Fluorescence Spectroscopy Study of Protoporphyrin IX in Optical Tissue Simulating Liquid Phantoms

materials

Article Fluorescence Spectroscopy Study of Protoporphyrin IX in Optical Tissue Simulating Liquid Phantoms

Huihui Lu 1,*, Francesco Floris 2 , Marc Rensing 2 and Stefan Andersson-Engels 1,3

1 Biophotonics @ Tyndall, IPIC, Tyndall National Institute, University College Cork, T12 R5CP Cork, Ireland; [email protected] 2 Photonics Packaging Group, IPIC, Tyndall National Institute, University College Cork, T12 R5CP Cork, Ireland; francesco.ﬂ[email protected] (F.F.); [email protected] (M.R.) 3 Department of Physics, University College Cork, T12 K8AF Cork, Ireland * Correspondence: [email protected]

Received: 27 March 2020; Accepted: 28 April 2020; Published: 2 May 2020

Abstract: Fluorescence spectroscopy has been extensively investigated for disease diagnosis. In this framework, optical tissue phantoms are widely used for validating the biomedical device system in a laboratory environment outside of clinical procedures. Moreover, it is fundamental to consider that there are several scattering components and chromophores inside biological tissues and the interplay between scattering and absorption may result in a distortion of the emitted fluorescent signal. In this work, the photophysical behaviour of a set of liquid, tissue-like phantoms containing different compositions was analysed: phosphate buffer saline (PBS) was used as the background medium, low fat milk as a scatterer, Indian ink as an absorber and protoporphyrin IX (PpIX) dissolved in dimethyl formamide (DMF) as a fluorophore. We examined the collected data in terms of the impact of surfactant Tween-20 on the background medium, scattering effects and combination of scattering and absorption within a luminescent body on PpIX. The results indicated that the intrinsic emission peaks are red shifted by the scattering particles or surfactant, whilst the scattering agent and the absorbent can alter the emission intensity substantially. We corroborated that phantoms containing higher surfactant content (>0.5% Tween 20) are essential to prepare stable aqueous phantoms.

Keywords: protoporphyrin IX; optical phantoms; ﬂuorescence spectroscopy; optical properties; tissue diagnostics

1. Introduction Fluorescence spectroscopy is seen as a promising diagnostic tool thanks to its inherent rapid response, combined with high sensitivity and specificity rate. This makes it an ideal method for disease diagnosis in the fields of biological and biomedical applications, specifically for tissue diseases. When tissues are illuminated with ultra-violet or visible light, biological molecules can absorb the energy carried by the impinging light and re-emit it at a longer wavelength. This can be used to capture the features of endogenous or exogenous fluorophores in the form of an injectable fluorescent molecule to differentiate tumour and normal tissues in a variety of organ systems [1–3]. Most common endogenous tissue fluorophores include tryptophan, tyrosine, phenylalanine, collagen, elastin, reduced nicotinamide adenine dinucleotide (NADH), flavin adenine dinucleotide (FAD) and porphyrins [4]. Each of these molecules has unique excitation and emission spectra. There are also various scattering components and chromophores in biological tissues. Photons incident on a tissue undergo substantial scattering and absorption when emerging from the surface. The interplay of absorption and scattering can result in distortion of the fluorescent signal from tissues [5,6]. Disentangling effects of scattering and absorption have already been investigated. Müller et al. [7] developed a model based on photon-migration theory

Materials 2020, 13, 2105; doi:10.3390/ma13092105 www.mdpi.com/journal/materials Materials 2020, 13, 2105 2 of 10 and information from simultaneously acquired fluorescence and reflectance spectra, in order to extract the intrinsic fluorescence from turbid media. Wu et al. [8] used a non-negative matrix factorization algorithm to generate a fluorescence excitation emission matrix that corresponded to the fluorophores in biological tissue, including tryptophan, collagen, elastin, NADH and FAD. Protoporphyrin IX (PpIX), a heterocyclic organic compound consisting of four pyrrole rings, is the final product in the heme biosynthetic pathway [9]. In the field of neurosurgery, administration of an exogenous fluorophores or a fluorophores precursor, such as 5-aminolevulonic acid (5-ALA), is used to help guide surgery [1,4,10]. 5-ALA is taken up by glioma cells where a breakdown of the blood-brain barrier (BBB) has occurred, but not in normal brain tissue. This permits fluorescence discrimination of the tumour and normal tissue. Researchers in this field have developed a variety of approaches to quantify PpIX fluorescence emission from tissue [11,12]. Optical tissue phantoms are widely used for validating prototype medical device systems in laboratory environments outside of clinical procedures. In this framework, our goal is to offer an additional procedure for optical fluorescent tissue phantoms in order to validate an integrated optical probe system for tissue characterization, based upon diffuse reflectance and fluorescence. With this in mind, the aim of this paper is to find and study a set of liquid, tissue-like phantoms with different compositions. We used PBS as the background medium, low fat milk as a scatterer, Indian ink as an absorber and PpIX as a fluorophore. PpIX can be dissolved in many organic solvents (e.g., dimethyl sulfoxide (DMSO), DMF, etc.), but it is very difficult to dissolve in water, and we are interested in water-based background media. Surfactants such as Tween are commonly added to aqueous PpIX phantoms to prevent aggregation [11–14]. They have an effect on the fluorescence spectra [13,15–17]; therefore, the role of Tween on background medium was also investigated.

2. Materials and Methods

2.1. Instrumentation PpIX has a fluorescence emission property due to π-electron delocalization over the conjugated double-bond system. PpIX can be excited in different spectral regions, each offering advantages and disadvantages [13,18]. Commonly, blue light is used to excite the PpIX in fluorescence-guided surgery [12,19]. In this spectral region, blue light generates efficient fluorescence, however, the light is highly absorbed by haemoglobin. Excitement in the red wavelength yields a less efficient fluorescent signal, but avoids light attenuation by the blood and achieves deeper sampling. Using red light excitation has been applied for PpIX concentration monitoring during photodynamic therapy [20]. In this study, light at a wavelength of 415 nm was chosen for excitation. During the measurements, excitation light from a Fiber-Coupled LED (Thorlabs M415F3, Output power ~21.3 mW, FWHM 14 nm, Spot size ~1.3 mm) was projected on the samples and fluorescent emissions were collected using a QEPro spectrometer from Ocean Optics (Duiven, Netherlands), with a detection range from 400 to 1000 nm. Measurements were taken at an angle perpendicular to the excitation light [21]. Liquid phantom samples were placed in transparent plastics cuvettes with a 10 mm path length. To minimise inner-filter effects from the background medium, the background media without dye were recorded prior to each fluorescence measurement and subtracted from the raw data.

