Nanoscale View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
PAPER
Cite this: DOI: 10.1039/c4nr03952a
View Journal
Dual core quantum dots for highly quantitative ratiometric detection of trypsin activity in cystic fibrosis patients† Iván Castelló Serrano,‡a Georgiana Stoica,*‡a Alba Matas Adamsa and Emilio Palomares*a,b We present herein two colour encoded silica nanospheres (2nanoSi) for the fluorescence quantitative ratiometric determination of trypsin in humans. Current detection methods for cystic fibrosis diagnosis are slow, costly and suffer from false positives. The 2nanoSi proved to be a highly sensitive, fast (minutes), and single-step approach nanosensor for the screening and diagnosis of cystic fibrosis, allowing the quantification of trypsin concentrations in a wide range relevant for clinical applications (25–350 μg L−1). Furthermore, as trypsin is directly related to the development of cystic fibrosis (CF), different human genotypes, i.e. CF homozygotic, CF heterozygotic, and unaffected, respectively, can be determined using our
Received 14th July 2014, Accepted 10th September 2014
2nanoSi nanospheres. We anticipate the 2nanoSi system to be a starting point for non-invasive, easy-touse and cost effective ratiometric fluorescent biomarkers for recessive genetic diseases like human cystic
DOI: 10.1039/c4nr03952a
fibrosis. In a screening program in which the goal is to detect disease and also the carrier status, early
www.rsc.org/nanoscale
diagnosis could be of great help.
Introduction The detection of human recessive diseases has been dominated by the use of fluorescent biomarkers, based on organic dyes, helping researchers to study and analyse gene expression,1 cell cycle,2 and enzymatic activity.3 Among several proteolytic enzymes, trypsin has attracted much attention, as it is a target in the study of various important human recessive diseases including cystic fibrosis (CF).4 Recent advances in the field of nanoscience and the evaluation of enzyme proteolytic processes have taken advantage of short peptide substrates5,6 that can be specifically cleaved by trypsin7 allowing a new route to analyze and trace enzymatic activity, and therefore its concentration in human serum, blood or tissues. However, many of these studies8,9 focused just on qualitative data for the presence of the homozygous recessive genotype (the individuals with the disease) and, therefore, cannot differentiate between heterozygotic human carriers (those individuals that
a Institute of Chemical Research of Catalonia (ICIQ), Avinguda del Països Catalans 16, 43007 Tarragona, Spain. E-mail: [email protected], [email protected] b Catalan Institution for Research and Advanced Studies (ICREA), Passeig de Lluis Companys 23, 08010 Barcelona, Spain † Electronic supplementary information (ESI) available. See DOI: 10.1039/ c4nr03952a ‡ These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2014
carry an allele for the disease but do not express it) and the unaffected persons. Of particular interest, in these bio-analytical systems, is the use of simple Förster fluorescence resonance energy transfer (FRET),10–12 a process based on two fluorophores where initially one fluorophore can transfer energy to the second one. This pair of fluorophores is known as the energy donor and energy acceptor pair, respectively. The efficiency of this energy transfer process is inversely proportional to the sixth power of the distance between the donor and the acceptor making FRET extremely sensitive to short distances between the two fluorophores.13–15 Pioneering work by other research groups has focussed on QD-QD,16 QD-dye,17 QD-dye-dye (in a two-step FRET process18–20 or dual-donor FRET21,22 systems), rare earth-QDdye,23 or QD-plasmonic nanoparticles.24,25 Besides, the use of peptide-conjugated quantum dots proved to be successful for proteolytic enzyme assays.10,18 We took this approach one step further by protecting the quantum dots with silica shells26 and including a fixed standard probe inside the nanosensor, thus upgrading the properties of the system from qualitative sensing to accurate ratiometric determination of enzymatic activity. Due to the properties of quantum dots and the efficiency of the FRET process, our system can be used as an excellent probe for the high-throughput trypsin assay in human faeces as an alternative to the traditional approaches for CF diagnosis27 and other pancreatic-related diseases.
Nanoscale
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Paper
Nanoscale
Scheme 1 Pictorial description of the 2nanoSi system with QD660 (red) at the core of the nanosilica sphere, QD540 (green) at the silica coating layer and the attached small TAMRA-labelled peptide on the outer surface of the silica coating layer.
Pancreatic insufficiency (PI) is commonly associated with diseases such as cystic fibrosis (the most common cause) or pancreatitis,28 in which the patients have a shortage of the digestive enzymes necessary to break down food. Besides, certain gastrointestinal diseases, such as stomach ulcers and Crohn’s disease, may also induce PI. However, all these diseases have their own clinical and genetic features in addition to the biochemical markers. For example, the trypsin concentration is always lower in cystic fibrosis,28 while it is higher in Crohn’s disease.29 Fecal chymotrypsin and fecal elastase 1 (FE1) are other biochemical markers used to assess exocrine PI.30 Both trypsin and chymotrypsin are exclusively of pancreatic origin, and their quantitative measurement can be used to indicate pancreatic function. FE1 is not affected by bacterial degradation (as trypsin or chymotrypsin) or pancreatic enzyme supplementation, thus being useful to determine the pancreatic function in CF. However, low FE1 levels can be found in children with short gut syndrome as well as in patients with the Shwachman–Diamond syndrome who otherwise have normal pancreatic function.31 Therefore, the trypsin test is a promising method to help detect CF in symptomatic newborns or infants, or anyone to determine whether a person is a CF carrier (i.e. has a CF gene mutation). Early diagnosis of cystic fibrosis has the potential to dramatically improve the life quality and survival of millions of people dealing with this chronic illness, of genetic origin, by developing early and personalised treatment, preventing malnutrition and allowing access to counseling. Many countries have established a default screening for newborns that involves quantifying trypsinogen in the blood of the babies (immunoreactive trypsinogen test, or IRT), which typically makes use of fluorescence-based immunoassays. However, these tests still have the drawbacks of immunoassays with respect to the specificity, time-effectiveness and userfriendliness. Additionally, due to a considerable percentage of false positives inherent to this test, babies with abnormal results must be either tested again for IRT or, alternatively, undergo genetic testing to determine whether they have at
Nanoscale
least one mutated copy of the cystic fibrosis gene (human carriers).32,33 Infants with a second abnormal newborn IRT screen or carrier genotype need a sweat test to confirm the cystic fibrosis diagnosis. Sweat testing however cannot be performed accurately in the first few days after birth, and it may be difficult to obtain an adequate sweat sample during the first two to three weeks after birth, especially for premature babies. If the volume of sweat obtained is inadequate, the test should be repeated. The overall procedure is long (weeks to months) and sometimes considered invasive by the parents.34 We present herein for the first time a well-designed and cost effective nanosystem, based on FRET, with an internal reference that unequivocally allows the ratiometric determination of trypsin in the range of enzyme concentrations relevant for clinical diagnosis of cystic fibrosis in minutes, with high sensitivity. Our method provides specific information about whether or not the individuals are CF carriers (heterozygotic genotype), and therefore about their risk of having a child affected with CF (homozygotic genotype). CF is the most life threatening recessive genetic condition affecting Caucasian children and diagnosis is often made at a point when significant and irreversible damage has already occurred. The challenge is to identify a detection method that will allow for early diagnosis and treatment with sufficient time to adopt lifestyle changes or use agents to prevent or retard disease development. A short peptide substrate previously marked with TAMRA, a rhodamine-derived (carboxy-tetramethyl-rhodamine) dye with fluorescence emission at λ = 575 nm (see Scheme 1), is anchored to the surface of a silica nanosphere containing two types of CdSe quantum dot nanocrystals with different fluorescence emission wavelengths.
