table 1. Statistical significance of difference among groups was assessed by Student’s t statistic (t-test). The p-values are reported in Tables 2 and 3. 3.2. Electron spin resonance investigation ESR investigation has been initially performed on pure lactoferrin in phosphate buffer solution at 298 K, as a reference sample for studying the presence of iron centers in human tears (Fig. 2a). This protein is highly abundant in tears and it has high affinity in binding Fe3+ centers, conferring to the system relevant bacteriostatic and antioxidant properties [29–31]. After this preliminary investigation on lactoferrin, ESR spectra were carried out directly on human tears of one CL wearer and one non-wearer (Fig. 2b). 4. Discussion
Fig. 1. Statistical frequency distributions of dH,avg and Ipd data for non-wearers, for CL wearers with the CL in-situ, and for the same CL wearers after (25 ± 5) min from CL removal.
3. Results 3.1. Photon correlation spectroscopy analyses Fig. 1 shows the statistical frequency distributions for CL wearers and non-wearers of dH,avg and Ipd . Each figure shows the three Table 1 Mean and standard deviation (SD) of dH,avg and Ipd for non-wearers, for CL wearers with the CL in-situ, and for the same CL wearers after (25 ± 5) min from CL removal.
Non-wearers CL wearers (CL in-situ) CL wearers (after 25 min from CL removal)
dH,avg (nm)
Ipd
mean
SD
mean
SD
258.9 316.9 312.4
30.7 55.9 66.9
0.220 0.209 0.209
0.072 0.062 0.086
As far as PCS results are concerned, the first point to discuss is the significance of dH,avg . As already mentioned, the standard report for size measurements is an intensity distribution, while volume and number distributions are generated using established theories of light scattering. When the sample is monomodal (i.e. only one peak is detected) and monodisperse (i.e. the width of the distribution is negligible) and dispersed particles are spherical, the conversion from intensity to volume and number is less critical. Such samples typically have polydispersity Ipd below 0.1. On the contrary, the technique is not applicable in case of polydispersity higher than 0.5. For intermediate polydispersity, dH,avg can be deduced only from the intensity distribution and the obtained results can only be used for comparative purposes between different samples. Polydispersity of tear samples was found to be about 0.2 (Fig. 1b and Table 1). Therefore, for tear samples, dH,avg deduced from the intensity distribution can be used for comparative studies, as it is often the case for emulsions, polymers, proteins, and other samples [28]. Both for CL wearers and for non-wearers, dH,avg was found to be few hundreds of nanometers, but the above considerations clarify that this diameter is not attributable to a specific particle size of tear content. Moreover, in general, the hydrodynamic diameter depends, not only on the size of a particle core, but also on any surface structure, as well as ions in the medium. Indeed, ions in a medium are expected to influence the Brownian motion of suspended particles, namely PCS results. This aspect is expected to influence PCS results on tears, due to their complex composition. Another aspect which may affect PCS results is restricted diffusion due to high concentration. Restricted diffusion is due to the presence of other particles which hinders free particle Brownian motion. This can
S. Picarazzi et al. / Colloids and Surfaces B: Biointerfaces 157 (2017) 26–30
Fig. 2. ESR spectra of (a) pure lactoferrin in phosphate buffer solution and (b) tears of non wearers (i, black line) and wearers (ii, red line) of CLs (senofilcon A). A rectangle in (b) highlights the spectral region where a change of ESR signal for CL wearer is detectable.
be excluded for the tear samples under investigation with concentration (10 ± 2) V /W because no substantial differences were found varying the concentration from (3 ± 1) V /W to (10 ± 2) V /W . A clear statistical difference (p< 0.0001) was found between CL wearers with CL in-situ and non-wearers, dH,avg being ∼60 nm higher for CL wearers. A similar scenario was detected after 25 min from CL removal with a similar mean difference of about 60 nm compared to non-wearers. Although, as discussed above, the dH,avg measured number is not real, the difference has a physical meaning, which is attributed to a change of the state of aggregation/interaction of tear content due to CL wear. Indeed, PCS is very sensitive to the presence of aggregates and to interactions between suspended particles, which modify their free motion. Interestingly, the scenario is similar when the CL is in-situ and after 25 min from its removal. When a CL is placed on the eye, it strongly alters the arrangement of the tear film. Indeed, the tear film is a layer of few micrometers covering conjunctival and corneal epithelia, while CLs have a thickness of tens or hundreds of micrometers. Our results are not attributable to the presence of a CL placed on the eye during tear collection, but to an intrinsic change of the tear film proper-
29
ties observed also after its removal. These PCS results confirm the hypothesis of intrinsic changes of the tear film reported by Guillon et al. [22], who measured an higher tear evaporation rate for CL wearers compared to non-wearers, even after 1 day without CL. ESR preliminary results confirmed these interpretations. In the ESR spectrum of lactoferrin (Fig. 2a), an intense signal at g = 4.3 is easily detectable. As reported by Ainscough et al. [32] the position and the lineshape allows to attribute this resonance to Fe3+ centers with orthorhombic symmetry [32,33]. The weakest signals at g ∼ 6 and g ∼ 2 are instead associated to iron (III) species with axial symmetry [34]. These spectral features can be connected to the three dimensional structure of protein, which presents two homologous lobes binding a single Fe3+ ion. Similarly to lactoferrin, tears (Fig. 2b) show the signal at g = 4.3 of orthorhombic Fe3+ centers and the resonances of iron species with axial symmetry (g ∼ 6 and g ∼ 2). However, the intensity