2.2. Liquid Phantom Samples Preparation Tween-20, PBS, PpIX powder, and DMF were purchased from Merck (Wicklow, Ireland). Milk (Dawn® low fat milk, produced in Kerry, Ireland) and Indian ink (Winsor and Newton® black India ink, Cork, Ireland) were used as scatterer and absorber. PpIX powder was dissolved in DMF to prepare PpIX stock solution as 10 mg/mL. Dilutions from this stock into PBS solution were used to achieve the desired PpIX concentration. A series of ﬂuorescence emission spectra of PpIX were measured for diﬀerent concentrations of PpIX. Materials 2020, 13, 2105 3 of 10

In order to investigate the role of surfactant in background medium, 100 µL of PpIX (10 mg/mL) stock solution were diluted in 9.9 ml PBS solution containing different amount of Tween 20 (10%, 5%, 2.5%, 1.25%, 0.5%, 0.05%, 0.005%, 0.0005%, and 0%). The selection of the optimal scatterers and absorbers in tissue phantoms strongly depended on specific applications [14,22–24]. In this paper, milk and Indian ink were introduced into PpIX solutions to investigate the influence of the scattering and the absorption on PpIX fluorescence emission spectra. One set contained different concentrations of milk without the background absorber. Another two sets contained both milk and Indian ink as a background absorber, and milk was used at two 1 concentrations to simulate low and high scattering coefficients (130 and 260 cm− ). Indian ink was selected at five concentrations to simulate different absorption level coefficients (0.01, 0.29, 0.56, 1 1.11 and 2.21 cm− ). The properties of the scattering coefficient of milk and the absorption coefficient of Indian ink were determined by combination of the diffuse reflectance and transmittance spectroscopy measurement setup described in our previous work, [25,26] and were chosen within a range typical for biological tissue.

3. Results and Discussion

3.1. Effect of Surfactant on PpIX Fluorescence in Non-Turbid Phantom The influence of Tween 20 on fluorescence spectra was investigated by diluting the same volume of PpIX stock solution in PBS buffer containing different volume fractions of Tween 20 (0.5%, 0.05%, 0.005%, 0.0005%, and 0%). Emission spectra are shown in Figure1. It is observed that PpIX emission spectrum had two distinctive peaks at wavelengths of 622 nm and 684 nm, with the sample containing Materials 2020, 13no, x surfactants. FOR PEER EmissionREVIEW peaks at 630 nm from PpIX solutions containing surfactant were red shifted 4 of 11 (~10 nm). In addition, the intensity of the entire band significantly expanded.

x105 2.0 Shift Tween 20 Volume fraction 1.8 Peak1 0 (%) 1.6 0.0005 (%) 0.005 (%) 1.4 0.05 (%) 1.2 0.5 (%) 1.0 0.8 0.6 Peak2

Emisson intensity (a.u.) intensity Emisson 0.4 No Tween 0.2 0.0 600 620 640 660 680 700 720 740 Wavelength (nm)

Figure 1. Fluorescence spectra (415 nm excitation) of PpIX in PBS buffer with different Tween 20 Figure 1. Fluorescencevolume fraction. spectra (415 nm excitation) of PpIX in PBS buffer with different Tween 20 volume fraction. To further investigate the effect of Tween 20 on PpIX fluorescence, PpIX solution in PBS buffer containing higher volume fractions of Tween 20 (10%, 5%, 2.5%, 1.25%, 0.5%, 0.05%, 0.005%, 0.0005%, Tween / PpIX Molar ratio and 0%) were prepared, and the luminescence study was performed with these solutions. The emission 0 10-2 10-1 100 101 102 103 peaks at 630 nm were106 plotted as a function of Tween 20 volume fraction and Tween/PpIX molar ratio in Figure2. The emission intensities were observed to increase proportionally with the Tween volume fraction between 0.0005% and 1.25%. The PpIX solution containing Tween showed a strongly enhanced emission signal, with an approximate 49-fold increase at a Tween volume fraction of 0.5%, 105

104

Emission Intensity Emission

Peak1 @630 nm (a.u.)Peak1

No Tween

103 0 10-4 10-3 10-2 10-1 100 101 Tween volume fraction (v/v%)

Figure 2. PpIX fluorescence peak intensity at 630 nm as a function of Tween 20 volume fraction (bottom x-axis) and Tween/PpIX molar ratio (top x-axis).

Further studies were carried out by measuring a series of fluorescence emission spectra for different concentrations of PpIX in PBS solutions containing 0.5% (v/v) Tween 20. Emission peak intensity as a function of PpIX concentration is plotted in Figure 3. It reveals a maximum intensity around 125 µg/mL (PpIX/Tween molar ratio is 0.046) and shows that the emission signal decreased when going to higher PpIX concentration. PpIX has a fluorescence emission property due to π- electron delocalization over the conjugated double-bond system. Porphyrin presents high aggregates in a pH ranging from 3 to 7 [15,16], and aggregation through the occurrence of π-π stacking interactions. Figure 2 shows the molar ratio of Tween/PpIX to reach equilibrium was around 27 (at this point Tween 20 volume fraction is 0.5%, PpIX concentration is 100 µg/mL). Fluorescence emission intensity was expected to decrease due to the PpIX samples beginning to aggregate at higher concentrations. For further quantitative studies, PpIX concentration below 125 µg/mL was selected because a monotonic relationship between signal and dye concentration is preferred.