Experimental Materials Trypsin (13 000–20 000 units mg−1), selenium powder (99.99%,), 1-octadecene (90%), tetraethyl orthosilicate (TEOS, 99.999%), 3-aminopropyl-trimethoxysilane (APTMS, 97%), (3-trihydroxysilyl) propyl methylphosphonate (THPMP), and phosphate buffer solution (PBS) were purchased from Sigma-Aldrich. Ammonia was purchased from Fluka. Synthesis of QD. The procedure to obtain CdSe QDs was using TOPO-TOP as a capping ligand.35 A selenium solution was prepared by mixing 0.4 g of selenium powder, 10 mL of TOP and 0.2 mL of anhydrous toluene and stirred under argon. A mixture of 20 g of TOPO and 0.25 g of cadmium acetate dihydrate was placed in a round-bottomed flask, stirred and heated to 150 °C under argon and the temperature was increased to 320 °C. When the mixture reached 320 °C the selenium solution was quickly added and then cooled to 270 °C. The reaction was run for a specific time (depending on the size of the QDs that we expected to achieve). After stopping the reaction, samples were cooled and washed with ethanol and acetone three times and stored in chloroform. Synthesis of QD@silica nanospheres. The experiments were performed using CdSe core quantum dots with emission wavelengths at 540 nm (green (2.7 nm)) and 660 nm (red (5.6 nm)). Next, single and two color quantum dot embedded silica nanospheres, i.e. QD540@silica and QD660@silica@QD540@silica, respectively, were prepared by the reverse microemulsion method as detailed elsewhere.36,37 Amine functionalization was done by adding 30 μL of APTMS and 80 μL of THPMP to the reaction and allowing to stir for 24 hours at room temperature. Peptide-functionalization of QD@silica nanospheres. The positive surface of the QD540@silica nanospheres (given by APTMS) was functionalized with two different TAMRA-labelled peptides: Pro-active: NH2-Cys-Lys-Arg-Val-Lys-TAMRA In-active: NH2-Cys-Lys-Pro-Val-Lys-TAMRA The two different sequences of TAMRA-labelled peptides were particularly designed for our study by Biotrend. Glutaraldehyde was used to bind the amine functional groups of the peptides with the amine groups of the silica nanosphere surface as described in the following methodology. 2 mL of QD@silica-APTMS in PBS and 0.25 mL of glutaraldehyde (0.25 mg mL−1) were mixed and shaken for 4 h at 37 °C. Then the solution was centrifuged to remove the unreacted glutaraldehyde and finally the pellet was resuspended in PBS. A determined amount of TAMRA-labelled peptides was added and shaken for 20 hours at 4 °C. After this time, the solution was centrifuged to remove the unbound peptides. In the end, the TAMRA-labelled peptide-functionalized QD@silica nanospheres were resuspended in 5 mL PBS and stored. Enzymatic digestion. A determined concentration of trypsin between 25 and 350 μg L−1 was added to the labelled peptidesnanospheres and mixed for a period of time: from 30 s to 120 min, at 37 °C. After this time, the mixture was heated to 70 °C and held for 1 h in order to denaturalize the enzyme and stop the reaction. If not measured immediately after the
This journal is © The Royal Society of Chemistry 2014
Paper
enzymatic digestion, it is recommended to keep the samples in a freezer until the analysis. Characterization techniques The formation of the QDs@silica nanospheres was directly visualized by transmission electron microscopy (TEM) using a JEOL 1011 microscope. The UV-Visible and fluorescence spectra were recorded using a 1 cm path length quartz cell on a Shimadzu UV Spectrophotometer 1700 and an AmincoBowman Series 2 luminescence spectrometer. Preparation of the biological samples Patients. Eligible patients included males and females from four families (11 subjects in total, A–K in Table S1†) with a confirmed diagnosis of cystic fibrosis. Four of the subjects were taking daily optimal doses of a pancreatic enzyme to increase the protein absorption. Written informed consent was obtained from all patients or their parents, and assent was also obtained from minors. Stool sample collection. Trypsin is firmly attached to the solid matter of feces and after the homogenization the supernatant shows very little enzymatic activity.38,39 Therefore, all the fecal specimens analyzed have to be sufficiently solid to be handled with a spatula. As the enzymes are very well embedded in the solid matter of the feces, we can correlate the measurements in concentration per gram with the concentration per liter in the duodenum.40 1 g of solid feces per participant, stored at −8 °C, was added to 3 mL of PBS ( pH 7.2) and vigorously stirred until a homogeneous solution was obtained. For the measurements, the mixture was heated up to 37 °C and 1 mL of TAMRA-labelled peptides-2nanoSi nanospheres in PBS was added. The reaction was run for 10 min, and then heated up to 70 °C to stop the enzymatic activity by denaturalization for 1 h. Finally, the samples were measured by fluorescence spectrometry. Statistical analysis. Each sample was measured five times and the mean values were reported with their standard deviation (SD) for each participant. The measurements were carried out in five consecutive days, with one set per day, each set of measurements consisting of 11 individual samples.
Results and discussion In brief, the 2nanoSi system consists of a 77 nm silica nanosphere with a nanocrystal CdSe (CdSe660) core with the luminescence emission maximum at λ = 660 nm, and a second shell of silica embedding the second type of CdSe nanocrystal with a luminescence emission maximum at λ = 540 nm (CdSe540). Upon light excitation at λ = 405 nm, both types of CdSe quantum dots undergo fluorescence emission. Moreover, due to the FRET process between the CdSe540@2nanoSi, as the energy donor, and the TAMRA dye, as the energy acceptor, it is possible to register TAMRA fluorescence emission at λ = 575 nm. Control experiments show negligible emission of
Nanoscale
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Paper
TAMRA upon excitation at λ = 405 nm (see Fig. S1†). In the presence of trypsin, the proteolytic activity of the enzyme leads to the cleavage of the peptide and the FRET process is disrupted. Thus, by measuring the signal increase at λ = 540 nm that corresponds to the CdSe540@2nanoSi, it is possible to correlate the changes in the emission intensity with the enzyme activity using an internal reference, in which fluorescence emission remains constant throughout the proteolytic process. In this case, the internal reference is the CdSe660@2nanoSi with fluorescence emission at λ = 660 nm. Hence, the reference CdSe660 material has the advantage of not being influenced by the TAMRA concentration and the excitation intensity as it occurs with commercial immuno-fluorescence intensity based measurements. FRET is observed when the “donor” and “acceptor” are sufficiently close to each other and when there is sufficient spectral overlap between the emission spectrum of the donor and the absorption spectrum of the acceptor. The average distance between TAMRA and CdSe660 is around 40 nm, while the photoluminescence emission spectrum of the label and the photoluminescence absorption spectrum of the first QD population (CdSe660) have a spectral overlap below 30% area/area, conditions which effectively preclude FRET between the label (as donor) and the CdSe660 (as the acceptor). The use of a single quantum dot fluorophore in our system (1nanoSi, Fig. S2 and S3†) already allowed for the qualitative analysis of the trypsin proteolytic activity. Nonetheless, a key figure of merit of our approach is to be able to quantify the enzymatic activity and, thus, the use of an internal reference in our fluorescent system. As mentioned above, we added the second population of QDs at the core of the silica nanospheres with fluorescence emission at λ = 660 nm (Fig. S4†). Thus, the use of two QD fluorophores with distinguishable emission spectra allows us to establish a measurable ratio, namely I540/I660. Moreover, the emission intensity at λ = 660 nm remains constant as the QD660 at the silica nanospheres does not influence the FRET process. The key feature in our 2nanoSi is that due to the absorption properties of the nanocrystal quantum dots, both fluorophores can be excited at the same wavelength (λ = 405 nm), but only the nanocrystals in the outer layer emitting at 540 nm undergo FRET with the TAMRA dye. Despite the small overlap between the emission of QD660 and the absorption of TAMRA, the distance between them (around 38 nm) does not allow energy transfer to occur. On the other hand, the distance and the overlap between QD540 and TAMRA are sufficient for FRET to occur. By TCSPC and steady-state emission spectroscopy, we determined a FRET efficiency of 47% between the QD540 and the TAMRA dye. The calculated R0 (the distance at which the efficiency of FRET is 50%) is 4.982 nm, and the distance between CdSe540 and TAMRA dye (r) is 5.10 nm calculated from the fluorescence measurements. Next, the number of TAMRA-labelled peptides attached to the silica nanosphere surface was found to be one via the Förster model.41 For the complete calculations, see the ESI.† This result clearly indicates that with only one QD@silica – peptide-dye donor–acceptor pair, our system displays a