and the intensity, broadness and asymmetry of these features are much higher, suggesting that iron ions are more magnetically diluted and less coupled than in pure lactoferrin. Moreover, regardless the CLs wear, the spectrum of tears shows also the presence of an isotropic weak signal at g = 1.92, which, instead, is not detectable in pure lactoferrin. This signal is usually observed in Fe-S proteins and it has been associated to Fe-S clusters [34]. In the present case, we may attribute this signal to Fe-S clusters generated by the interaction of the Fe3+ centers with the S–S bridges of lysozyme, particularly abundant in the tear film [35]. However, further investigations are ongoing to clarify the origin of this feature. By comparing the spectra of tears of CL wearers and non-wearers, the signal intensity for the CL wearer appears decreased, in particular as concerns the resonances attributed to magnetically diluted Fe3+ centers. More in detail, the signal at g = 4.3 widens, almost overlapping the resonance at g ∼ 6 related to the species with axial symmetry. This overlapping can be detected by looking at the slope of the curves (their first derivative) in the range 1000–1100 G in Fig. 2b. Only the black curve (i, non-wearer) shows a minimum at ∼1000 G, while the red curve shows positive slope in the whole spectral range. These variations in intensity and width of the ESR spectral lines are usually attributable to a change of the surroundings of the paramagnetic centers and/or to their interaction (magnetic coupling) with neighboring radical species. In this case, we may suggest the occurrence of a change in the coordination or in the local environment of Fe3+ centers connected to aggregation phenomena of the proteins in the lacrimal film induced by the wear of CLs. This aggregation may promote magnetic coupling among the paramagnetic centers and, in turn, a consequent decay in intensity and line-broadening of the ESR signals. These preliminary results seem to support the differences observed by PCS between wearers and non-wearers and suggest possible modifications of the aggregation state of the tear components induced by CL wear.
5. Conclusions PCS detects intensity fluctuations of scattered light by a sample. It is applicable to analyze human tears for comparative purposes. It measures the polydispersity index and the mean hydrodynamic diameter of tear content. This diameter is not attributable to the size of a specific component, but it provides information on the presence of aggregates and interactions between tear constituents, which modify their free motion. The sensitivity of the technique makes it possible to discriminate between CL wearers and nonwearers (p < 0.0001). The hydrodynamic diameter was measured to be (316.9 ± 55.9) nm with the CL in-situ and (312.4 ± 66.9) nm after (25 ± 5) min from their removal, to be compared with (258.9 ± 30.7) nm for non-wearers. For future developments, PCS could be extensively applied to compare the clinical effects of different types of CLs and to develop more compatible CL materials, CL maintenance solu-
30
S. Picarazzi et al. / Colloids and Surfaces B: Biointerfaces 157 (2017) 26–30
tions, and artificial tears. The correlation with ocular pathologies could also be explored for diagnostic purposes. Preliminary ESR results support the differences observed by PCS between CL wearers and non-wearers. In fact, ESR signals with weaker intensity and an increased broadness were observed for tears of CL wearers compared to those of non-wearers. This behavior may be associated to changes in the coordination or in the local environment of Fe3+ centers, probably connected to aggregation phenomena of the tear proteins induced by CL wear. Acknowledgment The authors gratefully acknowledge prof. Alessandro Borghesi (University of Milano Bicocca) for helpful discussion and for critically reading the paper. References [1] S. Mishima, A. Gasset, S.D. Klyce, J.L. Baum, Determination of tear volume and tear flow, Invest. Ophthalmol. Visual Sci. 5 (1996) 264–276. [2] L. Battat, A. Macri, D. Dursun, S.C. Pflugfelder, Effects of laser in situ keratomileusis on tear production, clearance, and the ocular surface, Ophalmology 108 (2001) 1230–1235. [3] P.E. King-Smith, B.A. Fink, N. Fogt, K.K. Nichols, R.M. Hill, G.S. Wilson, The thickness of the human precorneal tear film: evidence from reflection spectra, Invest. Ophthalmol. Visual Sci. 41 (2000) 3348–3359. [4] S. Mishima, Some physiological aspects of the precorneal tear film, Arch. Ophthalmol. (1965) 233–241. [5] S. Patel, K.E. Boyd, J. Burns, Age, stability of the precorneal tear film and the refractive index of tears, Cont. Lens Ant. Eye 23 (2000) 44–47. [6] M. Yamada, M. Kawai, H. Mochizuki, Y. Hata, Y. Mashima, Fluorophotometric measurement of the buffering action of human tears in vivo, Curr. Eye Res. 17 (1997) 482–486. [7] J.E. Terry, R.M. Hill, Human tear osmotic pressure diurnal variations and the closed eye, Arch. Ophthalmol. 96 (1978) 120–122. [8] V. Ng, P. Cho, S. Mak, A. Lee, Variability of tear protein levels in normal young adults: between-day variation, Graefes Arch. Clin. Exp. Ophthalmol. 238 (2000) 892–899. [9] R.A. Sack, K.O. Tan, A. Tan, Diurnal tear cycle: evidence for a nocturnal inflammatory constitutive tear fluid, Invest. Ophthalmol. Visual Sci. 33 (1992) 626–640. [10] V. Ng, P. Cho, The relationship between total tear protein concentrations determined by different methods and standards, Graefes Arch. Clin. Exp. Ophtalmol. 238 (2000) 571–576. [11] A. Posa, L. Bräuer, M. Schicht, F. Garreis, S. Beileke, F. Paulsen, Schirmer strip vs. capillary tube method: non-invasive methods of obtaining proteins from tear fluid, Ann. Anat. 195 (2013) 137–142. [12] A.M. Bright, B.J. Tighe, The composition and interfacial properties of tears: tear substitutes and tear models, J. Br. Contact Lens Assoc. 16 (1993) 57–66. [13] J. Baguet, F. Somer, V. Claudon-Eyl, T.M. Duc, Characterization of lachrymal component accumulation on worn soft contact lens surface by atomic force microscopy, Biomaterials 16 (1995) 3–9.
