Letter Cite This: Anal. Chem. XXXX, XXX, XXX−XXX
pubs.acs.org/ac
Recombinant Fusion Streptavidin as a Scaffold for DNA Nanotetrads for Nucleic Acid Delivery and Telomerase Activity Imaging in Living Cells Zhi-Mei Huang, Mei-Ya Lin, Chong-Hua Zhang, Zhenkun Wu,* Ru-Qin Yu, and Jian-Hui Jiang* State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha, Hunan 410082, P. R. China Downloaded via 79.110.17.36 on July 21, 2019 at 05:36:20 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: Efficient platforms for intracellular delivery of nucleic acids are essential for biomedical imaging and gene regulation. We develop a recombinant fusion streptavidin as a novel protein scaffold for DNA nanotetrads for highly efficient nucleic acid delivery and telomerase activity imaging in living cells via cross-linking hybridization chain reaction (cHCR). The recombinant streptavidin protein is designed to fuse with multiple SV40 NLS (nuclear localization signal) and NES (nuclear export signal) domains and prepared through Escherichia coli expression. The recombinant NLS-SA protein allows facile assembly with four biotinylated DNA probes via high-affinity noncovalent interactions, forming a well-defined DNA tetrad nanostructure. The DNA nanotetrads are demonstrated to confer efficient cytosolic delivery of nucleic acid via a caveolar mediated endocytosis pathway, allowing efficient escape from lysosomal degradation. Moreover, the nanotetrads enable efficient cHCR assembly in response to telomerase in vitro and in cellulo, affording ultrasensitive detection and spatially resolved imaging for telomerase with a detection limit as low as 90 HeLa cells/mL. The fluorescence brightness obtained in live cell imaging is found to be dynamically correlated to telomerase activity and the inhibitor concentrations. Therefore, the proposed strategy may provide a highly efficient platform for nucleic acid delivery and imaging of biomarkers in living cells.
T
binding capacity. Two strategies have been developed for constructing protein-based DNA nanostructures. One utilizes the thiol or amine groups on the protein surface to functionalize DNA via covalent linkages, such as proteinbased spherical nucleic acids.14−16 This approach requires relatively complicated chemical conjugation steps, leading to low yield in the synthesis of protein−DNA nanostructures. The number and position of DNA on its surface are also difficult to be precisely controlled. An alternative strategy employs high-affinity noncovalent bindings for synthesizing DNA nanostructures.17 Streptavidin and its variants provide a facile and robust scaffold to construct protein−DNA nanostructures.18,19 These proteins have precisely engineered interacting sites on the surface for noncovalent binding to biotin with a dissociation constant on the order of 10−15 M.20 The SA scaffolded DNA nanostructures show important properties such as ease in synthesis, well-defined structure, and cell membrane permeability. However, their interactions with the cell membrane are mainly contributed by the surface
elomerase, a ribonucleoprotein reverse transcriptase that adds TTAGGG repeats to telomeres to protect the chromosomes, plays a critical role in uncontrollable proliferation and immortalization of cells.1 Overexpressed telomerase has close association with tumor progression and is detected in >90% of human cancer cells, indicating that telomerase is a useful biomarker for cancer.2 Approaches to monitoring telomerase activity in cells represent critical tools for tumor biology, cancer diagnostics, and drug discovery.3,4 Nucleic acid probe-based strategies afford useful platforms for telomerase imaging in living cells, including cationic liposome−DNA complexes,5 mesoporous silica nanoparticle−DNA probes,6 spherical nucleic acids,7−11 graphene oxide−DNA probes,12 and DNA polymer probes.13 Despite the important advances, the methods still suffer from issues such as dose-dependent cytotoxicity, complicated probe preparation, inferior sensitivity, or limited delivery efficiency due to endo/lysosomal entrapment. Robust and efficient approaches for cytosolic nucleic acid delivery remain highly demanding for cellular imaging and theranostics. Protein scaffolded DNA nanostructures have emerged as a promising platform for delivery of functional proteins and nucleic acids due to their good biocompatibility and high © XXXX American Chemical Society