Nanoscale
Nanoscale
remarkable FRET efficiency of approx. 50%. Previous studies with one pair of QD donor–dye acceptor reported lower efficiency, i.e. 22%.11 These authors required up to five acceptor molecules to increase the efficiency up to 58%. Our system is highly FRET efficient and the design applies to different types of quantum dot populations as long as the distance and spectral overlap between the emission spectrum of the donor and the absorption spectrum of the acceptor are particularly adequate. Fig. 1 shows the changes in the fluorescence emission of 2nanoSi in the presence of two different trypsin concentrations, 25 μg L−1 (Fig. 1a) versus 350 μg L−1 (Fig. 1b). As expected, increasing the enzyme concentration leads to faster recovery of the original emission at 540 nm (Fig. 1b). The complete enzyme concentration study can be found in Fig. S8.† The trypsin concentration for the different cystic fibrosis genotypes is undeniably tissue specific. Consequently, the
Fig. 1 2nanoSi emission spectra (a) after digestion with 25 μg L−1 of trypsin, and (b) after digestion with 350 μg L−1. The figure legends correspond to the enzymatic digestion time in min (0–30 min).
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Nanoscale
Fig. 2 Kinetics of the enzymatic digestion as a function of trypsin concentration using the ratio between PL emission peaks of both QD540 and QD660 of the 2nanoSi system.
trypsin concentration in fecal samples displays the following trend: CF-diagnosed 0–30 μg g−1, while in a non-CF medical check it is above 80 μg g−1.42 On the other hand, the distribution of the serum trypsin concentration in patients with cystic fibrosis43 is homozygotic (0–90 μg L−1), normal (160–340 μg L−1), and heterozygotic (91–349 μg L−1), respectively. Coincidentally, the trypsin ranges in stool samples highly match those in blood, i.e. non-CF > 80 μg g−1 stool versus heterozygote and healthy > 90 μg L−1 blood. Taking into account these values of trypsin, and considering the specificity of the 2nanoSi, we performed the study of the enzymatic digestion using the “pro-active” labelled peptide in a wide range of trypsin concentrations (25–350 μg L−1) that fall in the ranges found for CF patients mentioned above. Fig. 2 shows the ratio between the PL emission peaks of both QD540 and QD660 (I540/I660) during enzymatic digestion for 120 min in different enzyme concentrations. As expected, there is a direct relationship between the enzymatic digestion rate and the enzyme concentration that allowed us to unequivocally differentiate between the different enzyme concentrations used. In other words, the higher the trypsin concentration the shorter the time we need to reach a plateau in our kinetic measurements. Hence, by analysing the trypsin activity at early digestion times using our 2nanoSi ratiometric system (for example 10 min) we can differentiate between the different trypsin concentrations as shown in Fig. 3. As can be seen, the calibration curve proved to be highly linear (χ2 = 0.99) and allowed us to determine the trypsin concentration at levels that are clinically relevant for the cystic fibrosis prognosis, i.e. CF homozygotic (0–90 μg L−1) and nonCF (>90 μg L−1) including both CF heterozygotic and normal, respectively. The trypsin concentration in human faecal samples was determined based on the calibration curve plotted in the
This journal is © The Royal Society of Chemistry 2014
Paper
Fig. 3 Calibration curve obtained by fitting the fluorescence I540/I660 experimental values from Fig. 2 at the enzyme digestion time of 10 min. Red, orange and green correspond to CF homozygotic, CF heterozygotic and normal, respectively.
Fig. 4 Results of trypsin determination in human subject samples obtained with the proposed 2nanoSi nanospheres. The symbols correspond to participants in the study without specific medication (Δ), participants in the study under treatment with supplementary pancreatic enzymes (+), and participants in the study not taking the supplementary pancreatic enzymes (o).
coordinates: I540/I660 (−) versus trypsin concentration (μg L−1), following the equation: I 540 =I 660 ¼ 1:6811 þ 0:004625 ½trypsin
The results of trypsin determination performed with the proposed model fitted with the CF heterozygotic range (triangle (Δ) and cross (+) symbols in Fig. 4), indicating that all
Nanoscale
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Paper
the subjects are carrying at least one CF mutation. Nonetheless, taking into account that four of the participants (marked by (+) in Fig. 4) have a confirmed diagnosis of cystic fibrosis and are taking supplementary pancreatic enzymes (8–10 capsules of Creon® 25000 or 20 capsules of Creon® 10000 daily, respectively), the additional enzymatic activity (61–92%) has to be subtracted from the total trypsin concentration above. This observation is in good agreement with previous studies in the literature in which the CPA (coefficient of protein absorption) increased in the range 59–90% after supplementary pancreatic enzyme uptake.44,45 Accordingly, the values obtained for these four particular patients can now be found in the CF homozygotic range (symbol (o) in Fig. 4). (For the whole study see Fig. S9 and Tables S1 and S2 in ESI.†) The rest of the samples, provided by the parents of the four CF homozygotic patients, were considered as the control. Each parent of a patient with this condition is by definition a heterozygote, in other words he/ she carries only one copy of the CF gene but does not develop the disease.
Conclusions By combining the specificity of biomolecular interactions with the tunability of quantum dot and organic dye optical properties, we have developed for the first time a detection system capable of cystic fibrosis prognosis, both qualitatively and quantitatively. The trypsin enzymatic activity was reported via the FRET process, which mediates changes in the quantum dot emission spectra, between the organic dye TAMRA (λem = 575 nm) and green CdSe quantum dots (λem = 540 nm). Crossreactivity and signal overlap were avoided by designing enzyme-specific peptidic sequences. The 2nanoSi proved to be a fast and highly sensitive single-step nanosensor, allowing the quantification of trypsin concentrations in a wide range (25–350 μg L−1), which is clinically relevant for the diagnosis of cystic fibrosis (CF). This range covers different human genotypes, i.e. CF homozygotic (0–90 μg L−1) and CF heterozygotic (91–349 μg L−1). Using our 2nanoSi nanospheres in biological samples, four participants in the study were confirmed homozygotic, while the other seven were heterozygotic for CF. The biomarker has been shown to provide significant prognostic information, and hence represents a promising system for proof-of-concept demonstration of the proposed approach. We anticipate the 2nanoSi system to be a starting point for noninvasive, easy-to-use and cost effective ratiometric fluorescent biomarkers for recessive genetic diseases like human cystic fibrosis.