[14] J.H. Teichroeb, J.A. Forrest, V. Ngai, J.W. Martin, L. Jones, J. Medley, Imaging protein deposits on contact lens materials, Optom. Vis. Sci. 85 (2008) 1151–1164. [15] M. Lira, L. Santos, E. Yebra-Pimentel, M. Elisabete, C.D. Real Oliveira, Comparative study of silicone-hydrogel contact lenses surfaces before and after wear using atomic force microscopy, J. Biomed. Mater. Res. B: Appl. Biomater. 85 (2008) 361–367. [16] G. Young, R. Garofalo, O. Harmer, S. Peters, The effect of soft contact lens care products on lens modulus, Contact Lens Anterior Eye 33 (2010) 210–214. [17] M. Berry, C. Purslow, P.J. Murphy, H. Pult, Contact lens materials, mucin fragmentation and relation to symptoms, Cornea 31 (2012) 770–776. [18] M. Bettuelli, S. Trabattoni, M. Fagnola, S. Tavazzi, L. Introzzi, S. Farris, Surface properties and wear performances of siloxane-hydrogel contact lenses, J. Biomed. Mat. Res. B: Appl. Biomater. 101B (2013) 1585–1593. [19] M. Lira, S. Franco, J.B. Vazquez-Dorrio, José Benito, M.E. Real Oliveira, M.F. Costa, Surface roughness and refractive index changes in contact lens induced by lens care systems, Eye Contact Lens 40 (2014) 140–147. [20] S. Tavazzi, M. Tonveronachi, M. Fagnola, F. Cozza, L. Ferraro, A. Borghesi, M. Ascagni, S. Farris, Wear effects on microscopic morphology and hyaluronan uptake in siloxane-hydrogel contact lenses, J. Biomed. Mater. Res. B: Appl. Biomater. 103 (2015) 1092. [21] Mostafa M. Bahgat, Precorneal tear film changes due to soft contact lens wear, Indian J. Ophthalmol. 33 (1985) 177–179. [22] M. Guillon, C. Maissa, Contact lens wear affects tear film evaporation, Eye Contact Lens 34 (2008) 326–330. [23] C. Purslow, J. Wolffsohn, J. Rubido, The effect of contact lens wear on dynamic ocular surface temperature, Contact Lens Anterior Eye 28 (2005) 29–36. [24] D. Martin, I. Fatt, The presence of a contact lens induces a very small increase in the corneal surface temperature, Acta Ophthalmol. 64 (1986) 512–518. [25] M. Reim, N.F. Schrage, J. Becker, Interactions between ocular surface fluid and cornea related to contact lenses, Eur. J. Ophthalmol. 11 (2001) 105–115. [26] A. Mann, B. Tighe, Contact lens interactions with the tear film, Exp. Eye Res. 117 (2013) 88–92. [27] M. Azharuddin, J. Khandelwal, H. Datta, A.K. Dasgupta, S.O. Raja, Dry eye a protein conformational disease, Invest. Ophthalmol. Vis. Sci. 56 (2015) 1423–1429. [28] User manual zetasizer nanoseries, Malvern (2013). [29] J.V. Bannister, W.H. Bannister, H.A.O. Hill, P.J. Thornalley, Enhanced production of hydroxyl radicals by the xanthine- xanthine oxidase reaction in the presence of lactoferrin, Biochim. Biophys. Acta 715 (1982) 116–120. [30] H. Shinmoto, S. Dosako, I. Nakajima, Antioxidant activity of bovine lactoferrin on iron/ascorbate induced lipid peroxidation, Biosci. Biotechnol. Biochem. 56 (1992) 2079–2080. [31] V. Pastori, S. Tavazzi, M. Lecchi, Lactoferrin-loaded contact lenses: eye protection against oxidative stress, Cornea 34 (2015) 693–697. [32] E.W. Ainscough, A.M. Brodie, J.E. Plowman, S.J. Bloor, J.S. Loehr, T.M. Loehr, Studies on human lactoferrin by electron paramagnetic resonance, fluorescence, and resonance Raman spectroscopy, Biochemistry 19 (1980) 4072–4079. [33] G.G. Borisenko, V.E. Kagan, C.J. Hsia, N.F. Schor, Interaction between 6-hydroxydopamine and transferrin: let my iron go, Biochemistry 39 (2000) 3392–3400. [34] R. Cammack, C. Cooper, Electron paramagnetic resonance spectroscopy of iron complexes and iron-containing proteins, Methods Enzymol. 227 (1993) 353–384. [35] S. Patel, L. Anderson, K. Cairney, The refractive index of tears in normals and soft lens wearers, Adv. Exp. Med. Biol. 350 (1994) 417–420.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Protocols
Photon correlation spectroscopy applied to tear analysis S. Picarazzi a , M. Lecchi b,c , V. Pastori b,c , M. D’Arienzo a,c , R. Scotti a,c , S. Tavazzi a,c,∗ a
University of Milano Bicocca, Materials Science Department, Via R. Cozzi 55, I-20125 Milan, Italy University of Milano Bicocca, Department of Biotechnology and Bioscience, Piazza Della Scienza 2, I-20126 Milan, Italy c University of Milano Bicocca, COMiB, Via R. Cozzi 55, I-20125 Milan, Italy b
a r t i c l e
i n f o
Article history: Received 21 February 2017 Received in revised form 19 May 2017 Accepted 22 May 2017 Available online 24 May 2017 Keywords: Tear Photon correlation spectroscopy Electron spin resonance Contact lens
a b s t r a c t This study aims to deepen the knowledge on tear film properties by the development of a protocol for analyses of Photon Correlation Spectroscopy (PCS) on human tears and by the comparison between PCS results obtained on tears of contact lens wearers and non-wearers. Tears (5 L) were collected by a glass capillary. The analyses provide the hydrodynamic diameter of tear components by analyzing intensity fluctuations in time of scattered light. PCS appears a promising technique for studying tear features and for shedding light on specific eye conditions, such as on the clinical effects of CL wear. In fact, statistical difference (p < 0.001) was found between the measured mean hydrodynamic diameter of tear components of wearers and non-wearers, the resulting value significantly higher for CL wearers. The scenario does not substantially change after (25 ± 5) min from the CL removal. The difference is attributed to changes in the interactions between tear constituents due to CL wear. In order to get deeper insights on the influence of CL wear on aggregation and structure of tear components, a preliminary Electron Spin Resonance (ESR) investigation was performed, monitoring Fe3+ species. ESR spectra on tears of both CL wearers and non-wearers showed the presence of intense signals, probably associated to iron (III) centers in proteins such as lactoferrin, and a weaker resonance attributable to Fe3+ species interacting with S–S bridges of lysozyme. Differences in ESR spectra between CL wearers and non-wearers were detected and tentatively ascribed to changes in coordination or in local environment of Fe3+ centers connected to aggregation phenomena induced by CL wear, which promote their interaction with other neighboring iron species. © 2017 Published by Elsevier B.V.