Materials 2020, 13, x FOR PEER REVIEW 4 of 11

x105 2.0 Shift Tween 20 Volume fraction 1.8 Peak1 0 (%) Materials 2020, 13, 2105 1.6 0.0005 (%) 4 of 10 0.005 (%) 1.4 0.05 (%) compared with that containing1.2 no surfactant. The emission intensities increase 0.5 (%) by 91% and 98% of the maximum signal in the sample1.0 containing 0.5% and 1.25% Tween respectively, compared with that containing no surfactant. Our experimental results agree with previous studies from Marois et al. [13], 0.8 who showed a PpIX sample containing no surfactant with a reduction of 98% fluorescence emission, Peak2 compared with that containing0.6 surfactant. The intensity of molecular fluorescence is greatly influenced by the solvent effect [27, (a.u.) intensity Emisson 280.4] and other medium variables, such as pH [28,29]. The purpose of our No Tween phantom was to simulate0.2 biological tissue, therefore PBS buffer with a pH kept at 7 was used as a background medium throughout0.0 the study. The enhancement of the fluorescence emission may be due to (1) the non-ionic600 surfactant620 increasing640 the660 solubility680 of700 hydrophobic720 PpIX;740 (2) at higher concentrations, micelles start to form and PpIX structuredWavelength in micelles (nm) have different emission intensities. For Tween volume fraction in the range of 0.5% to 1.25%, the emission intensity exhibited a variation of only 7% between two concentrations. For Tween volume fraction in the range of 1.25% to 10%, Figureemission 1. Fluorescence intensity plateaued spectra and (415 slightly nm decreased.excitation) Tween of PpIX 20 volume in PBS fraction buffer (0.5% withv/v) wasdifferent therefore Tween 20 volumeemployed fraction for. further studies.

Tween / PpIX Molar ratio 0 10-2 10-1 100 101 102 103 106

105

104

Emission Intensity Emission

Peak1 @630 nm (a.u.)Peak1

No Tween

103 0 10-4 10-3 10-2 10-1 100 101 Tween volume fraction (v/v%)

Figure 2. PpIX fluorescence peak intensity at 630 nm as a function of Tween 20 volume fraction (bottom Figure 2. PpIXx-axis) and fluorescence Tween/PpIX molar peak ratio intensity (top x-axis). at 630 nm as a function of Tween 20 volume fraction (bottom x-axis) and Tween/PpIX molar ratio (top x-axis). Further studies were carried out by measuring a series of fluorescence emission spectra for different concentrations of PpIX in PBS solutions containing 0.5% (v/v) Tween 20. Emission peak intensity as a Furtherfunction studies of PpIX were concentration carried is out plotted by in measuring Figure3. It reveals a series a maximum of fluorescence intensity around 125 emissionµg/mL spectra for different concentrations(PpIX/Tween molar of ratio PpIX is 0.046) in and PBS shows solution that thes emissioncontaining signal 0.5%decreased (v/v when) Tween going to 20. higher Emission peak intensity asPpIX a funct concentration.ion of PpIX PpIX has concentration a fluorescence emission is plotted property in Figure due to π -electron3. It reveals delocalization a maximum over intensity the conjugated double-bond system. Porphyrin presents high aggregates in a pH ranging from 3 to around 1257 [µg15,16/mL], and (PpIX/Tween aggregation through molar the ratio occurrence is 0.046) of π- πandstacking shows interactions. that the Figure emission2 shows signal the decreased when goingmolar to ratio higher of Tween PpIX/PpIX concentration. to reach equilibrium PpIX was around has a 27 fluorescence (at this point Tween emission 20 volume property fraction due to π- electron delocalizationis 0.5%, PpIX concentration over the conjugated is 100 µg/mL). double Fluorescence-bond emission system intensity. Porphyrin was expected presents to decrease high aggregates due to the PpIX samples beginning to aggregate at higher concentrations. For further quantitative in a pH rangingstudies, PpIX from concentration 3 to 7 below[15,16 125], µandg/mL wasaggregation selected because through a monotonic the relationship occurrence between of π-π stacking interactions.signal Figure and dye 2 shows concentration the molar is preferred. ratio of Tween/PpIX to reach equilibrium was around 27 (at this point Tween 20 volume fraction is 0.5%, PpIX concentration is 100 µg/mL). Fluorescence emission intensity was expected to decrease due to the PpIX samples beginning to aggregate at higher concentrations. For further quantitative studies, PpIX concentration below 125 µg/mL was selected because a monotonic relationship between signal and dye concentration is preferred.

Materials 2020, 13, 2105 5 of 10 Materials 2020, 13, x FOR PEER REVIEW 5 of 11

PpIX / Tween Molar ratio 10-6 10-5 10-4 10-3 10-2 10-1 100 105 Peak1 @ 630nm Peak2 @ 700nm 104

103

102

101

Emission Peak Intensity (a.u.) Peak Emission

100 10-3 10-2 10-1 100 101 102 103 104 PpIX (g/ml)

Figure 3. Emission peak intensity as a function of concentrations of PpIX in PBS solutions containing Figure 3. Emission0.5% (v peak/v) Tween intensity 20 (bottom as x-axis) a function and PpIX of/Tween concentrations molar ratio (top of x-axis). PpIX in PBS solutions containing 0.5% (v/v) Tween 20 (bottom x-axis) and PpIX/Tween molar ratio (top x-axis). 3.2. Eﬀect of Scattering on PpIX Fluorescence