Acknowledgements We would like to thank ICIQ, ICREA, Fundació La Caixa, and the Spanish MICINN project CTQ2010-18859 for financial support. EP also thanks the EU for the ERCstg Polydot. We are
Nanoscale
Nanoscale
thankful to Núria Mir for facilitating the collection of the biological samples from the participants in the study.
Notes and references 1 N. Steuerwald, J. Cohen, R. J. Herrera and C. A. Brenner, Mol. Hum. Reprod., 1999, 11, 1034–1039. 2 A. W. Murray and M. N. Kirschner, Nature, 1989, 339, 575– 580. 3 A. Henning, H. Bakirci and W. M. Nau, Nat. Methods, 2007, 4, 629–632. 4 S. S. Antunovic, M. Lukac and D. Vujovic, Clin. Pharmacol. Ther., 2013, 93, 86–97. 5 O. Schilling and C. M. Overall, Nat. Biotechnol., 2008, 26, 685–694. 6 D. A. Thomas, P. Francis, C. Smith, S. Ratcliffe, N. J. Ede and C. Kay, Proteomics, 2006, 6, 2112–2120. 7 K. E. Sapsford, D. Farrell, S. Sun, A. Rasooly, H. Mattoussi and I. L. Medintz, Sens. Actuators, B, 2009, 139, 13–21. 8 M. Chalfie, Y. Tu, G. Euskirchen, W. W. Ward and D. C. Prasher, Science, 1994, 263, 802–805. 9 M. Chelfie and S. R. Kain, in Methods of Biochemical Analysis, ed. M. Chalfie and S. R. Kain, Wiley, New York, 2005, vol. 47. 10 W. R. Algar, A. Malonski, J. R. Deschamps, J. B. BlancoCanosa, K. Susumu, M. H. Stewart, B. J. Johnson, P. E. Dawson and I. L. Medintz, Nano Lett., 2012, 12, 3793– 3802. 11 I. L. Medintz, H. T. Uyeda, E. R. Goldman and H. Mattoussi, Nat. Mater., 2006, 4, 435–446. 12 Y. P. Ho and K. W. Leong, Nanoscale, 2010, 2, 60–68. 13 I. L. Medintz and H. Mattoussi, Phys. Chem. Chem. Phys., 2009, 11, 17–45. 14 H. Zhang, G. Feng, Y. Guo and D. Zhou, Nanoscale, 2013, 5, 10307–10315. 15 S. Chatterjee, J. B. Lee, N. V. Valappil, D. Luob and V. M. Menon, Nanoscale, 2012, 4, 1568–1571. 16 C. Y. Chen, C. T. Cheng, C. W. Lai, P. W. Wu, K. C. Wu, P. T. Chou, Y. H. Chou and H. T. Chiu, Chem. Commun., 2006, 263–265. 17 G. B. Kim and Y. P. Kim, Theranostics, 2012, 2, 127–138. 18 I. L. Medintz, A. R. Clapp, H. Mattoussi, E. R. Goldman, B. Fisher and J. M. Mauro, Nat. Mater., 2003, 2, 630–638. 19 H. H. Chen, Y. P. Ho, X. Jiang, H. Q. Mao, T. H. Wang and K. W. Leong, Nano Today, 2009, 4, 125–134. 20 R. Alam, J. Zylstra, D. M. Fontaine, B. R. Branchinib and M. M. Maye, Nanoscale, 2013, 5, 5303–5306. 21 H. Zhang and D. Zhou, Chem. Commun., 2012, 48, 5097– 5099. 22 K. Ouadahi, K. Sbargoud, E. Allarda and C. Larpent, Nanoscale, 2012, 4, 727–732. 23 W. R. Algar, A. P. Malanoski, K. Susumu, M. H. Stewart, N. Hildebrandt and I. L. Medintz, Anal. Chem., 2012, 84, 10136–10146.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Nanoscale
24 S. B. Lowe, J. A. G. Dick, B. E. Cohen and M. M. Stevens, ACS Nano, 2012, 6, 851–857. 25 P. L. Xavier, K. Chaudhari, P. K. Verma, S. K. Pal and T. Pradeep, Nanoscale, 2010, 2, 2769–2776. 26 G. Stoica, I. Castelló Serrano, A. Figuerola, I. Ugarte, R. Pacios and E. Palomares, Nanoscale, 2012, 4, 5409–5419. 27 A. F. Heeley and D. Watson, Clin. Chem., 1983, 29, 2011– 2018. 28 J. F. Pohl and D. J. Easley, Pract. Gastroenterol., 2003, 38–48. 29 T. Midtvedt, E. Zabarovsky, E. Norin, J. Bark, R. Gizatullin, V. Kashuba, O. Ljungqvist, V. Zabarovska, R. Möllby and I. Ernberg, PLoS One, 2013, 8, 1–9. 30 J. Leus, S. Van Biervliet and E. Robberecht, Eur. J. Pediatr., 2000, 159, 563–568. 31 S. Bebarry, L. Ellis, M. Corey, M. Marcon and P. Durie, J. Pediatr., 2002, 141, 84–90. 32 DELFIA Neonatal IRT assay, PerkinElmer Inc., 2009. 33 B. A. Lindau-Shepard and K. A. Pass, Clin. Chem., 2010, 56, 445–450. 34 A. Mishra, R. Greaves and J. Massie, Clin. Biochem. Rev., 2005, 26, 135–153. 35 J. Aldana, Y. A. Wang and X. G. Peng, J. Am. Chem. Soc., 2001, 112, 8844–8850.
This journal is © The Royal Society of Chemistry 2014
Paper
36 I. Castelló Serrano, Q. Ma and E. Palomares, J. Mater. Chem., 2011, 21, 17673–17679. 37 Q. Ma, I. Castelló Serrano and E. Palomares, Chem. Commun., 2011, 47, 7071–7073. 38 B. J. Haverback, B. J. Dyce, P. J. Gutentag and D. W. Montgomery, Gastroenterology, 1963, 44, 588–597. 39 W. P. Dick, Am. J. Dig. Dis., 1967, 12, 310–317. 40 J. S. Smith, I. Ediss, M. A. Mullinger and A. Bogoch, Can. Med. Assoc. J., 1971, 104, 691–694. 41 A. R. Clapp, I. L. Medintz, M. Mauro, B. R. Fisher, M. G. Bawendi and H. Mattoussi, J. Am. Chem. Soc., 2004, 126, 301–310. 42 G. J. Barbero, M. S. Sibling, J. M. Marino and R. Seibel, Am. J. Dis. Child, 1966, 112, 536–540. 43 P. Dandona, M. E. Hodson and J. C. Batten, J. Clin. Pathol., 1983, 36, 760–792. 44 M. M. Perez, A. D. Newcomer, C. G. Moertel, C. L. W. Go and E. P. Dimagno, Cancer, 1983, 52, 346–352. 45 M. W. Konstan, R. C. Stern, J. R. Trout, J. M. Sherman, H. Eigen, J. S. Wagener, C. Duggan, M. E. B. Wohl and P. Colin, Aliment. Pharmacol. Ther., 2004, 20, 1365– 1371.