1. Introduction The tear film is a highly specialized structure that performs several functions such as optical, mechanical, and defensive functions. Its volume is typically (6 ± 2) L [1]. The tear production was estimated to be 1–2 L/min and increases up to 4 L/min after stimulation of the ocular surface [2]. Under normal physiological conditions, the thickness of tear film is a few micrometers [3]. After blinking the thickness is reduced by about 20% after 5 s and it is reduced by 50% after 30 s [4]. The average refractive index is 1.337 [5]. The pH is slightly alkaline, with an average value between 7.4 and 7.5 [6]. The osmolarity, under normal conditions, is expected to be lower than 320 mOsm/kg [7]. The properties of the tear film change during the day [8,9]; some of these variations can depend on the method used to extract tears from the eye [10,11]. From the
∗ Corresponding author at: University of Milano Bicocca, COMiB & Materials Science Department, Via R. Cozzi 55, I-20125 Milan, Italy. E-mail address: [email protected] (S. Tavazzi). http://dx.doi.org/10.1016/j.colsurfb.2017.05.057 0927-7765/© 2017 Published by Elsevier B.V.
biochemical point of view, the main components of tear film are lipids, proteins, mucins and electrolytes [12]. The application of a contact lens (CL) can change the structure, the composition, the physical and chemical properties and the behavior of the normal tear film. Many studies have been reported in the literature concerning the properties of new and worn CLs [13–20]. As far as the tear properties of CL wearers are concerned, Bahgat showed that tear break up times and tear production are affected by CL wear [21]. Guillon et al. [22] demonstrated that wearing CLs produces a greater evaporation than that experienced by non-wearers. This behavior may be associated to an intrinsic change of the tear film, the effect being partially present also 24 h after CL wear. Another variation observed in the tear film was an increase of the ocular surface temperature immediately following CL wear, the temperature being significantly higher compared to non-wearers (37.1 ± 1.7 ◦ C versus 35.0 ± 1.1 ◦ C), predominantly in the continuous wear group (38.6 ± 1.0 ◦ C) [23]. Similar trends were also observed in a study reported many years ago by Martin et al. [24]. Remarkably, Reim and Schrage [25] showed that wearing CLs causes an increase of glucose and lactate levels in the cornea
S. Picarazzi et al. / Colloids and Surfaces B: Biointerfaces 157 (2017) 26–30
and in tears, while ATP and glycogen in the cornea were found to decrease. The presence of CLs can also affect the properties of lacrimal proteins, which can undergo structural modifications such as unfolding, as recently discussed by Mann and Tighe [26]. This process also causes the loss of their biological functions and the formation of protein aggregates. The same authors also discussed the possible variation both in type and concentration of the tear film components, the immobilization of lipid on the CL surface and their possible oxidative degradation, the deposition on the CL and denaturation of proteins, the stimulation of cascade processes leading either to the generation of additional proteins and peptides or to an increase in fraction of components. In general, there are significant gaps in the understanding to what extent CLs induce tear changes and, in turn, to what extent these changes are responsible for discomfort and other negative effects. Prompted by this background and aiming to better understand tear properties and the consequences of CL wear, we have developed a new methodological tool to reveal tear film changes between CL wearers and non-wearers based on Photon Correlation Spectroscopy (PCS), also known as Dynamic Light Scattering (DLS). Azharuddin et al. [27] recently reported the results of DLS analyses on tears collected by Schirmer’s strips, extracted and centrifugated. They found differences between patients with dry eye and healthy subjects and attributed these differences to a higher abundance of aggregated proteins in dry eye patients. Apart from these preliminary outcomes, many aspects deserve to be investigated. In this work, a novel method for tear analyses by PCS has been established. The method was applied for measuring the tear properties of CLs wearers and non-wearers. In detail, the hydrodynamic diameter of the main tear components was determined and a comprehensive comparison between the results obtained for CLs wearers and non-wearers was achieved. In order to get deeper insights on the influence of CLs wear on the aggregation and structure of tear constituents, in particular proteins, a preliminary Electron Spin Resonance (ESR) investigation was performed, to our knowledge for the first time, on both tears of CL wearers and non-wearers, monitoring the Fe3+ species. The results, besides revealing the presence of different iron centers associated to proteins and enzymes, supported the PCS data and highlighted the validity of ESR for characterizing metal binding sites in tears.