3.2. Effect of ScatteringTo investigate on PpIX the impactFluorescence of scattering on the intensity and the spectral shape of a PpIX fluorescence spectrum, low fat milk was used as the scatterer and was introduced into the aqueous phantom. The phantoms contained 100 µg/mL PpIX in PBS buffer containing 0.5% (v/v) Tween 20 with milk To investigate the impact of scattering on the intensity and the spectral shape1 of a PpIX between 0% and 80% (v/v), corresponding to the scattering coefficient range, µs, from 0 cm− to 1 fluorescence spectrum,264 cm− at alow wavelength fat milk of 415 was nm. used The measured as the scatter fluorescenceer and spectra was are introduced shown in Figure into4a,b, the aqueous phantom. Theshowing phantoms that the contain normaliseded spectra 100 µg shape/mL (normalised PpIX in to PBS their peakbuffer heights) contain were alling very 0.5% similar (v/v to ) Tween 20 the spectrum with no scatterer added, and had two distinctive peaks at wavelengths of 632 nm and 701 with milk between 0% and 80% (v/v), corresponding to the scattering coefficient range, µs, from 0 cm−1 nm. This observation showed that PpIX fluorescence was not red-shifted by the presence of the scatterer. to 264 cm−1 at Thisa wavelength seemed not to beof intuitive 415 nm. in terms The of measured previous work fluorescence from Ahmed et al. spectra [5], which are showed shown red shifts in Figure 4a-b, showing that ofthe the normalised emission spectra spectra peaks in shape the presence (normalised of a random scatterto their in luminescent peak heights) bodies, withwere respect all very similar to the spectrumto the with intrinsic no fluorescencescatterer added emission., and The three had main two components distinctive of milk, peaks besides at water,wavelengths are lipids, of 632 nm protein and lactose. The most abundant membrane lipids are the phospholipids [30]. Phospholipids and 701 nm. Thishave amphipathicobservation properties, showed and that therefore PpIX we hypothesisedfluorescence that was this component not red- inshifted the milk by may the act presence of the scatterer. Thisas a surfactant. seemed Similar not to experiments be intuitive were furtherin terms carried of out previous in the PBS work buffer containing from Ahmed no surfactant. et al. [5], which Figure4c,d show the red shifts in the emission spectra in the presence of the scatterer, with emission showed red shiftspeak wavelength of the emission at ~630 nm spectra and ~700 nm.peaks These in results the arepresence very similar of toa whatrandom we have scatter observed in in luminescent bodies, with respectthe previous to section the intrinsic (Section 3.1 fluoresand Figurecence1) on fluorescence emission. spectrum The three of PpIX main in PBS components bu ffer with of milk, besides waterdi, fferentare Tween lipids, 20 volume protein fraction. and This lactose. confirmed The that milk most in the abundant phantom may actmembrane as a surfactant lipids in are the the aqueous tissue-like phantom. phospholipids [30]PpIX. Phospholipids fluorescence peak have intensities amphipathic at ~630 nmand properties ~700 nm were, and plotted therefore as a function we hypothesi of the sed that this componentscattering in the coe milkfficient may as reported act as in a Figure surfactant.5 with left y-axis.Similar Relative experiments emission intensities were atfurther 630 nm ofcarried out in the PBS bufferthe con PpIXtaining sample no containing surfactant. different Figure amounts 4c of– scatterer,d show compared the red with shifts that containingin the emission no scatterer, spectra in the were plotted as a function of scattering coefficient, which are shown in Figure5 with right y-axis. presence of theIn scatter the case ofer the, with background emission medium peak as PBS wavelength buffer containing at ~630 Tween nm 20 (Figure and5 a),~700 the fluorescencenm. These results are very similar tointensity what decreased we have nearly observed exponentially, in with the increased previous scattering section coe ffi (cient.Section The 3.1 PpIX and sample Figure 1) on 1 1 fluorescence spectrumcontaining theof scattererPpIX in at PBS the scattering buffer with coefficient differentµs at 130 Tween cm− (µ 20s’ = volume13 cm− given fraction anisotropy. This confirmed factor g = 0.9) reduced 64% of fluorescence emission, compared with that containing no scatterer. In the that milk in thecase phantom of PBS buff ermay containing act as no a surfactant,surfactant Figure in5 theb shows aqueous the fluorescence tissue- intensitylike phantom. increased with

Materials 2020, 13, 2105 6 of 10

1 1 increased scattering coefficient at lower values of µs, from 0 cm− to 66 cm− . It is shown that the 1 PpIX sample containing the scatterer at scattering coefficient µs at 66 cm− had an increase of 400% 1 fluorescence emission compared with that containing no scatterer. For larger values of µs, from 66 cm− 1 to 264 cm− , the fluorescence intensity decreased with increased scattering coefficient after reaching the peak value. The normalised spectra in Figure4d indicate an increased absorption due to light scattering. Materials 2020, 13, x FOR PEER REVIEW 6 of 11 In this regime, light undergoes a longer random-walk path and experiences more absorption, hence the decreases in emission intensity.

x104 (a) 5.0 PBS with Tween 20 (b) PBS with Tween 20 1.0 -1 s cm 4.0 0 0.8 17 33 3.0 0.6 66 132 2.0 198 0.4 264

1.0 0.2

0.0 0.0 x103 5.0 (c) Shift PBS buffer (d) PBS buffer 1.0 4.0 0.8 Signal Normalised

Emission intensity (a.u.) intensity Emission 3.0 0.6

2.0 No scatterer 0.4

1.0 0.2

0.0 0.0 500 550 600 650 700 750 800 500 550 600 650 700 750 800 Wavelength (nm) Wavelength (nm)

Figure 4. Fluorescence spectra (415 nm excitation) of PpIX with diﬀerent concentrations of scattering

agent and their normalised spectra (normalised to each spectrum peak intensity); in PBS buﬀer Materials 2020, 13, x FOR PEER REVIEW 7 of 11 containingFigure 4. TweenFluorescence 20 (0.5% spectrav/v )((415a,b nm) and excitation) PBS bu ﬀofer PpIX containing with different no surfactant concentrations (c,d). of scattering agent and their normalised spectra (normalised to each spectrum peak intensity); in PBS buffer containing Tween 20 (0.5% v/v) (a,b) and PBS buffer containing no surfactant (c,d). x104 x103 5.0 1.0 5.0 8.0 (a) PBS with Tween 20 (b) PBS buffer PpIX fluorescence peak intensities Peak1 @630nm at ~630 nm and ~700 nm were plotted Peak1as a @630nm function7.0 of the 4.0 Peak 2 @700nm 0.8 4.0 Peak 2@700nm scattering coefficient as reported in Figure 5 with left y-axis. Relative emission intensities at6.0 630 nm 5.0 of the PpIX3.0 sample containing different amounts0.6 of scatter3.0 er, compared with that containing no scatterer, were plotted as a function of scattering coefficient, which are shown in Figure 5 with4.0 right 2.0 2.0 y-axis. In the case of the background medium0.4 as PBS buffer containing Tween 20 (Figure 3.05a), the fluorescence intensity decreased nearly exponentially, with increased scattering coefficient. The2.0 PpIX 1.0 0.2 1.0