Received: May 5, 2019 Accepted: July 4, 2019 Published: July 5, 2019 A
DOI: 10.1021/acs.analchem.9b02115 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
substrate for telomerase, and a pair of hairpin probes H1 and H2 are engineered for the detection of the primer extensions generated by telomerase. Three rSANTs, rSANT-TP, rSANTH1, and rSANT-H2, are separately constructed using the primer probe TP and the hairpin probes H1 and H2, respectively. When delivered into the cytosol independently by the three DNA nanotetrads, the primer TP is extended by telomerase in the cytosol, which then triggers HCR between hairpin probes H1 and H2. The alternating assembly of H1 and H2 in different DNA nanotetrads induce cross-linking of the reactant nanotetrads and form DNA nanoassembly products, with a highly localized and amplified fluorescence signal activated for telomerase activity imaging. To our knowledge, this is the first time that a recombinant fusion SA protein scaffolded DNA nanotetrads has been developed for efficient delivery of nucleic acid into living cells. Because of the superb efficiency of the delivery system and the highly amplified fluorescence signals, the developed strategy could provide a valuable platform for nucleic acid delivery and low abundant biomarker imaging. To obtain the recombinant fusion SA, a plasmid for rSA was constructed (Figure S1) and confirmed by sequencing (Figure S2). The recombinant fusion SA was then purified following overexpression in Escherichia coli (Figures 1A, S3, and S4). A
DNA probes, and high-affinity motifs to the cell membrane have not been introduced in the protein−DNA complex. Hence, this protein scaffolded DNA nanostructure still has limitation in nucleic acid delivery efficiency. We reason that a bioengineered fusion of SA with cellpenetrating peptides can afford improved nucleic acid delivery efficiency by introducing the high-affinity motifs to the cell membrane. Motivated by this hypothesis, we design a recombinant fusion protein of SA (rSA) with SV40 NLS (nuclear localization signal) as a potential scaffold to develop a novel DNA nanostructure for efficient delivery of DNA into living cells, as illustrated in Scheme 1A. In this design, the rSA Scheme 1. (A) Illustration of rSANTs and (B) rSANTs Mediated cHCR for Telomerase Activity Imaging in Living Cells
is engineered by fusion with two SV40 NLS domains followed by a NES (nuclear export signal) domain in each monomer of the tetramer protein. The NLS domains improve the interactions of the bioengineered protein with cell membrane and facilitate its delivery into cells. The NES domains help the cytosolic localization of the recombinant protein. Through secretion expression with Escherichia coli to obtain the recombinant protein, a DNA tetrad structure is synthesized by assembly of four biotinylated DNA probes on the protein surface through high-affinity interactions. Because of the added affinity of the NLS domains with the cell membrane, the rSADNA nanotetrad (rSANT) is demonstrated to allow efficient cytosolic delivery of nucleic acids into living cells via a caveolar mediated endocytosis pathway, allowing effective escape from lysosomal degradation. Moreover, its properties of noncovalent, high-affinity assembly, biomaterial composition, and precisely controlled binding sites make it a robust delivery system that enables facile synthesis, well-defined molecular structure, low cytotoxicity, and effective cytosolic nucleic acid delivery. On the basis of this efficient delivery system and the previous work on nonenzymatic DNA cascade circuits from others and our group,21−23 we further develop a rSANT mediated cross-linking hybridization chain reaction circuit for telomerase activity imaging in living cells, as illustrated in Scheme 1B. In the design of cross-linking hybridization chain reaction (cHCR), a primer probe TP is designed as the
Figure 1. Characterization of recombinant fusion SA and rSANTs. (A) SDS-PAGE gel image of commercial SA (lane 1) and recombinant fusion SA (lane 2). (B) Agarose gel image of DNA probes (lane 1) and rSANTs (lane 2). (C) Mean fluorescence intensity measured from confocal images for cellular internalization of SANTs and rSANTs. (D) Flow cytometry assay of HeLa cells with rSANTs incubation at different conditions.