2. Materials and methods 2.1. Tear collection Tears of 63 subjects were analyzed, 24 wearers of CLs (14 females and 10 males, age interval: 21–34 years) and 39 nonwearers (23 females and 16 males, age interval: 20–45 years). CL wearers wore CLs of different materials. The hydrogel materials belong to II FDA group (nelfilcon A, omafilcon A, hilafilcon B) and IV FDA group (etafilcon A, methafilcon A). Silicone-hydrogels are comfilcon A, lotrafilcon B, narafilcon A, galyfilcon A, balafilcon A, delefilcon A, senofilcon A. The procedure used for the collection of tears was to place, parallel to the lower meniscus tear, a glass capillary of 5 L. The capillary was filled by capillarity in a few minutes. For CL wearers, tears were collected with CLs in-situ and after (25 ± 5) min from their removal. The measurements were performed on samples of 5 L diluted with 45 mg (45 L) of deionized water (tear concentration (10 ± 2)%V /W ). The diluted tears were placed in 50 L cells. On each tear sample, three PCS acquisitions were carried out three times (nine acquisitions). For each triple acquisition, the Malvern software provided an average result and a quality report obtained by a number of quality tests on the three measurements. The data reported in the paper were taken from one triple acquisition, but
27
similar data were found in either two out of three or three out of three triple acquisitions. During the development of the PCS protocol, analyses were also performed on solutions with tear concentration (5 ± 1)%V /W and (3 ± 2)%V /W . PCS results were found to be independent on tear concentration. Lower tear concentrations could not be used because there was not enough scattered light to make measurements. Tear concentrations higher than (20 ± 2)%V /W were not taken into consideration due to the scarce repeatability of PCS results. 2.2. Photon correlation spectroscopy analyses The hydrodynamic diameter (dH ) of the main components of tear content was determined using a DLS Malvern Zetasizer ZS90 instrument. The Zetasizer system detects the Brownian motion of particles suspended in a solution by illuminating them with a laser and analyzing the intensity fluctuations of the scattered light as a function of time. The speed of movement is then used to determine dH , which is the diameter of hard spheres that would diffuse light at the same speed as the particles being measured. The relationship between dH and the particle speed is defined in the Stokes-Einstein kT equation dH = 3D , where k is the Boltzmann’s constant, T is the absolute temperature, is the viscosity, and D is the translational diffusion coefficient. The result is a particle size distribution calculated from the intensity of scattered light. The size distribution is displayed as relative intensity of light scattered by particles versus particle diameter. The most frequently used number to define dH is called z-average (dH,avg ). It is the intensity-weighted mean of the diameters calculated from the intensity particle size distribution. A second parameter is the polydispersity index (Ipd ), which describes the width of the intensity distribution. We also deduced dH,avg from (i) the fitting of the main peak of the intensity distribution and (ii) the particle size corresponding to maximum intensity of scattered light. These results are omitted here because a similar scenario was found as deduced from z-average. Although the fundamental size distribution generated by PCS is an intensity distribution, this can be converted, using Mie theory, to a volume distribution. This volume distribution can also be further converted to a number distribution. Here, volume and number particle size distributions are not taken into consideration because generated from the primary size distribution, which is the intensity one. Volume and number distributions are typically only recommended for comparison between relative amounts of different particles. The comparison of dH,avg for CL wearers and non-wearers was obtained keeping the same experimental conditions: the temperature was set to 25 ◦ C, the selected stabilization time was 60 s, the viscosity and the refractive index of solvent were assumed to be those of water (n = 1.330, v = 0.8872 cP). The algorithm used by the software for calculating the size distribution was the “General Purpose – NNLS (Non-negative least squares)” [28]. 2.3. Electron spin resonance investigation The electron spin resonance (ESR) investigation was performed by a Bruker EMX spectrometer operating at the X-band frequency and equipped with an Oxford cryostat working in the temperature range of 4–298 K. The capillary containing the tears samples were charged in quartz glass tubes and the spectra were recorded at 298 K. The spectra of pure lactoferrin were recorded at 298 K on a solution of the protein in a phosphate buffer (0.02 M at pH = 7.6). Three samples of lactoferrin in phosphate buffer solution were analyzed and the results were substantially equivalent. The g values were calculated by standardization with 2,2-diphenyl1-picrylhydrazyl (DPPH). Care was taken to always keep the most sensitive part of the ESR cavity (1 cm length) filled.
28
S. Picarazzi et al. / Colloids and Surfaces B: Biointerfaces 157 (2017) 26–30 Table 2 p-values of Student’s t statistics (t-test) to assess the statistical difference between mean values of dH,avg of paired (P) or unpaired (U) groups.
Non-wearers CL wearers (CL in-situ)
CL wearers (CL in-situ)
CL wearers (after 25 min from CL removal)
<0.0001 (U)
0.0009 (U) 0.7089 (P)
Table 3 p-values of Student’s t statistics (t-test) to assess the statistical difference between mean values of Ipd data of paired (P) or unpaired (U) groups.