Emission peak intensity (a.u.) peak intensity Emission (a.u.) peak intensity Emission −1 1.0−1 sample containing the scatterer at the scattering @ 630nm peak intensity Relative coefficient µ s at 130 cm (µ s’ = 13 cm given @ 630nm peak intensity Relative anisotropy 0.0factor g = 0.9) reduced 64% of fluorescence0.0 emission0.0 , compared with that containing0.0 no 0 50 100 150 200 250 300 0 50 100 150 200 250 300 scatterer. In the case of PBS -1 buffer containing no surfactant, Figure 5b shows-1 the fluorescence s(cm ) s(cm ) intensity increased with increased scattering coefficient at lower values of µ s, from 0 cm−1 to 66 cm−1 . It Figureis shown 5. thatPpIX the fluorescence PpIX sample peak containing intensity the at scatter ~630er nm at scattering (blue line) coefficient and ~700 µ nms at (red66 cm line)−1 had plotted an Figure 5. PpIX fluorescence peak intensity at ~630 nm (blue line) and ~700 nm (red line) plotted as a increasas a functione of 400% of fluorescence scattering coeemissionfficient compa (left y-axis).red with Relative that containing emission no intensities scatterer. For at 630larger nm values of PpIX function of scattering coefficient (left y-axis). Relative emission intensities at 630 nm of PpIX samples of samples µ s, from containing 66 cm−1 to di ff 264erent cm concentrations−1, the fluorescence of scatterer, intensity compared decrease withd thatwith containing increased no scattering scatterer as containing different concentrations of scatterer, compared with that containing no scatterer as a coefficienta function after of scatteringreaching the coe peakfficient value. (right The y-axis). normalised (a) in PBSspectra buff iner Figure containing 4d indicate Tween an 20 andincreased (b) PBS function of scattering coefficient (right y-axis). (a) in PBS buffer containing Tween 20 and (b) PBS absorption due to light scattering. In this regime, light undergoes a longer random-walk path and bufferbuffer containing containing no surfactant.no surfactant. experiences more absorption, hence the decreases in emission intensity. 3.3. Effect of Combined Scattering and Absorption on PpIX Fluorescence 3.3. Effect of Combined Scattering and Absorption on PpIX Fluorescence To investigateTo investigate absorption absorption effects effects on PpIX on PpIX fluorescence fluorescence in the in presence the presence of a scatterer, of a scatter weer conducted, we experimentsconducted using experiments another using two sets another ofphantoms two sets of containing phantoms containing milk and milkIndian and ink Indian as a inkbackground as a absorber.background Milk was absorb useder at. Milk two wasconcentrations used at two to concentrations simulate low toand simulate high scattering low and coehighffi scatteringcient (130 and coefficient (130 and 260 cm−1). Indian ink was selected at five concentrations to simulate different absorption coefficient levels (0.01, 0.29, 0.56, 1.11 and 2.21 cm−1). Figure 6 shows an overview via a 3D diagram of the fluorescence spectra of PpIX in PBS solutions, as a function of wavelength and absorption coefficients at the level of µ s equal to 130 cm−1 and 260 cm−1. To illustrate more clearly the absorption effects, PpIX emission peak intensity at ~630 nm and ~700 nm, plotted as a function of absorption coefficient, are also displayed in Figure 6c-d with the left y-axis. Relative emission intensities at 630 nm of the PpIX sample containing different concentrations of absorber, compared with that containing no absorber, were plotted as a function of absorption coefficient, which are shown in Figure 6c,d with the right y-axis. With increased absorption coefficient, the fluorescence intensity decreased. PpIX samples containing the absorber at the absorption coefficient µ a at 1.1 cm−1 and 2.2 cm−1, showed reductions by 88% and 97% fluorescence emission, respectively, compared with that containing no absorber. Marois et al. [13] used whole blood as a background absorber and identified that there is an interaction between erythrocytes in whole blood and Tween, which can complicate the stability of PpIX fluorescence. In our experiments, we did not observe any decrease in the fluorescence signal over eight hours in freshly prepared samples in PBS solution containing Tween 20. Using whole blood as an absorber in phantoms can provide realistic tissue spectra and oxygenation function, while ink provided nearly flat absorption spectra. The selection of the optimal absorber in tissue phantoms strongly depends on specific applications [14].

Materials 2020, 13, 2105 7 of 10

1 260 cm− ). Indian ink was selected at five concentrations to simulate different absorption coefficient 1 levels (0.01, 0.29, 0.56, 1.11 and 2.21 cm− ). Figure6 shows an overview via a 3D diagram of the fluorescence spectra of PpIX in PBS solutions, 1 as a function of wavelength and absorption coefficients at the level of µs equal to 130 cm− and 1 260 cm− . To illustrate more clearly the absorption effects, PpIX emission peak intensity at ~630 nm and ~700 nm, plotted as a function of absorption coefficient, are also displayed in Figure6c,d with the left y-axis. Relative emission intensities at 630 nm of the PpIX sample containing different concentrations of absorber, compared with that containing no absorber, were plotted as a function of absorption coefficient, which are shown in Figure6c,d with the right y-axis. With increased absorption coe fficient, the fluorescence intensity decreased. PpIX samples containing the absorber at the absorption coefficient 1 1 µa at 1.1 cm− and 2.2 cm− , showed reductions by 88% and 97% fluorescence emission, respectively, compared with that containing no absorber. Marois et al. [13] used whole blood as a background absorber and identified that there is an interaction between erythrocytes in whole blood and Tween, which can complicate the stability of PpIX fluorescence. In our experiments, we did not observe any decrease in the fluorescence signal over eight hours in freshly prepared samples in PBS solution containing Tween 20. Using whole blood as an absorber in phantoms can provide realistic tissue spectra and oxygenation function, while ink provided nearly flat absorption spectra. The selection of the optimalMaterials 20 absorber20, 13, x FOR in PEER tissue REVIEW phantoms strongly depends on specific applications [14]. 8 of 11