Nanoscale
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
PAPER
Cite this: DOI: 10.1039/c4nr03952a
View Journal
Dual core quantum dots for highly quantitative ratiometric detection of trypsin activity in cystic fibrosis patients† Iván Castelló Serrano,‡a Georgiana Stoica,*‡a Alba Matas Adamsa and Emilio Palomares*a,b We present herein two colour encoded silica nanospheres (2nanoSi) for the fluorescence quantitative ratiometric determination of trypsin in humans. Current detection methods for cystic fibrosis diagnosis are slow, costly and suffer from false positives. The 2nanoSi proved to be a highly sensitive, fast (minutes), and single-step approach nanosensor for the screening and diagnosis of cystic fibrosis, allowing the quantification of trypsin concentrations in a wide range relevant for clinical applications (25–350 μg L−1). Furthermore, as trypsin is directly related to the development of cystic fibrosis (CF), different human genotypes, i.e. CF homozygotic, CF heterozygotic, and unaffected, respectively, can be determined using our
Received 14th July 2014, Accepted 10th September 2014
2nanoSi nanospheres. We anticipate the 2nanoSi system to be a starting point for non-invasive, easy-touse and cost effective ratiometric fluorescent biomarkers for recessive genetic diseases like human cystic
DOI: 10.1039/c4nr03952a
fibrosis. In a screening program in which the goal is to detect disease and also the carrier status, early
www.rsc.org/nanoscale
diagnosis could be of great help.
Introduction The detection of human recessive diseases has been dominated by the use of fluorescent biomarkers, based on organic dyes, helping researchers to study and analyse gene expression,1 cell cycle,2 and enzymatic activity.3 Among several proteolytic enzymes, trypsin has attracted much attention, as it is a target in the study of various important human recessive diseases including cystic fibrosis (CF).4 Recent advances in the field of nanoscience and the evaluation of enzyme proteolytic processes have taken advantage of short peptide substrates5,6 that can be specifically cleaved by trypsin7 allowing a new route to analyze and trace enzymatic activity, and therefore its concentration in human serum, blood or tissues. However, many of these studies8,9 focused just on qualitative data for the presence of the homozygous recessive genotype (the individuals with the disease) and, therefore, cannot differentiate between heterozygotic human carriers (those individuals that
a Institute of Chemical Research of Catalonia (ICIQ), Avinguda del Països Catalans 16, 43007 Tarragona, Spain. E-mail: [email protected], [email protected] b Catalan Institution for Research and Advanced Studies (ICREA), Passeig de Lluis Companys 23, 08010 Barcelona, Spain † Electronic supplementary information (ESI) available. See DOI: 10.1039/ c4nr03952a ‡ These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2014
carry an allele for the disease but do not express it) and the unaffected persons. Of particular interest, in these bio-analytical systems, is the use of simple Förster fluorescence resonance energy transfer (FRET),10–12 a process based on two fluorophores where initially one fluorophore can transfer energy to the second one. This pair of fluorophores is known as the energy donor and energy acceptor pair, respectively. The efficiency of this energy transfer process is inversely proportional to the sixth power of the distance between the donor and the acceptor making FRET extremely sensitive to short distances between the two fluorophores.13–15 Pioneering work by other research groups has focussed on QD-QD,16 QD-dye,17 QD-dye-dye (in a two-step FRET process18–20 or dual-donor FRET21,22 systems), rare earth-QDdye,23 or QD-plasmonic nanoparticles.24,25 Besides, the use of peptide-conjugated quantum dots proved to be successful for proteolytic enzyme assays.10,18 We took this approach one step further by protecting the quantum dots with silica shells26 and including a fixed standard probe inside the nanosensor, thus upgrading the properties of the system from qualitative sensing to accurate ratiometric determination of enzymatic activity. Due to the properties of quantum dots and the efficiency of the FRET process, our system can be used as an excellent probe for the high-throughput trypsin assay in human faeces as an alternative to the traditional approaches for CF diagnosis27 and other pancreatic-related diseases.
Nanoscale
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Paper
Nanoscale
Scheme 1 Pictorial description of the 2nanoSi system with QD660 (red) at the core of the nanosilica sphere, QD540 (green) at the silica coating layer and the attached small TAMRA-labelled peptide on the outer surface of the silica coating layer.
Pancreatic insufficiency (PI) is commonly associated with diseases such as cystic fibrosis (the most common cause) or pancreatitis,28 in which the patients have a shortage of the digestive enzymes necessary to break down food. Besides, certain gastrointestinal diseases, such as stomach ulcers and Crohn’s disease, may also induce PI. However, all these diseases have their own clinical and genetic features in addition to the biochemical markers. For example, the trypsin concentration is always lower in cystic fibrosis,28 while it is higher in Crohn’s disease.29 Fecal chymotrypsin and fecal elastase 1 (FE1) are other biochemical markers used to assess exocrine PI.30 Both trypsin and chymotrypsin are exclusively of pancreatic origin, and their quantitative measurement can be used to indicate pancreatic function. FE1 is not affected by bacterial degradation (as trypsin or chymotrypsin) or pancreatic enzyme supplementation, thus being useful to determine the pancreatic function in CF. However, low FE1 levels can be found in children with short gut syndrome as well as in patients with the Shwachman–Diamond syndrome who otherwise have normal pancreatic function.31 Therefore, the trypsin test is a promising method to help detect CF in symptomatic newborns or infants, or anyone to determine whether a person is a CF carrier (i.e. has a CF gene mutation). Early diagnosis of cystic fibrosis has the potential to dramatically improve the life quality and survival of millions of people dealing with this chronic illness, of genetic origin, by developing early and personalised treatment, preventing malnutrition and allowing access to counseling. Many countries have established a default screening for newborns that involves quantifying trypsinogen in the blood of the babies (immunoreactive trypsinogen test, or IRT), which typically makes use of fluorescence-based immunoassays. However, these tests still have the drawbacks of immunoassays with respect to the specificity, time-effectiveness and userfriendliness. Additionally, due to a considerable percentage of false positives inherent to this test, babies with abnormal results must be either tested again for IRT or, alternatively, undergo genetic testing to determine whether they have at
Nanoscale
least one mutated copy of the cystic fibrosis gene (human carriers).32,33 Infants with a second abnormal newborn IRT screen or carrier genotype need a sweat test to confirm the cystic fibrosis diagnosis. Sweat testing however cannot be performed accurately in the first few days after birth, and it may be difficult to obtain an adequate sweat sample during the first two to three weeks after birth, especially for premature babies. If the volume of sweat obtained is inadequate, the test should be repeated. The overall procedure is long (weeks to months) and sometimes considered invasive by the parents.34 We present herein for the first time a well-designed and cost effective nanosystem, based on FRET, with an internal reference that unequivocally allows the ratiometric determination of trypsin in the range of enzyme concentrations relevant for clinical diagnosis of cystic fibrosis in minutes, with high sensitivity. Our method provides specific information about whether or not the individuals are CF carriers (heterozygotic genotype), and therefore about their risk of having a child affected with CF (homozygotic genotype). CF is the most life threatening recessive genetic condition affecting Caucasian children and diagnosis is often made at a point when significant and irreversible damage has already occurred. The challenge is to identify a detection method that will allow for early diagnosis and treatment with sufficient time to adopt lifestyle changes or use agents to prevent or retard disease development. A short peptide substrate previously marked with TAMRA, a rhodamine-derived (carboxy-tetramethyl-rhodamine) dye with fluorescence emission at λ = 575 nm (see Scheme 1), is anchored to the surface of a silica nanosphere containing two types of CdSe quantum dot nanocrystals with different fluorescence emission wavelengths.
