Enzymatically Amplified Mass Tags for Tissue Mass Spectrometry

Article pubs.acs.org/ac

Enzymatically Amplified Mass Tags for Tissue Mass Spectrometry Imaging Rui Hong, Jan True, and Christopher Bieniarz* Ventana Medical Systems, Inc., Technology and Applied Research, 1910 E. Innovation Park Drive, Oro Valley, Arizona 85755, United States S Supporting Information *

ABSTRACT: Tissue mass spectrometry imaging (MSI) is a rapidly developing technology which promises to provide biomarker molecular information within tissue context, which is an unmet medical need in the era of personalized medicine. However, challenges associated with tissue specimens as well as the MSI technical limitations have hindered the practical applications of this technology. We report here a mass tag based MSI method that combines the strength of signal amplification by immuno-enzymatic reactions with the superior detection characteristics of mass spectrometry to enable matrix-free MSI of protein biomarkers in formalin fixed paraffin embedded (FFPE) tissues. The method involves binding of the target protein with a primary antibody with high affinity and specificity, followed by binding with a secondary antibody−enzyme conjugate. Enzyme substrates suitable for mass spectrometry detection are locally deposited at the site of the target through enzymatically catalyzed transformation. The precipitates thus serve as mass tags to be detected in mass spectrometry to represent the target biomolecule in tissue. Two enzymes and various substrates have been identified and successfully used to demonstrate the feasibility of this novel MSI method to image protein targets in FFPE tissue samples.

W

methods with optical output/signal in terms of multiplexing and quantification capabilities. Modern mass spectrometers are capable of providing very high mass resolution which makes it possible to detect hundreds of molecules with discrete mass over charge (m/z). Various mass spectrometry based methods have been developed and widely applied in proteomic research. 5 One of the emerging techniques in mass spectrometry for tissue analysis is matrix-assisted laser desorption ionization-mass spectrometry imaging (MALDIMSI).6−8 The method is promising in providing molecular information of proteins,9 peptides,10 drugs,11 lipids,12 and other metabolites,13 while preserving tissue contextual information.14,15 However, for detection of proteins in tissue samples, especially FFPE tissues, several limitations originating from the nature of tissue samples as well as the MALDI-MSI process itself still persist and impede the wider biomedical and clinical applications of MSI.16 The first challenge is that the size of the molecules detectable by MALDI is rather limited. Usually, a protein with molecular weight greater than 50 kDa is unlikely to be detected, especially in tissue.17,18 However, many proven cancer biomarkers and drug targets exhibit much higher molecular weight than 50 kDa. For example, HER2 is a 185 KDa protein and thus undetectable by MALDI. The rather low natural abundance of many cancer biomarkers also makes them beyond the detection limit of MALDI. The detectability of

ith the growing understanding of cancer complexity and heterogeneity, the need for detection of multiple protein biomarkers on the same specimen has become even more urgent.1,2 Cancer tissue samples are extremely complex on morphological, genomic, and proteomic levels. Tools to provide molecular information associated with cancer are thus critical to both diagnosis and targeted treatment of cancers. The ability to obtain multiplexed and quantitative protein molecular information within the tissue context is one of the most challenging unmet medical needs to make personalized medicine a reality.3 Of particular importance is the protein detection in formalin fixed paraffin embedded (FFPE) tissue samples, because the majority of tissue specimens are preserved by this method. Immunohistochemistry (IHC) has been the standard technique for protein detection in tissue samples. The process of IHC usually involves the binding of target protein with a cognate antibody directly or indirectly conjugated to an enzyme, such as horseradish peroxidase (HRP) or alkaline phosphatase (AP). A colored substance is then deposited at the site of the target indicating the presence of the protein of interest.4 Even though IHC is a rather mature method, its capability in providing critical diagnostic and prognostic information has been even more valuable in the era of personalized medicine. For example, IHC of human epidermal growth factor receptor 2 (HER2) has been used to guide targeted therapeutics, such as humanized monoclonal antibody trastuzumab (Herceptin) in breast cancer treatment. As an emerging technology, mass spectrometry imaging may provide an alternative method to detect proteins in tissues. Mass spectrometry offers several potential advantages over © 2014 American Chemical Society

Received: August 26, 2013 Accepted: January 13, 2014 Published: January 13, 2014 1459

dx.doi.org/10.1021/ac402718f | Anal. Chem. 2014, 86, 1459−1467

Analytical Chemistry

Article

biomarkers that do not exhibit certain enzymatic activity. In addition, multistep chemical synthesis was required to generate the dendritic mass tag complex, which may limit its practical applications. In view of the challenges in protein detection in tissues, especially in FFPE tissues by MSI, and the limitations of the previously reported methods, we have developed a mass tag based MSI method that allows for specific detection of virtually any protein biomarkers in situ as long as cognate antibody is available. The method features significant signal amplification and thus leads to successful matrix-free LDI-MSI in FFPE tissue samples. A broad range of proteins can be detected regardless of their size, function, or ionization efficiency. The use of an antibody to specifically label the marker of interest also circumvents the need for protein digestion in cross-linked FFPE tissue samples. As shown in Scheme 1, the key aspect of the method is to use the amplification capability of enzymes routinely used in