(a) (b)

x103 x103 7.0 7.0 -1 1.0  260 cm-1 1.8 s 130 cm s 6.0 Peak1 @630nm 6.0 Peak1 @630nm 1.6 Peak2 @700nm Peak2 @700nm 0.8 1.4 5.0 5.0 1.2 4.0 0.6 4.0 1.0 3.0 3.0 0.8 0.4 0.6 2.0 2.0 0.2 0.4

Emission peak intensity (a.u.) peak intensity Emission 1.0 (a.u.) peak intensity Emission 1.0

Relative peak intensity @ 630nm peak intensity Relative 0.2

Relative peak intensty @ peak intensty 630 nm Relative 0.0 0.0 0.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5 -1 -1 a (cm ) a (cm )

FigureFigure 6. Fluorescence 6. Fluorescence spectra spectra (415 (415 nm nm excitation) excitation) of PpIX in in PBS PBS solution solution as asa function a function of wavelength of wavelength at lowerat lower (a) and (a) and higher higher (b) ( scatteringb) scattering coe coefficients.fficients. Emission intensities intensities as asa function a function of absorption of absorption coefficoefficientcient at lower at lower (c) and(c) and higher higher (d ()d scattering) scattering coecoefficientsfficients.. Relative Relative emission emission intensities intensities at 63 at0 n 630m nm of PpIX sample containing different concentrations of absorber compared with that containing no of PpIX sample containing different concentrations of absorber compared with that containing no absorber is shown in right y-axis. absorber is shown in right y-axis. 3.4. Temporal Stability Study of PpIX Fluorescence We investigated phantom stability in terms of aggregation, by comparing PpIX prepared in PBS buffer containing different concentration of Tween 20 (0.5%, 0.05%, 0.005%, 0.0005%, and 0%) stored at room temperature (20 °C). It is clearly shown in Figure 7 that there was no precipitation observed in PBS containing 0.5% (v/v) Tween 20 at day seven after samples were freshly prepared. To further characterise the temporal stability of PpIX fluorescence signal in PBS containing surfactant, the sample was measured when it was freshly prepared and seven days after preparation. The measured fluorescence spectra of PpIX prepared in PBS containing two different concentrations of Tween 20 (0.5%, 0.05%) are shown in Figure 8. In PBS containing 0.5% (v/v) Tween 20 medium, the spectra of PpIX aqueous sample measured at day seven after preparation was nearly identical to that of the freshly prepared sample. Its spectra showed good stability in terms of the fluorescence signal intensity, while in PBS containing a lower concentration of Tween 20 (0.05% v/v), the fluorescence signal decreased at day seven after preparation.

Materials 2020, 13, 2105 8 of 10

3.4. Temporal Stability Study of PpIX Fluorescence We investigated phantom stability in terms of aggregation, by comparing PpIX prepared in PBS buﬀer containing diﬀerent concentration of Tween 20 (0.5%, 0.05%, 0.005%, 0.0005%, and 0%) stored at room temperature (20 ◦C). It is clearly shown in Figure7 that there was no precipitation observed in Materials 2020,PBS 13, x containing FOR PEER 0.5% REVIEW (v/v) Tween 20 at day seven after samples were freshly prepared. 9 of 11

PBS contains Tween 20 PBS 0.5% 0.05% 0.005% 0.0005% 0%

Materials 2020, 13, x FOR PEER REVIEW 9 of 11 Figure 7. PpIX aqueous samples prepared in PBS containing different volume fractions of Tween 20 at Figure 7. PpIXday sevenaqueous after preparation.samples preparedPBS containsin PBS Tween contain 20 ing differentPBS volume fractions of Tween 20 at day seven after preparation. 0.5% 0.05% 0.005% 0.0005% 0% To further characterise the temporal stability of PpIX fluorescence signal in PBS containing surfactant, the sample was measured when it was freshly prepared and seven days after preparation. The measured fluorescence spectra of PpIX prepared in PBS containing two different concentrations of Tween 20 (0.5%, 0.05%) are shown in Figure8. In PBS containing 0.5% ( v/v) Tween 20 medium, the spectra of PpIX aqueous sample measured at day seven after preparation was nearly identical to that of the freshly prepared sample. Its spectra showed good stability in terms of the fluorescence Figure 7. PpIX aqueous samples prepared in PBS containing different volume fractions of Tween 20 signal intensity, while in PBS containing a lower concentration of Tween 20 (0.05% v/v), the fluorescence at day seven after preparation. signal decreased at day seven after preparation.

(a) (b)