Experimental Materials Trypsin (13 000–20 000 units mg−1), selenium powder (99.99%,), 1-octadecene (90%), tetraethyl orthosilicate (TEOS, 99.999%), 3-aminopropyl-trimethoxysilane (APTMS, 97%), (3-trihydroxysilyl) propyl methylphosphonate (THPMP), and phosphate buffer solution (PBS) were purchased from Sigma-Aldrich. Ammonia was purchased from Fluka. Synthesis of QD. The procedure to obtain CdSe QDs was using TOPO-TOP as a capping ligand.35 A selenium solution was prepared by mixing 0.4 g of selenium powder, 10 mL of TOP and 0.2 mL of anhydrous toluene and stirred under argon. A mixture of 20 g of TOPO and 0.25 g of cadmium acetate dihydrate was placed in a round-bottomed flask, stirred and heated to 150 °C under argon and the temperature was increased to 320 °C. When the mixture reached 320 °C the selenium solution was quickly added and then cooled to 270 °C. The reaction was run for a specific time (depending on the size of the QDs that we expected to achieve). After stopping the reaction, samples were cooled and washed with ethanol and acetone three times and stored in chloroform. Synthesis of QD@silica nanospheres. The experiments were performed using CdSe core quantum dots with emission wavelengths at 540 nm (green (2.7 nm)) and 660 nm (red (5.6 nm)). Next, single and two color quantum dot embedded silica nanospheres, i.e. QD540@silica and QD660@silica@QD540@silica, respectively, were prepared by the reverse microemulsion method as detailed elsewhere.36,37 Amine functionalization was done by adding 30 μL of APTMS and 80 μL of THPMP to the reaction and allowing to stir for 24 hours at room temperature. Peptide-functionalization of QD@silica nanospheres. The positive surface of the QD540@silica nanospheres (given by APTMS) was functionalized with two different TAMRA-labelled peptides: Pro-active: NH2-Cys-Lys-Arg-Val-Lys-TAMRA In-active: NH2-Cys-Lys-Pro-Val-Lys-TAMRA The two different sequences of TAMRA-labelled peptides were particularly designed for our study by Biotrend. Glutaraldehyde was used to bind the amine functional groups of the peptides with the amine groups of the silica nanosphere surface as described in the following methodology. 2 mL of QD@silica-APTMS in PBS and 0.25 mL of glutaraldehyde (0.25 mg mL−1) were mixed and shaken for 4 h at 37 °C. Then the solution was centrifuged to remove the unreacted glutaraldehyde and finally the pellet was resuspended in PBS. A determined amount of TAMRA-labelled peptides was added and shaken for 20 hours at 4 °C. After this time, the solution was centrifuged to remove the unbound peptides. In the end, the TAMRA-labelled peptide-functionalized QD@silica nanospheres were resuspended in 5 mL PBS and stored. Enzymatic digestion. A determined concentration of trypsin between 25 and 350 μg L−1 was added to the labelled peptidesnanospheres and mixed for a period of time: from 30 s to 120 min, at 37 °C. After this time, the mixture was heated to 70 °C and held for 1 h in order to denaturalize the enzyme and stop the reaction. If not measured immediately after the
This journal is © The Royal Society of Chemistry 2014
Paper
enzymatic digestion, it is recommended to keep the samples in a freezer until the analysis. Characterization techniques The formation of the QDs@silica nanospheres was directly visualized by transmission electron microscopy (TEM) using a JEOL 1011 microscope. The UV-Visible and fluorescence spectra were recorded using a 1 cm path length quartz cell on a Shimadzu UV Spectrophotometer 1700 and an AmincoBowman Series 2 luminescence spectrometer. Preparation of the biological samples Patients. Eligible patients included males and females from four families (11 subjects in total, A–K in Table S1†) with a confirmed diagnosis of cystic fibrosis. Four of the subjects were taking daily optimal doses of a pancreatic enzyme to increase the protein absorption. Written informed consent was obtained from all patients or their parents, and assent was also obtained from minors. Stool sample collection. Trypsin is firmly attached to the solid matter of feces and after the homogenization the supernatant shows very little enzymatic activity.38,39 Therefore, all the fecal specimens analyzed have to be sufficiently solid to be handled with a spatula. As the enzymes are very well embedded in the solid matter of the feces, we can correlate the measurements in concentration per gram with the concentration per liter in the duodenum.40 1 g of solid feces per participant, stored at −8 °C, was added to 3 mL of PBS ( pH 7.2) and vigorously stirred until a homogeneous solution was obtained. For the measurements, the mixture was heated up to 37 °C and 1 mL of TAMRA-labelled peptides-2nanoSi nanospheres in PBS was added. The reaction was run for 10 min, and then heated up to 70 °C to stop the enzymatic activity by denaturalization for 1 h. Finally, the samples were measured by fluorescence spectrometry. Statistical analysis. Each sample was measured five times and the mean values were reported with their standard deviation (SD) for each participant. The measurements were carried out in five consecutive days, with one set per day, each set of measurements consisting of 11 individual samples.
Results and discussion In brief, the 2nanoSi system consists of a 77 nm silica nanosphere with a nanocrystal CdSe (CdSe660) core with the luminescence emission maximum at λ = 660 nm, and a second shell of silica embedding the second type of CdSe nanocrystal with a luminescence emission maximum at λ = 540 nm (CdSe540). Upon light excitation at λ = 405 nm, both types of CdSe quantum dots undergo fluorescence emission. Moreover, due to the FRET process between the CdSe540@2nanoSi, as the energy donor, and the TAMRA dye, as the energy acceptor, it is possible to register TAMRA fluorescence emission at λ = 575 nm. Control experiments show negligible emission of
Nanoscale
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Paper
TAMRA upon excitation at λ = 405 nm (see Fig. S1†). In the presence of trypsin, the proteolytic activity of the enzyme leads to the cleavage of the peptide and the FRET process is disrupted. Thus, by measuring the signal increase at λ = 540 nm that corresponds to the CdSe540@2nanoSi, it is possible to correlate the changes in the emission intensity with the enzyme activity using an internal reference, in which fluorescence emission remains constant throughout the proteolytic process. In this case, the internal reference is the CdSe660@2nanoSi with fluorescence emission at λ = 660 nm. Hence, the reference CdSe660 material has the advantage of not being influenced by the TAMRA concentration and the excitation intensity as it occurs with commercial immuno-fluorescence intensity based measurements. FRET is observed when the “donor” and “acceptor” are sufficiently close to each other and when there is sufficient spectral overlap between the emission spectrum of the donor and the absorption spectrum of the acceptor. The average distance between TAMRA and CdSe660 is around 40 nm, while the photoluminescence emission spectrum of the label and the photoluminescence absorption spectrum of the first QD population (CdSe660) have a spectral overlap below 30% area/area, conditions which effectively preclude FRET between the label (as donor) and the CdSe660 (as the acceptor). The use of a single quantum dot fluorophore in our system (1nanoSi, Fig. S2 and S3†) already allowed for the qualitative analysis of the trypsin proteolytic activity. Nonetheless, a key figure of merit of our approach is to be able to quantify the enzymatic activity and, thus, the use of an internal reference in our fluorescent system. As mentioned above, we added the second population of QDs at the core of the silica nanospheres with fluorescence emission at λ = 660 nm (Fig. S4†). Thus, the use of two QD fluorophores with distinguishable emission spectra allows us to establish a measurable ratio, namely I540/I660. Moreover, the emission intensity at λ = 660 nm remains constant as the QD660 at the silica nanospheres does not influence the FRET process. The key feature in our 2nanoSi is that due to the absorption properties of the nanocrystal quantum dots, both fluorophores can be excited at the same wavelength (λ = 405 nm), but only the nanocrystals in the outer layer emitting at 540 nm undergo FRET with the TAMRA dye. Despite the small overlap between the emission of QD660 and the absorption of TAMRA, the distance between them (around 38 nm) does not allow energy transfer to occur. On the other hand, the distance and the overlap between QD540 and TAMRA are sufficient for FRET to occur. By TCSPC and steady-state emission spectroscopy, we determined a FRET efficiency of 47% between the QD540 and the TAMRA dye. The calculated R0 (the distance at which the efficiency of FRET is 50%) is 4.982 nm, and the distance between CdSe540 and TAMRA dye (r) is 5.10 nm calculated from the fluorescence measurements. Next, the number of TAMRA-labelled peptides attached to the silica nanosphere surface was found to be one via the Förster model.41 For the complete calculations, see the ESI.† This result clearly indicates that with only one QD@silica – peptide-dye donor–acceptor pair, our system displays a