Non-wearers CL wearers (CL in-situ)
CL wearers (CL in-situ)
CL wearers (after 25 min from CL removal)
0.5135 (U)
0.5972 (U) 0.9923 (P)
distributions for non-wearers, CL wearers with the CL in-situ, and the same CL wearers after (25 ± 5) min from CL removal. Mean and standard deviation deduced from these data are reported in Table 1. Statistical significance of difference among groups was assessed by Student’s t statistic (t-test). The p-values are reported in Tables 2 and 3. 3.2. Electron spin resonance investigation ESR investigation has been initially performed on pure lactoferrin in phosphate buffer solution at 298 K, as a reference sample for studying the presence of iron centers in human tears (Fig. 2a). This protein is highly abundant in tears and it has high affinity in binding Fe3+ centers, conferring to the system relevant bacteriostatic and antioxidant properties [29–31]. After this preliminary investigation on lactoferrin, ESR spectra were carried out directly on human tears of one CL wearer and one non-wearer (Fig. 2b). 4. Discussion
Fig. 1. Statistical frequency distributions of dH,avg and Ipd data for non-wearers, for CL wearers with the CL in-situ, and for the same CL wearers after (25 ± 5) min from CL removal.
3. Results 3.1. Photon correlation spectroscopy analyses Fig. 1 shows the statistical frequency distributions for CL wearers and non-wearers of dH,avg and Ipd . Each figure shows the three Table 1 Mean and standard deviation (SD) of dH,avg and Ipd for non-wearers, for CL wearers with the CL in-situ, and for the same CL wearers after (25 ± 5) min from CL removal.
Non-wearers CL wearers (CL in-situ) CL wearers (after 25 min from CL removal)
dH,avg (nm)
Ipd
mean
SD
mean
SD
258.9 316.9 312.4
30.7 55.9 66.9
0.220 0.209 0.209
0.072 0.062 0.086
As far as PCS results are concerned, the first point to discuss is the significance of dH,avg . As already mentioned, the standard report for size measurements is an intensity distribution, while volume and number distributions are generated using established theories of light scattering. When the sample is monomodal (i.e. only one peak is detected) and monodisperse (i.e. the width of the distribution is negligible) and dispersed particles are spherical, the conversion from intensity to volume and number is less critical. Such samples typically have polydispersity Ipd below 0.1. On the contrary, the technique is not applicable in case of polydispersity higher than 0.5. For intermediate polydispersity, dH,avg can be deduced only from the intensity distribution and the obtained results can only be used for comparative purposes between different samples. Polydispersity of tear samples was found to be about 0.2 (Fig. 1b and Table 1). Therefore, for tear samples, dH,avg deduced from the intensity distribution can be used for comparative studies, as it is often the case for emulsions, polymers, proteins, and other samples [28]. Both for CL wearers and for non-wearers, dH,avg was found to be few hundreds of nanometers, but the above considerations clarify that this diameter is not attributable to a specific particle size of tear content. Moreover, in general, the hydrodynamic diameter depends, not only on the size of a particle core, but also on any surface structure, as well as ions in the medium. Indeed, ions in a medium are expected to influence the Brownian motion of suspended particles, namely PCS results. This aspect is expected to influence PCS results on tears, due to their complex composition. Another aspect which may affect PCS results is restricted diffusion due to high concentration. Restricted diffusion is due to the presence of other particles which hinders free particle Brownian motion. This can
S. Picarazzi et al. / Colloids and Surfaces B: Biointerfaces 157 (2017) 26–30
Fig. 2. ESR spectra of (a) pure lactoferrin in phosphate buffer solution and (b) tears of non wearers (i, black line) and wearers (ii, red line) of CLs (senofilcon A). A rectangle in (b) highlights the spectral region where a change of ESR signal for CL wearer is detectable.
be excluded for the tear samples under investigation with concentration (10 ± 2) V /W because no substantial differences were found varying the concentration from (3 ± 1) V /W to (10 ± 2) V /W . A clear statistical difference (p< 0.0001) was found between CL wearers with CL in-situ and non-wearers, dH,avg being ∼60 nm higher for CL wearers. A similar scenario was detected after 25 min from CL removal with a similar mean difference of about 60 nm compared to non-wearers. Although, as discussed above, the dH,avg measured number is not real, the difference has a physical meaning, which is attributed to a change of the state of aggregation/interaction of tear content due to CL wear. Indeed, PCS is very sensitive to the presence of aggregates and to interactions between suspended particles, which modify their free motion. Interestingly, the scenario is similar when the CL is in-situ and after 25 min from its removal. When a CL is placed on the eye, it strongly alters the arrangement of the tear film. Indeed, the tear film is a layer of few micrometers covering conjunctival and corneal epithelia, while CLs have a thickness of tens or hundreds of micrometers. Our results are not attributable to the presence of a CL placed on the eye during tear collection, but to an intrinsic change of the tear film proper-
29