bright bond with molecular weight of ∼20 kDa was observed, indicating the successful expression of rSA. The interactions between rSA and biotinylated DNA probes were studied using agarose gel (Figure 1B). A single new band with much larger size was displayed, demonstrating the successful formation of rSANTs. The cell permeability of rSANTs was then studied. rSANTs assembled with Cy5-labeled hairpin probes were separately incubated with four different types of cells, and the cell uptake efficiency was further compared with the standard SA-DNA nanotetrads (SANTs) under the same conditions. HeLa cells incubated with rSANTs displayed bright fluorescence signals, which were several times higher (∼5.1-fold for MCF-7, ∼4.0B
DOI: 10.1021/acs.analchem.9b02115 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry fold for HepG2, ∼4.5-fold for HeLa, and ∼6.2-fold for L02) than those treated with SANTs (Figures 1C and S5), indicating the enhanced delivery efficiency of rSANTs. Interesting, almost no fluorescence signals were observed in the nucleus of cells treated with rSANTs. We speculated that the explanation could be the strong electrostatic interactions between positively charged SV40 NLS peptides and negatively charged DNA could mask the recognition of SV40 NLS to importin protein. This hypothesis was supported by a previous study in which covalently linked NLS to linear DNA did not result in nuclear accumulation.24 The cellular uptake dynamic of rSANTs was explored using flow cytometry (Figure S6). With the addition of rSANTs into cells, the fluorescence was gradually increased and saturated at 60 min, suggesting the rapid and efficient cellular uptake of rSANTs. The delivery efficiency of DNA probes with various sizes as well as RNA probes was investigated (Figure S7). All three rSANTs with different sizes of DNA displayed an obvious fluorescence peak redshift, and with an increase in the size of DNA, the uptake efficiency slightly decreased. A possible explanation could be that the SV40 could neutralize a greater proportion of the negative charges with a smaller size of the DNA, making it easier to enter into cells. A huge fluorescence signal increase was observed for rSANTs carrying RNA probes, suggesting the ability of rSANTs for RNA delivery. We also compared the cellular uptake of rSANTs with commercial transfection agents Lipofectamine 3000 (Figure S8). A similar efficiency was observed from rSANTs, suggesting the potential of rSANTs to be a useful platform for nucleic acid delivery. The cell uptake mechanism of rSANTs was further investigated. As shown in Figure 1D, HeLa cells pretreated at 4 °C showed an obvious fluorescence signal peak blueshift, suggesting rSANTs were involved in a temperature-dependent cell uptake process. When HeLa cells were pretreated with NaN3, an inhibitor for ATPase, a huge fluorescence peak blueshift was also observed, confirming the energy-dependent endocytosis pathway for rSANTs. To explore the cellular internalization pathways of rSANTs, we pretreated HeLa cells with four kinds of inhibitors:25 methyl-β-cyclodextrin (M-βCD, lipid-raft mediated endocytosis inhibitor), chlorpromazine (CPZ, clathrin mediated endocytosis inhibitor), wortmannin (macropinocytosis mediated endocytosis inhibitor), and nystatin (caveolar mediated endocytosis inhibitor). It was found that HeLa cells pretreated with nystatin showed a concentration-dependent fluorescence peak blueshift; meanwhile, M-β-CD, CPZ, and wortmannin had almost no effect on intracellular uptake efficiency of rSANTs (Figures 1D and S9). These studies indicated that the cellular internalization rSANTs occurred via a caveolar mediated endocytosis pathway. The subcellular localization of rSANTs was explored. The fluorescence of Cy5 showed typical cytosolic localization and did not colocalize well with Lysotracker (Figure S10). This result was consistent with the previous report that caveosomes did not fuse with lysosomes,26 suggesting the rSANTs are a promising tool for cytosolic nucleic acid delivery. Subsequently, the biocompatibility of rSANTs was studied. HeLa cells were incubated with various concentrations of rSANTs, followed by the MTS assay. As shown in Figure S11, HeLa cells treated with rSANTs retained good viability, indicating the low cytotoxicity of rSANTs.
Having demonstrated the efficient nucleic acid delivery of rSANTs with low cytotoxicity, we then explored the performance of rSANTs mediated cHCR for telomerase activity imaging in living cells. The feasibility of cHCR for telomerase activity detection was first investigated in vitro (Figure 2A).