Nanoscale
Nanoscale
remarkable FRET efficiency of approx. 50%. Previous studies with one pair of QD donor–dye acceptor reported lower efficiency, i.e. 22%.11 These authors required up to five acceptor molecules to increase the efficiency up to 58%. Our system is highly FRET efficient and the design applies to different types of quantum dot populations as long as the distance and spectral overlap between the emission spectrum of the donor and the absorption spectrum of the acceptor are particularly adequate. Fig. 1 shows the changes in the fluorescence emission of 2nanoSi in the presence of two different trypsin concentrations, 25 μg L−1 (Fig. 1a) versus 350 μg L−1 (Fig. 1b). As expected, increasing the enzyme concentration leads to faster recovery of the original emission at 540 nm (Fig. 1b). The complete enzyme concentration study can be found in Fig. S8.† The trypsin concentration for the different cystic fibrosis genotypes is undeniably tissue specific. Consequently, the
Fig. 1 2nanoSi emission spectra (a) after digestion with 25 μg L−1 of trypsin, and (b) after digestion with 350 μg L−1. The figure legends correspond to the enzymatic digestion time in min (0–30 min).
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Nanoscale
Fig. 2 Kinetics of the enzymatic digestion as a function of trypsin concentration using the ratio between PL emission peaks of both QD540 and QD660 of the 2nanoSi system.
trypsin concentration in fecal samples displays the following trend: CF-diagnosed 0–30 μg g−1, while in a non-CF medical check it is above 80 μg g−1.42 On the other hand, the distribution of the serum trypsin concentration in patients with cystic fibrosis43 is homozygotic (0–90 μg L−1), normal (160–340 μg L−1), and heterozygotic (91–349 μg L−1), respectively. Coincidentally, the trypsin ranges in stool samples highly match those in blood, i.e. non-CF > 80 μg g−1 stool versus heterozygote and healthy > 90 μg L−1 blood. Taking into account these values of trypsin, and considering the specificity of the 2nanoSi, we performed the study of the enzymatic digestion using the “pro-active” labelled peptide in a wide range of trypsin concentrations (25–350 μg L−1) that fall in the ranges found for CF patients mentioned above. Fig. 2 shows the ratio between the PL emission peaks of both QD540 and QD660 (I540/I660) during enzymatic digestion for 120 min in different enzyme concentrations. As expected, there is a direct relationship between the enzymatic digestion rate and the enzyme concentration that allowed us to unequivocally differentiate between the different enzyme concentrations used. In other words, the higher the trypsin concentration the shorter the time we need to reach a plateau in our kinetic measurements. Hence, by analysing the trypsin activity at early digestion times using our 2nanoSi ratiometric system (for example 10 min) we can differentiate between the different trypsin concentrations as shown in Fig. 3. As can be seen, the calibration curve proved to be highly linear (χ2 = 0.99) and allowed us to determine the trypsin concentration at levels that are clinically relevant for the cystic fibrosis prognosis, i.e. CF homozygotic (0–90 μg L−1) and nonCF (>90 μg L−1) including both CF heterozygotic and normal, respectively. The trypsin concentration in human faecal samples was determined based on the calibration curve plotted in the
This journal is © The Royal Society of Chemistry 2014
Paper
Fig. 3 Calibration curve obtained by fitting the fluorescence I540/I660 experimental values from Fig. 2 at the enzyme digestion time of 10 min. Red, orange and green correspond to CF homozygotic, CF heterozygotic and normal, respectively.
Fig. 4 Results of trypsin determination in human subject samples obtained with the proposed 2nanoSi nanospheres. The symbols correspond to participants in the study without specific medication (Δ), participants in the study under treatment with supplementary pancreatic enzymes (+), and participants in the study not taking the supplementary pancreatic enzymes (o).
coordinates: I540/I660 (−) versus trypsin concentration (μg L−1), following the equation: I 540 =I 660 ¼ 1:6811 þ 0:004625 ½trypsin
The results of trypsin determination performed with the proposed model fitted with the CF heterozygotic range (triangle (Δ) and cross (+) symbols in Fig. 4), indicating that all
Nanoscale
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Paper
the subjects are carrying at least one CF mutation. Nonetheless, taking into account that four of the participants (marked by (+) in Fig. 4) have a confirmed diagnosis of cystic fibrosis and are taking supplementary pancreatic enzymes (8–10 capsules of Creon® 25000 or 20 capsules of Creon® 10000 daily, respectively), the additional enzymatic activity (61–92%) has to be subtracted from the total trypsin concentration above. This observation is in good agreement with previous studies in the literature in which the CPA (coefficient of protein absorption) increased in the range 59–90% after supplementary pancreatic enzyme uptake.44,45 Accordingly, the values obtained for these four particular patients can now be found in the CF homozygotic range (symbol (o) in Fig. 4). (For the whole study see Fig. S9 and Tables S1 and S2 in ESI.†) The rest of the samples, provided by the parents of the four CF homozygotic patients, were considered as the control. Each parent of a patient with this condition is by definition a heterozygote, in other words he/ she carries only one copy of the CF gene but does not develop the disease.
Conclusions By combining the specificity of biomolecular interactions with the tunability of quantum dot and organic dye optical properties, we have developed for the first time a detection system capable of cystic fibrosis prognosis, both qualitatively and quantitatively. The trypsin enzymatic activity was reported via the FRET process, which mediates changes in the quantum dot emission spectra, between the organic dye TAMRA (λem = 575 nm) and green CdSe quantum dots (λem = 540 nm). Crossreactivity and signal overlap were avoided by designing enzyme-specific peptidic sequences. The 2nanoSi proved to be a fast and highly sensitive single-step nanosensor, allowing the quantification of trypsin concentrations in a wide range (25–350 μg L−1), which is clinically relevant for the diagnosis of cystic fibrosis (CF). This range covers different human genotypes, i.e. CF homozygotic (0–90 μg L−1) and CF heterozygotic (91–349 μg L−1). Using our 2nanoSi nanospheres in biological samples, four participants in the study were confirmed homozygotic, while the other seven were heterozygotic for CF. The biomarker has been shown to provide significant prognostic information, and hence represents a promising system for proof-of-concept demonstration of the proposed approach. We anticipate the 2nanoSi system to be a starting point for noninvasive, easy-to-use and cost effective ratiometric fluorescent biomarkers for recessive genetic diseases like human cystic fibrosis.