ties observed also after its removal. These PCS results confirm the hypothesis of intrinsic changes of the tear film reported by Guillon et al. [22], who measured an higher tear evaporation rate for CL wearers compared to non-wearers, even after 1 day without CL. ESR preliminary results confirmed these interpretations. In the ESR spectrum of lactoferrin (Fig. 2a), an intense signal at g = 4.3 is easily detectable. As reported by Ainscough et al. [32] the position and the lineshape allows to attribute this resonance to Fe3+ centers with orthorhombic symmetry [32,33]. The weakest signals at g ∼ 6 and g ∼ 2 are instead associated to iron (III) species with axial symmetry [34]. These spectral features can be connected to the three dimensional structure of protein, which presents two homologous lobes binding a single Fe3+ ion. Similarly to lactoferrin, tears (Fig. 2b) show the signal at g = 4.3 of orthorhombic Fe3+ centers and the resonances of iron species with axial symmetry (g ∼ 6 and g ∼ 2). However, the intensity and the intensity, broadness and asymmetry of these features are much higher, suggesting that iron ions are more magnetically diluted and less coupled than in pure lactoferrin. Moreover, regardless the CLs wear, the spectrum of tears shows also the presence of an isotropic weak signal at g = 1.92, which, instead, is not detectable in pure lactoferrin. This signal is usually observed in Fe-S proteins and it has been associated to Fe-S clusters [34]. In the present case, we may attribute this signal to Fe-S clusters generated by the interaction of the Fe3+ centers with the S–S bridges of lysozyme, particularly abundant in the tear film [35]. However, further investigations are ongoing to clarify the origin of this feature. By comparing the spectra of tears of CL wearers and non-wearers, the signal intensity for the CL wearer appears decreased, in particular as concerns the resonances attributed to magnetically diluted Fe3+ centers. More in detail, the signal at g = 4.3 widens, almost overlapping the resonance at g ∼ 6 related to the species with axial symmetry. This overlapping can be detected by looking at the slope of the curves (their first derivative) in the range 1000–1100 G in Fig. 2b. Only the black curve (i, non-wearer) shows a minimum at ∼1000 G, while the red curve shows positive slope in the whole spectral range. These variations in intensity and width of the ESR spectral lines are usually attributable to a change of the surroundings of the paramagnetic centers and/or to their interaction (magnetic coupling) with neighboring radical species. In this case, we may suggest the occurrence of a change in the coordination or in the local environment of Fe3+ centers connected to aggregation phenomena of the proteins in the lacrimal film induced by the wear of CLs. This aggregation may promote magnetic coupling among the paramagnetic centers and, in turn, a consequent decay in intensity and line-broadening of the ESR signals. These preliminary results seem to support the differences observed by PCS between wearers and non-wearers and suggest possible modifications of the aggregation state of the tear components induced by CL wear.
5. Conclusions PCS detects intensity fluctuations of scattered light by a sample. It is applicable to analyze human tears for comparative purposes. It measures the polydispersity index and the mean hydrodynamic diameter of tear content. This diameter is not attributable to the size of a specific component, but it provides information on the presence of aggregates and interactions between tear constituents, which modify their free motion. The sensitivity of the technique makes it possible to discriminate between CL wearers and nonwearers (p < 0.0001). The hydrodynamic diameter was measured to be (316.9 ± 55.9) nm with the CL in-situ and (312.4 ± 66.9) nm after (25 ± 5) min from their removal, to be compared with (258.9 ± 30.7) nm for non-wearers. For future developments, PCS could be extensively applied to compare the clinical effects of different types of CLs and to develop more compatible CL materials, CL maintenance solu-
30
S. Picarazzi et al. / Colloids and Surfaces B: Biointerfaces 157 (2017) 26–30
tions, and artificial tears. The correlation with ocular pathologies could also be explored for diagnostic purposes. Preliminary ESR results support the differences observed by PCS between CL wearers and non-wearers. In fact, ESR signals with weaker intensity and an increased broadness were observed for tears of CL wearers compared to those of non-wearers. This behavior may be associated to changes in the coordination or in the local environment of Fe3+ centers, probably connected to aggregation phenomena of the tear proteins induced by CL wear. Acknowledgment The authors gratefully acknowledge prof. Alessandro Borghesi (University of Milano Bicocca) for helpful discussion and for critically reading the paper. References [1] S. Mishima, A. Gasset, S.D. Klyce, J.L. Baum, Determination of tear volume and tear flow, Invest. Ophthalmol. Visual Sci. 5 (1996) 264–276. [2] L. Battat, A. Macri, D. Dursun, S.C. Pflugfelder, Effects of laser in situ keratomileusis on tear production, clearance, and the ocular surface, Ophalmology 108 (2001) 1230–1235. [3] P.E. King-Smith, B.A. Fink, N. Fogt, K.K. Nichols, R.M. Hill, G.S. Wilson, The thickness of the human precorneal tear film: evidence from reflection spectra, Invest. Ophthalmol. Visual Sci. 41 (2000) 3348–3359. [4] S. Mishima, Some physiological aspects of the precorneal tear film, Arch. Ophthalmol. (1965) 233–241. [5] S. Patel, K.E. Boyd, J. Burns, Age, stability of the precorneal tear film and the refractive index of tears, Cont. Lens Ant. Eye 23 (2000) 44–47. [6] M. Yamada, M. Kawai, H. Mochizuki, Y. Hata, Y. Mashima, Fluorophotometric measurement of the buffering action of human tears in vivo, Curr. Eye Res. 17 (1997) 482–486. [7] J.E. Terry, R.M. Hill, Human tear osmotic pressure diurnal variations and the closed eye, Arch. Ophthalmol. 96 (1978) 120–122. [8] V. Ng, P. Cho, S. Mak, A. Lee, Variability of tear protein levels in normal young adults: between-day variation, Graefes Arch. Clin. Exp. Ophthalmol. 238 (2000) 892–899. [9] R.A. Sack, K.O. Tan, A. Tan, Diurnal tear cycle: evidence for a nocturnal inflammatory constitutive tear fluid, Invest. Ophthalmol. Visual Sci. 33 (1992) 626–640. [10] V. Ng, P. Cho, The relationship between total tear protein concentrations determined by different methods and standards, Graefes Arch. Clin. Exp. Ophtalmol. 238 (2000) 571–576. [11] A. Posa, L. Bräuer, M. Schicht, F. Garreis, S. Beileke, F. Paulsen, Schirmer strip vs. capillary tube method: non-invasive methods of obtaining proteins from tear fluid, Ann. Anat. 195 (2013) 137–142. [12] A.M. Bright, B.J. Tighe, The composition and interfacial properties of tears: tear substitutes and tear models, J. Br. Contact Lens Assoc. 16 (1993) 57–66. [13] J. Baguet, F. Somer, V. Claudon-Eyl, T.M. Duc, Characterization of lachrymal component accumulation on worn soft contact lens surface by atomic force microscopy, Biomaterials 16 (1995) 3–9.