Figure 2. (A) Fluorescence signal of cHCR for telomerase assay: (a) response of rSANT-TP + rSANT-H1 + rSANT-H2; (b) response to heat-treated HeLa cell lysate; (c) response to QSG-7701 cell lysate; (d) response to 2 mM AZT-pretreated HeLa cell lysate; (e) response to HeLa cell lysate. (B) Fluorescence spectral responses to varying telomerase activities in HeLa cell lysates.
After incubating the mixture of rSANT-TP, rSANT-H1, and rSANT-H2 for 4 h, only a very weak fluorescence peak was observed, suggesting that cHCR did not occur in the absence of telomerase. In contrast, incubation of the mixture of rSANTs with the HeLa cell lysate gave intense fluorescence, indicating the initiation of cHCR triggered by the HeLa cell lysate. Control experiments by incubating the mixture of rSANTs with the heat-treated HeLa cell lysate or the QSG7701 cell lysate both showed negligibly increased signals. In addition, the real-time fluorescence assay showed similar results (Figure S12). These results validated that the cHCR specifically responded to the active telomerase, and coexisting cellular components had little interference with the assay. Another control experiment using the HeLa cell lysate, which was pretreated with 3′-azido-3′-deoxythymidine (AZT), a telomerase inhibitor,27 yielded substantially decreased fluorescence, confirming the specificity of cHCR to telomerase activity. The agarose gel assay was also performed to confirm the results (Figure S13). To confirm the formation of DNA assembly product of cHCR, atomic force microscopy (AFM) was used to directly visualize the product (Figure S14). Clumps of the cHCR products with a size of >1 μm were observed, clearly evidencing the formation of DNA assembly in the cHCR assay. The fluorescence responses of the cHCR strategy for the telomerase activity assay were then examined (Figure 2B). The fluorescence signals showed dynamic responses to telomerase activities in HeLa cell lysates ranging from 100 to 50 000 cells/ mL (Figure S15). A linear correlation was obtained for the fluorescence intensity to cell numbers ranging from 100 to 2000 cells. According to the 3σ/slope rule, a detection limit of 90 cells/mL was calculated. This detection limit was much better than existing telomerase activity imaging methods,8,12 indicating the high sensitivity of the proposed strategy. We next applied the cHCR for telomerase activity imaging in living cells. After incubating HeLa cells with rSANT-TP, rSANT-H1, and rSANT-H2 for 4 h, bright fluorescent spots appeared (Figure 3a). A control experiment by incubating the HeLa cells with rSANT-H1 and rSANT-H2 showed almost no fluorescence signal (Figure 3b), suggesting a specific response of cHCR to telomerase. In another control experiment using a C
DOI: 10.1021/acs.analchem.9b02115 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
Figure 3. Confocal microscopy images of rSANTs mediated cHCR for telomerase activity imaging. (a) rSANT-TP + rSANT-H1 + rSANT-H2; (b) rSANT-H1 + rSANT-H2; (c) control rSANT-TP + rSANT-H1 + rSANT-H2. (d) HeLa cells pretreated with AZT followed by incubation of rSANT-HP + rSANT-H1 + rSANT-H2.