Figure 8. Fluorescence spectra (415 nm excitation) of PpIX measured at freshly prepared and at seven days after preparation in PBS( abuffer) containing (a) Tween 20 (0.5%( bv/v) ) and (b) Tween 20 (0.05% v/v). FigureFigure 8.8. FluorescenceFluorescence spectraspectra (415(415 nmnm excitation)excitation) ofof PpIXPpIX measuredmeasured atat freshlyfreshly preparedprepared andand atat sevenseven 4. Conclusion daysdays afterafter preparationpreparation inin PBSPBS bubufferffer containing containing ( a(a)) Tween Tween 20 20 (0.5% (0.5%v /v/vv)) and and ( b(b)) Tween Tween 20 20 (0.05% (0.05%v /v/vv).). 4. Conclusions In this 4.paper Conclusion, we presented the experimental results of surfactant Tween 20 volume fraction changes on backgroundInIn thisthis paper,paper medium,, wewe presentedpresent scatteringed thethe experimentalexperimental and combination resultsresults ofof surfactantsurfactant of scattering TweenTween 2020, and volumevolume absorption fractionfraction within a changes on background medium, scattering and combination of scattering, and absorption within a luminescentchanges body on background PpIX fluorescence medium, scattering, observed and combination at the surface of scattering of the, and phantom absorption s withinamples a . The data luminescentluminescent bodybody onon PpIXPpIX fluorescence,fluorescence, observedobserved atat thethe surfacesurfaceof of thethe phantom phantom samples. samples. TheThe datadata indicated thatindicatedindicate milkd thatandthat milkmilk Indian andand IndianIndian ink can inkink cancanbe bereliablybe reliablyreliably used used asas as scatterersca scattertterer andander absorberabsorband absorber inin PpIXPpIXer fluorescent fluorescentin PpIX fluorescent phantoms. phantoms. Itphantoms. is shown It It is shownis that shown that red that red sh shifts redifts sh occur iftsoccur inoccur the in emissionin the the emission emissionspectra in spectra relation spectra in to the relation in intrinsic relation to fluorescence the intrinsic to the intrinsic emission in the presence of the scatterer and surfactant, whilst the scatterer and absorber can alter fluorescencefluorescence emission emissionin the presence in the presence of the of the scatter scattererer and surfactant, surfactant, whilst whilst the scatter theer scatter and absorber ander absorber can alter the emission intensity substantially. Phospholipids in the milk in the phantom act as a can alter thesurfactant emission in the intensity phantoms. substantially. The PpIX solution Phospholipids containing 0.5% in ( thev/v) of milk Tween in 20 the presents phanto a m act as a surfactant insignificantly the phantoms. enhanced The emission PpIX signal solution compared containing with that containing 0.5% ( v/v no ) surfactant. of Tween We 20 also presents a significantlycorroborated enhanced that emission the surfactant signal content compared (>0.5% Tween with 20) was that essential containing to prepare no stable surfactant. aqueous We also corroboratedPpIX that fluorescent the surfactant phantoms. content (>0.5% Tween 20) was essential to prepare stable aqueous PpIX fluorescentAuthor phantoms. Contributions: Conceptualization, H.L.; Data curation, H.L. and F.F.; Funding acquisition, S.A.-E.; Investigation, H.L.; Methodology, H.L.; Project administration, H.L.; Resources, H.L., F.F. and M.R.; Supervision, S.A.-E.; Validation, H.L.; Writing—original draft, H.L.; Writing—review and editing, H.L., F.F., M.R. and S.A.- Author Contributions:E. All authors haveConceptualization, read and agreed to the H published.L.; Data version curation, of the manuscript. H.L. and F.F.; Funding acquisition, S.A.-E.; Investigation, H.L.; Methodology, H.L.; Project administration, H.L.; Resources, H.L., F.F. and M.R.; Supervision, Funding: This work is supported by a research grant from the Science Foundation Ireland (SFI) under Grant S.A.-E.; Validation,Number H SFI/15/RP/2828.L.; Writing. —original draft, H.L.; Writing—review and editing, H.L., F.F., M.R. and S.A.- E. All authors have read and agreed to the published version of the manuscript. Conflicts of Interest: The authors declare no conflict of interest. Funding: This work is supported by a research grant from the Science Foundation Ireland (SFI) under Grant References Number SFI/15/RP/2828.

Conflicts of Interest: The authors declare no conflict of interest.

References

Materials 2020, 13, 2105 9 of 10 the emission intensity substantially. Phospholipids in the milk in the phantom act as a surfactant in the phantoms. The PpIX solution containing 0.5% (v/v) of Tween 20 presents a signiﬁcantly enhanced emission signal compared with that containing no surfactant. We also corroborated that the surfactant content (>0.5% Tween 20) was essential to prepare stable aqueous PpIX ﬂuorescent phantoms.