Acknowledgements We would like to thank ICIQ, ICREA, Fundació La Caixa, and the Spanish MICINN project CTQ2010-18859 for financial support. EP also thanks the EU for the ERCstg Polydot. We are
Nanoscale
Nanoscale
thankful to Núria Mir for facilitating the collection of the biological samples from the participants in the study.
Notes and references 1 N. Steuerwald, J. Cohen, R. J. Herrera and C. A. Brenner, Mol. Hum. Reprod., 1999, 11, 1034–1039. 2 A. W. Murray and M. N. Kirschner, Nature, 1989, 339, 575– 580. 3 A. Henning, H. Bakirci and W. M. Nau, Nat. Methods, 2007, 4, 629–632. 4 S. S. Antunovic, M. Lukac and D. Vujovic, Clin. Pharmacol. Ther., 2013, 93, 86–97. 5 O. Schilling and C. M. Overall, Nat. Biotechnol., 2008, 26, 685–694. 6 D. A. Thomas, P. Francis, C. Smith, S. Ratcliffe, N. J. Ede and C. Kay, Proteomics, 2006, 6, 2112–2120. 7 K. E. Sapsford, D. Farrell, S. Sun, A. Rasooly, H. Mattoussi and I. L. Medintz, Sens. Actuators, B, 2009, 139, 13–21. 8 M. Chalfie, Y. Tu, G. Euskirchen, W. W. Ward and D. C. Prasher, Science, 1994, 263, 802–805. 9 M. Chelfie and S. R. Kain, in Methods of Biochemical Analysis, ed. M. Chalfie and S. R. Kain, Wiley, New York, 2005, vol. 47. 10 W. R. Algar, A. Malonski, J. R. Deschamps, J. B. BlancoCanosa, K. Susumu, M. H. Stewart, B. J. Johnson, P. E. Dawson and I. L. Medintz, Nano Lett., 2012, 12, 3793– 3802. 11 I. L. Medintz, H. T. Uyeda, E. R. Goldman and H. Mattoussi, Nat. Mater., 2006, 4, 435–446. 12 Y. P. Ho and K. W. Leong, Nanoscale, 2010, 2, 60–68. 13 I. L. Medintz and H. Mattoussi, Phys. Chem. Chem. Phys., 2009, 11, 17–45. 14 H. Zhang, G. Feng, Y. Guo and D. Zhou, Nanoscale, 2013, 5, 10307–10315. 15 S. Chatterjee, J. B. Lee, N. V. Valappil, D. Luob and V. M. Menon, Nanoscale, 2012, 4, 1568–1571. 16 C. Y. Chen, C. T. Cheng, C. W. Lai, P. W. Wu, K. C. Wu, P. T. Chou, Y. H. Chou and H. T. Chiu, Chem. Commun., 2006, 263–265. 17 G. B. Kim and Y. P. Kim, Theranostics, 2012, 2, 127–138. 18 I. L. Medintz, A. R. Clapp, H. Mattoussi, E. R. Goldman, B. Fisher and J. M. Mauro, Nat. Mater., 2003, 2, 630–638. 19 H. H. Chen, Y. P. Ho, X. Jiang, H. Q. Mao, T. H. Wang and K. W. Leong, Nano Today, 2009, 4, 125–134. 20 R. Alam, J. Zylstra, D. M. Fontaine, B. R. Branchinib and M. M. Maye, Nanoscale, 2013, 5, 5303–5306. 21 H. Zhang and D. Zhou, Chem. Commun., 2012, 48, 5097– 5099. 22 K. Ouadahi, K. Sbargoud, E. Allarda and C. Larpent, Nanoscale, 2012, 4, 727–732. 23 W. R. Algar, A. P. Malanoski, K. Susumu, M. H. Stewart, N. Hildebrandt and I. L. Medintz, Anal. Chem., 2012, 84, 10136–10146.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 12 September 2014. Downloaded by KUNGL TEKNISKA HOGSKOLAN on 06/10/2014 10:13:07.
Nanoscale
24 S. B. Lowe, J. A. G. Dick, B. E. Cohen and M. M. Stevens, ACS Nano, 2012, 6, 851–857. 25 P. L. Xavier, K. Chaudhari, P. K. Verma, S. K. Pal and T. Pradeep, Nanoscale, 2010, 2, 2769–2776. 26 G. Stoica, I. Castelló Serrano, A. Figuerola, I. Ugarte, R. Pacios and E. Palomares, Nanoscale, 2012, 4, 5409–5419. 27 A. F. Heeley and D. Watson, Clin. Chem., 1983, 29, 2011– 2018. 28 J. F. Pohl and D. J. Easley, Pract. Gastroenterol., 2003, 38–48. 29 T. Midtvedt, E. Zabarovsky, E. Norin, J. Bark, R. Gizatullin, V. Kashuba, O. Ljungqvist, V. Zabarovska, R. Möllby and I. Ernberg, PLoS One, 2013, 8, 1–9. 30 J. Leus, S. Van Biervliet and E. Robberecht, Eur. J. Pediatr., 2000, 159, 563–568. 31 S. Bebarry, L. Ellis, M. Corey, M. Marcon and P. Durie, J. Pediatr., 2002, 141, 84–90. 32 DELFIA Neonatal IRT assay, PerkinElmer Inc., 2009. 33 B. A. Lindau-Shepard and K. A. Pass, Clin. Chem., 2010, 56, 445–450. 34 A. Mishra, R. Greaves and J. Massie, Clin. Biochem. Rev., 2005, 26, 135–153. 35 J. Aldana, Y. A. Wang and X. G. Peng, J. Am. Chem. Soc., 2001, 112, 8844–8850.
This journal is © The Royal Society of Chemistry 2014
Paper
36 I. Castelló Serrano, Q. Ma and E. Palomares, J. Mater. Chem., 2011, 21, 17673–17679. 37 Q. Ma, I. Castelló Serrano and E. Palomares, Chem. Commun., 2011, 47, 7071–7073. 38 B. J. Haverback, B. J. Dyce, P. J. Gutentag and D. W. Montgomery, Gastroenterology, 1963, 44, 588–597. 39 W. P. Dick, Am. J. Dig. Dis., 1967, 12, 310–317. 40 J. S. Smith, I. Ediss, M. A. Mullinger and A. Bogoch, Can. Med. Assoc. J., 1971, 104, 691–694. 41 A. R. Clapp, I. L. Medintz, M. Mauro, B. R. Fisher, M. G. Bawendi and H. Mattoussi, J. Am. Chem. Soc., 2004, 126, 301–310. 42 G. J. Barbero, M. S. Sibling, J. M. Marino and R. Seibel, Am. J. Dis. Child, 1966, 112, 536–540. 43 P. Dandona, M. E. Hodson and J. C. Batten, J. Clin. Pathol., 1983, 36, 760–792. 44 M. M. Perez, A. D. Newcomer, C. G. Moertel, C. L. W. Go and E. P. Dimagno, Cancer, 1983, 52, 346–352. 45 M. W. Konstan, R. C. Stern, J. R. Trout, J. M. Sherman, H. Eigen, J. S. Wagener, C. Duggan, M. E. B. Wohl and P. Colin, Aliment. Pharmacol. Ther., 2004, 20, 1365– 1371.
Nanoscale