control rSANT-TP, which could not be recognized by telomerase, no appreciable fluorescence was observed in the cells (Figure 3c), verifying the extension of the primer was essential for fluorescence signal generation. When HeLa cells were pretreated with AZT followed by incubation with rSANT-TP, rSANT-H1, and rSANT-H2 for 4 h (Figure 3d), fluorescence signals were rarely observed in cells. These results clearly evidenced that the cHCR assay had high specificity to telomerase. Time-dependent imaging showed that the fluorescent spots appeared in cells within 60 min and were saturated at 4 h (Figure S16). The Z-stack assay confirmed that the fluorescence signals were inside HeLa cells (Figure S17). The cHCR strategy was further used for monitoring the telomerase activity in living cells. HeLa cells were pretreated with various concentrations of AZT. Then, the AZT-treated cells were divided into two portions: one lysed for the in vitro fluorescence assay and the other, for imaging. The in vitro assay showed decreased fluorescence signals with increasing concentrations of AZT (Figure S18). The confocal imaging assay revealed that, as the concentration of AZT increased, the fluorescent signals inside the cells gradually decreased (Figure S19). These results demonstrated that the cHCR could be used for monitoring the telomerase activity changes in live cells, proving a potential platform for live cell screening of telomerase inhibitors. To investigate the quantitative ability of cHCR for telomerase activity, four cell lines, MCF-7, HepG2, HeLa, and QSG-7701 cells, were tested. As shown in Figure 4, cancer cells (MCF-7, HepG2, and HeLa) showed higher fluorescence signals than normal cells (QSG-7701), and different types of cancer cells also displayed different fluorescence intensities (Figure 4A,B). These results were consistent with previous reports for the telomerase activity assay.8,11 Moreover, the enzyme-linked immunosorbent assay (ELISA) was performed to further verify the results from confocal images (Figure 4C). Thus, these results demonstrated the ability of our cHCR strategy for quantitative telomerase activity imaging in living cells and distinguishing cancer cells from normal cells. In conclusion, we have developed a recombinant fusion SA as a novel scaffold for DNA nanotetrads for highly efficient nucleic acid delivery and telomerase activity imaging in living cells through cHCR. The proposed rSANTs showed important properties, including facile synthesis, good biocompatibility, and well-defined nanostructure. Moreover, the rSANTs displayed high efficiency in cytosolic nucleic acid delivery via a caveolar mediated endocytosis pathway. Benefiting from the
Figure 4. (A) Confocal microscopy images of (a) HeLa, (b) MCF-7, (c) HepG2, and (d) QSG-7701 cells incubated with rSANT-HP + rSANT-H1 + rSANT-H2. Telomerase activities measured by cHCR imaging (B) and the ELISA assay (C).
above unique properties, the developed rSANTs have been demonstrated to enable efficient cHCR assembly in response to telomerase, affording sensitive and spatially resolved telomerase activity imaging in living cells. In virtue of the advantages, the developed strategy may provide a promising platform for nucleic acid delivery and biomedical imaging.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.9b02115. Chemicals and instruments, plasmid DNA sequencing, SDA-PAGE and agarose gel electrophoresis assay, fluorescence spectra assay, MTS assay, confocal microscopy images, flow cytometry assays, subcellular location of and cellular toxicity assay for rSANTs, realtime fluorescence response, agarose gel imaging, AFM images, and Z-stack images (PDF)