Author Contributions: Conceptualization, H.L.; Data curation, H.L. and F.F.; Funding acquisition, S.A.-E.; Investigation, H.L.; Methodology, H.L.; Project administration, H.L.; Resources, H.L., F.F. and M.R.; Supervision, S.A.-E.; Validation, H.L.; Writing—original draft, H.L.; Writing—review and editing, H.L., F.F., M.R. and S.A.-E. All authors have read and agreed to the published version of the manuscript. Funding: This work is supported by a research grant from the Science Foundation Ireland (SFI) under Grant Number SFI/15/RP/2828. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Dsouza, A.V.; Lin, H.; Henderson, E.R.; Samkow, K.S.; Pogue, B.W. Review of fluorescence guided surgery systems: Identification of key performance capabilities beyond indocyanine green imaging. J. Biomed. Opt. 2016, 21, 080901. [CrossRef] 2. Breslin, T.M.; Xu, F.; Palmer, G.M.; Zhu, C.; Gilchrist, K.W.; Ramanujam, N. Autofluorescence and diffuse reflectance properties of malignant and benign breast tissue. Ann. Surg. Oncol. 2003, 11, 65–70. [CrossRef] 3. Jiang, Y.; Girard, E.J.; Pakiam, F.; Seibel, E.J. Calibration of fluorescence imaging for tumor surgical margin delineation: Multistep registration of fluorescence and histological images. J. Med. Imaging 2019, 6, 025005. 4. Toms, S.A.; Konrad, P.E.; Lin, W.-C.; Weil, R.J. Neuro-oncological applications of optical spectroscopy. Technol. Cancer Rese. Treat. 2006, 5, 231–238. [CrossRef][PubMed] 5. Ahmed, S.A.; Zhang, Z.-W.; Yoo, K.M.; Ali, M.A.; Alfano, R.R. Effect of multiple light scattering and self-absorption on the fluorescence and excitation spectra of dyes in random media. Appl. Opt. 1994, 33, 2746–2750. [CrossRef][PubMed] 6. Anand, B.S.S.; Sujatha, N. Effect of intralipid-10% in fluorescence distortion studies on liquid-tissue phantoms in UV range. J. Biophotonics 2011, 4, 92–97. [CrossRef][PubMed] 7. Müller, M.G.; Georgakoudi, I.; Zhang, Q.; Wu, J.; Feld, M.S. Intrinsic fluorescence spectroscopy in turbid media: Disentangling effects of scattering and absorption. Appl. Opt. 2001, 40, 4633–4646. [CrossRef] [PubMed] 8. Wu, B.; Gayen, S.K.; Xu, M. Fluorescence spectroscopy using excitation and emission matrix for quantification of tissue native fluorophores and cancer diagnosis. SPIE BiOS 2014, 8926, 89261. 9. Sachar, M.; Anderson, K.E.; Ma, X. Protoporphyrin IX: The good, the bad, and the ugly. J. Pharmacol. Exp. Ther. 2016, 356, 267–275. [CrossRef] 10. Valdés, P.A.; Roberts, D.W.; Lu, F.-K.; Golby, A. Optical technologies for intraoperative neurosurgical guidance. Neurosurg. Focus. 2016, 40, E8. [CrossRef] 11. Xie, H.; Xie, Z.; Mousavi, M.; Bendsoe, N.; Brydegarrd, M.; Axeisson, J.; Andersson-Engels, S. Design and validation of a fiber optic point probe instrument for therapy guidance and monitoring. J. Biomed. Opt. 2014, 19, 071408. [CrossRef] 12. Haj-Hosseini, N.; Richter, J.; Andersson-Engels, S.; Wårdell, K. Optical touch pointer for fluorescence guided glioblastoma resection using 5-aminolevulinic acid. Laser Surg. Med. 2010, 42, 9–14. [CrossRef][PubMed] 13. Marois, M.; Bravo, J.; Davis, S.C.; Kanick, S.C. Characterization and standardization of tissue-simulating protoporphyrin IX optical phantoms. J. Biomed. Opt. 2016, 21, 035003. [CrossRef][PubMed] 14. Pogue, B.W.; Patterson, M.S. Review of tissue simulating phantoms for optical spectroscopy, imaging and dosimetry. J. Biomed. Opt. 2006, 11, 041102. [CrossRef][PubMed] 15. Inamura, I.; Isshiki, M.; Araki, T. Bull. Solubilization of hemi in neutral and acidic aqueous solution by forming complexes with water-soluble macromolecules. Chem. Soc. Jpn. 1989, 62, 2413–2415. [CrossRef] 16. Scolaro, L.M.; Castriciano, M.; Romeo, A.; Patane, S.; Cefalı, E.; Allegrini, M. Aggregation behavior of protoporphyrin IX in aqueous solutions: Clear evidence of vesicle formation. J. Phys. Chem. B 2002, 106, 2453–2459. [CrossRef] 17. Maiti, N.C.; Krishna, M.M.G.; Britto, P.J.; Periasamy, N. Fluorescence dynamics of dye probes in micelles. J. Phys. Chem. B 1997, 101, 11051–11060. [CrossRef] Materials 2020, 13, 2105 10 of 10

18. Markwardt, N.K.; Hai-Hosseini, N.; Hollnburger, B.; Stepp, H.; Zelenkov, P.; Rühm, A. 405nm versus 633 nm for protoporphyrin IX excitation in fluorescence-guided sterootactic biopsy of brain tumors. J. Biophotonics 2016, 9, 901–912. [CrossRef] 19. Wagnieres, G.A.; Willem, M.; Wilson, B.C. ln vivo fluorescence spectroscopy and imaging for neuro oncological applications. Photochem. Photobiol. 1998, 68, 603–632. [CrossRef] 20. Johansson, A.; Faber, F.; Kniebühler, G.; Stepp, H.; Sroka, R.; Egensperger, R.; Beyer, W.; Kreth, F.-W. Protoporphyrin IX fluorescence and photobleaching during interstitial photodynamic therapy of malignant gliomas for early treatment prognosis. Laser Surg. Med. 2013, 45, 225–234. [CrossRef] 21. Lu, H.; Floris, F.; Rensing, M.; Andersson-Engels, S. Fluorescence spectroscopy study of protoporphyrin IX in tissue-like phantoms. SPIE-Intl. Soc. Opt. Eng. 2019, 11190, 111901S. 22. Wilson, B.C.; Jacques, S.L. Optical reflectance and transmittance of tissues: Principles and applications. IEEE J. Quantum Electron. 1990, 26, 2186–2199. [CrossRef] 23. Michels, R.; Foschum, F.; Kienle, A. Optical properties of fat emulsions. Opt. Express 2008, 16, 5907–5925. [CrossRef] 24. Ninni, P.D.; Martelli, F.; Zaccanti, G. The use of India ink in tissue-simulating phantoms. Opt. Express 2010, 18, 26854–26865. [CrossRef][PubMed] 25. Lu, H.; Gunther, J.; Andersson-Engles, S. Determination optical properties of tissue-like phantoms using diffuse reflectance and transmittance spectroscopy. Opt. Health Care Biomed. Opt. 2018, 10820, 108202X. 26. Gunther, J.; Lu, H.; Andersson-Engles, S. Combination of diffuse reflectance and transmittance spectroscopy to obtain optical properties of liquid phantoms. Opt. Eng. 2020, 59, 024109. [CrossRef] 27. Zehentbauer, F.M.; Moretto, C.; Stephen, R.; Thevar, T.; Gilchrist, J.R.; Pokrajac, D.; Richard, K.L.; Kiefer, J. Fluorescence spectroscopy of Rhodamine 6G: Concentration and solvent effects. Spectrochim. Acta A 2014, 121, 147–151. [CrossRef] 28. Suah, F.B.M.; Ahmad, M.; Mehamod, F.S. Effect of non-ionic surfactants to the Al (III)-morin complex and its application in determination of Al (III) ions: A preliminary study. Malays. J. Anal. Sci. 2017, 21, 793–800. 29. Yang, R.-H.; Wang, K.-M.; Xiao, D.; Yang, X.-H.; Li, H.-M. A selective optical chemical sensor for the determination of Tween-60 based on fluorescence enhancement of tetraphenyporphyrin. Anal. Chim. Acta 2000, 404, 205–211. [CrossRef] 30. Stocker, S.; Foschun, F.; Krauter, P.; Bergmann, F.; Hohmann, A.; Happ, C.S.; Kienle, A. Broadband optical properties of milk. Appl. Spectrosc. 2017, 71, 951–962. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).