[14] J.H. Teichroeb, J.A. Forrest, V. Ngai, J.W. Martin, L. Jones, J. Medley, Imaging protein deposits on contact lens materials, Optom. Vis. Sci. 85 (2008) 1151–1164. [15] M. Lira, L. Santos, E. Yebra-Pimentel, M. Elisabete, C.D. Real Oliveira, Comparative study of silicone-hydrogel contact lenses surfaces before and after wear using atomic force microscopy, J. Biomed. Mater. Res. B: Appl. Biomater. 85 (2008) 361–367. [16] G. Young, R. Garofalo, O. Harmer, S. Peters, The effect of soft contact lens care products on lens modulus, Contact Lens Anterior Eye 33 (2010) 210–214. [17] M. Berry, C. Purslow, P.J. Murphy, H. Pult, Contact lens materials, mucin fragmentation and relation to symptoms, Cornea 31 (2012) 770–776. [18] M. Bettuelli, S. Trabattoni, M. Fagnola, S. Tavazzi, L. Introzzi, S. Farris, Surface properties and wear performances of siloxane-hydrogel contact lenses, J. Biomed. Mat. Res. B: Appl. Biomater. 101B (2013) 1585–1593. [19] M. Lira, S. Franco, J.B. Vazquez-Dorrio, José Benito, M.E. Real Oliveira, M.F. Costa, Surface roughness and refractive index changes in contact lens induced by lens care systems, Eye Contact Lens 40 (2014) 140–147. [20] S. Tavazzi, M. Tonveronachi, M. Fagnola, F. Cozza, L. Ferraro, A. Borghesi, M. Ascagni, S. Farris, Wear effects on microscopic morphology and hyaluronan uptake in siloxane-hydrogel contact lenses, J. Biomed. Mater. Res. B: Appl. Biomater. 103 (2015) 1092. [21] Mostafa M. Bahgat, Precorneal tear film changes due to soft contact lens wear, Indian J. Ophthalmol. 33 (1985) 177–179. [22] M. Guillon, C. Maissa, Contact lens wear affects tear film evaporation, Eye Contact Lens 34 (2008) 326–330. [23] C. Purslow, J. Wolffsohn, J. Rubido, The effect of contact lens wear on dynamic ocular surface temperature, Contact Lens Anterior Eye 28 (2005) 29–36. [24] D. Martin, I. Fatt, The presence of a contact lens induces a very small increase in the corneal surface temperature, Acta Ophthalmol. 64 (1986) 512–518. [25] M. Reim, N.F. Schrage, J. Becker, Interactions between ocular surface fluid and cornea related to contact lenses, Eur. J. Ophthalmol. 11 (2001) 105–115. [26] A. Mann, B. Tighe, Contact lens interactions with the tear film, Exp. Eye Res. 117 (2013) 88–92. [27] M. Azharuddin, J. Khandelwal, H. Datta, A.K. Dasgupta, S.O. Raja, Dry eye a protein conformational disease, Invest. Ophthalmol. Vis. Sci. 56 (2015) 1423–1429. [28] User manual zetasizer nanoseries, Malvern (2013). [29] J.V. Bannister, W.H. Bannister, H.A.O. Hill, P.J. Thornalley, Enhanced production of hydroxyl radicals by the xanthine- xanthine oxidase reaction in the presence of lactoferrin, Biochim. Biophys. Acta 715 (1982) 116–120. [30] H. Shinmoto, S. Dosako, I. Nakajima, Antioxidant activity of bovine lactoferrin on iron/ascorbate induced lipid peroxidation, Biosci. Biotechnol. Biochem. 56 (1992) 2079–2080. [31] V. Pastori, S. Tavazzi, M. Lecchi, Lactoferrin-loaded contact lenses: eye protection against oxidative stress, Cornea 34 (2015) 693–697. [32] E.W. Ainscough, A.M. Brodie, J.E. Plowman, S.J. Bloor, J.S. Loehr, T.M. Loehr, Studies on human lactoferrin by electron paramagnetic resonance, fluorescence, and resonance Raman spectroscopy, Biochemistry 19 (1980) 4072–4079. [33] G.G. Borisenko, V.E. Kagan, C.J. Hsia, N.F. Schor, Interaction between 6-hydroxydopamine and transferrin: let my iron go, Biochemistry 39 (2000) 3392–3400. [34] R. Cammack, C. Cooper, Electron paramagnetic resonance spectroscopy of iron complexes and iron-containing proteins, Methods Enzymol. 227 (1993) 353–384. [35] S. Patel, L. Anderson, K. Cairney, The refractive index of tears in normals and soft lens wearers, Adv. Exp. Med. Biol. 350 (1994) 417–420.