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. *E-mail:
[email protected]. ORCID
Zhenkun Wu: 0000-0001-6458-5659 Ru-Qin Yu: 0000-0002-7412-8360 Jian-Hui Jiang: 0000-0003-1594-4023 Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS This work was supported by NSFC (21527810, 21521063). REFERENCES
(1) Hanahan, D.; Weinberg, R. A. Cell 2011, 144, 646−674. (2) Shay, J. W.; Bacchetti, S. Eur. J. Cancer 1997, 33, 787−791. (3) Harley, C. B. Nat. Rev. Cancer 2008, 8, 167−179.
D
DOI: 10.1021/acs.analchem.9b02115 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry (4) Blasco, M. A. Nat. Rev. Genet. 2005, 6, 611−622. (5) (a) Zhuang, Y.; Huang, F.; Xu, Q.; Zhang, M.; Lou, X.; Xia, F. Anal. Chem. 2016, 88, 3289−3294. (b) Yan, L. W.; Hui, J. J.; Liu, Y. R.; Guo, Y. H.; Liu, L.; Ding, L.; Ju, H. X. Biosens. Bioelectron. 2016, 86, 1017−1023. (6) Qian, R. C.; Ding, L.; Ju, H. X. J. Am. Chem. Soc. 2013, 135, 13282−13285. (7) Liu, Z. J.; Zhao, J.; Zhang, R. L.; Han, G. M.; Zhang, C.; Liu, B. H.; Zhang, Z. P.; Han, M. Y.; Gao, X. H. ACS Nano 2018, 12, 3629− 3637. (8) Qian, R. C.; Ding, L.; Yan, L. W.; Lin, M. F.; Ju, H. X. J. Am. Chem. Soc. 2014, 136, 8205−8208. (9) Xue, J.; Chen, F.; Bai, M.; Cao, X. W.; Huang, P.; Zhao, Y. X. Anal. Chem. 2019, 91, 4696−4701. (10) Ma, Y.; Wang, Z. H.; Zhang, M.; Han, Z. H.; Chen, D.; Zhu, Q. Y.; Gao, W. D.; Qian, Z. Y.; Gu, Y. Q. Angew. Chem., Int. Ed. 2016, 55, 3304−3308. (11) Yang, X. J.; Zhang, K.; Zhang, T. T.; Xu, J. J.; Chen, H. Y. Anal. Chem. 2017, 89, 4216−4222. (12) Hong, M.; Xu, L. D.; Xue, Q. W.; Li, L.; Tang, B. Anal. Chem. 2016, 88, 12177−12182. (13) Zhu, X. L.; Ye, H. Y.; Liu, J. W.; Yu, R. Q.; Jiang, J. H. Anal. Chem. 2018, 90, 13188−13192. (14) Brodin, J. D.; Auyeung, E.; Mirkin, C. A. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 4564−4569. (15) Brodin, J. D.; Sprangers, A. J.; McMillan, J. R.; Mirkin, C. A. J. Am. Chem. Soc. 2015, 137, 14838−14841. (16) Cigler, P.; Lytton-Jean, A. K.; Anderson, D. G.; Finn, M. G.; Park, S. Y. Nat. Mater. 2010, 9, 918−922. (17) (a) Zhang, C.; Tian, C.; Guo, F.; Liu, Z.; Jiang, W.; Mao, C. D. Angew. Chem., Int. Ed. 2012, 51, 3382−3385. (b) Lacroix, A.; Edwardson, T. G.; Hancock, M. A.; Dore, M. D.; Sleiman, H. F. J. Am. Chem. Soc. 2017, 139, 7355−7362. (c) Zhu, Q.; Chai, Y. Q.; Zhuo, Y.; Yuan, R. Biosens. Bioelectron. 2015, 68, 42−48. (d) Liu, G.; Wan, Y.; Gau, V.; Zhang, J.; Wang, L. H.; Song, S. P.; Fan, C. H. J. Am. Chem. Soc. 2008, 130, 6820−6825. (18) Huang, D. J.; Huang, Z. M.; Xiao, H. Y.; Wu, Z. K.; Tang, L. J.; Jiang, J. H. Chem. Sci. 2018, 9, 4892−4897. (19) Yan, H.; Park, S. H.; Finkelstein, G.; Reif, J. H.; LaBean, T. H. Science 2003, 301, 1882−1884. (20) Niemeyer, C. M. Trends Biotechnol. 2002, 20, 395−401. (21) Li, J.; Liu, S.; Sun, L.; Li, W.; Zhang, S. Y.; Yang, S.; Li, J.; Yang, H. H. J. Am. Chem. Soc. 2018, 140, 16589−16595. (22) Zhang, H. Q.; Li, F.; Dever, B.; Li, X. F.; Le, X. C. Chem. Rev. 2013, 113, 2812−2841. (23) Wu, Z.; Fan, H.; Satyavolu, N. S. R.; Wang, W.; Lake, R.; Jiang, J. H.; Lu, Y. Angew. Chem. 2017, 129, 8847−8851. (24) Van der Aa, M. A. E. M.; Koning, G. A.; d’Oliveira, C.; Oosting, R. S.; Wilschut, K. J.; Hennink, W. E.; Crommelin, D. J. A. J. Gene Med. 2005, 7, 208−217. (25) (a) Mout, R.; Ray, M.; Tonga, Y. G.; Lee, Y. W.; Tay, T.; Sasaki, K.; Rotello, V. M. ACS Nano 2017, 11, 2452−2458. (b) Chen, T. T.; Yi, J. T.; Zhao, Y. Y.; Chu, X. J. Am. Chem. Soc. 2018, 140, 9912−9920. (26) Dalal, C.; Saha, A.; Jana, N. R. J. Phys. Chem. C 2016, 120, 6778−6786. (27) White, L. K.; Wright, W. E.; Shay, J. W. Trends Biotechnol. 2001, 19, 114−120.
E
DOI: 10.1021/acs.analchem.9b02115 Anal. Chem. XXXX, XXX, XXX−XXX