certain protein targets can be further complicated due to variance in their ionization efficiency in the MALDI process. Taking these factors together, only the most abundant and relatively small proteins are detectable by MALDI-MSI, with the signal intensity not necessarily correlated to the protein abundance. In addition, the spatial resolution of MALDI-MSI is profoundly affected by a combination of the matrix application, target abundance, and instrument sensitivity. To achieve high spatial resolution, ideally to single cell level (about 10 μm), matrix crystals must be homogeneous19 and in the same range as a single cell. Otherwise, the signal will diffuse to the size of the crystals, and spatial information within the matrix crystal would be lost, which would abrogate spatial resolution gain by reducing laser focus diameter. Although significant improvements have been made in matrix application protocols,20,21 a matrix-free imaging method would facilitate the development of high resolution MSI. Furthermore, an additional challenge associated with the higher spatial resolution is the drastically reduced sampling area, which makes detection of low abundant targets even more difficult. Due to these obstacles, MALDIMSI has been a method more suitable for tissue profiling and biomarker discovery than targeted detection of biomolecules of interest.22,23 To achieve specific or targeted protein detection in tissue, especially in FFPE tissue by MSI, approaches involving the use of mass tags have been successfully developed. A mass tag is an encoding molecule with a unique mass to charge ratio (m/z) that is used to represent the target protein molecule. Usually, they are small molecules ( BT474 ≈ SKOV3 > MDA-MB-453.37 Fast Blue BB, deposited through AP catalysis, was used as the mass tag for detecting HER2 expression in the FFPE cell lines (see Figure S1, Supporting Information, for images). The deposition process was identical to that described previously for tissue. For each cell line, three regions were selected and imaged, and the overall average of the mass tag intensity of each region was summarized (Figure 9). The detected MSI intensity levels of the cell lines are in agreement with the known HER2 expression levels in the cell lines examined. Although the absolute amount of protein cannot be determined due to lack of calibration, the results indicated a strong correlation between literature reported protein abundance trends and the MSI signals. In recent years, there has been increasing interest in mass spectrometric imaging and its use for visualization of protein targets in histopathological samples. The capability to provide

Figure 5. Mass spectrometric image of detection of Ki67 in FFPE tonsil tissue using silver based mass tag: (a) optical image; (b) heat map of m/z = 216 for Ag2+; (c) heat map of m/z = 322 for Ag3+; (d) overlaid heat map of m/z = 216 and 322; and (e) overall average mass spectrum of the imaged region, showing the dominating peaks of Ag2+ and Ag3+, with no background signal observed.

the photocleavable products. In general, if electron donating groups (e.g., methoxy, ethoxy, etc.) are present in the diazonium salts, the azo products are readily cleaved by the UV laser (entries 1 to 4 in Table 1) and detected by mass spectrometry. However, if electron withdrawing groups (e.g., nitro and diazonium etc.) are present as substituents in the conjugated aromatic system, no photocleavage is observed (entries 9 to 11 in Table 1). Under high laser power, parent azo compounds were observed in MS. Substituents with weak electronic effects resulted in intermediate photocleavage efficiency, as both the parent compounds and the photocleaved cations were observed (entries 5 to 8). The identification of such a trend provides additional opportunities for the rational

Figure 6. AP catalyzed deposition of azo dyes and the mass tag detected under LDI conditions. 1464

dx.doi.org/10.1021/ac402718f | Anal. Chem. 2014, 86, 1459−1467

Analytical Chemistry

Article

Table 1. Screening of Diazonium Salts to Generate LDI Detectable Azo Mass Tags for LDI-MSI

both molecular and contextual information by MSI is very attractive for the detection of cancer biomarkers in tissue specimens. The potential to achieve high level multiplexed and quantitative protein detection in tissue makes MSI a unique and powerful new technology promising to provide information previously unavailable in cancer diagnostics. However, the promise of the applications for clinical tissue diagnostics have been slow to develop, largely due to the challenges associated with tissue specimens, especially FFPE samples, and the technological limitations of MSI methods. A suitable strategy such as mass tag based MSI will facilitate the development of the technology and its utilization in medical practice. The methodology we report here combines enzymatic amplification and antibody immuno-recognition with mass spectrometric detection of small molecules (mass tags), to overcome the difficulties in MSI for FFPE tissues. The resulting successful detection of protein targets in FFPE tissue under matrix-free LDI conditions is advantageous in many ways relative to previously reported methods. The method we present uses

enzymes rather than a carrier (i.e. dendrimers) to achieve amplification and, as such, eliminates the need for labor intensive chemical synthesis of a dendrimer-mass tag complex. The flexibility and generality of the method also make it appealing for fundamental research as well as practical applications. In this report, only single targets are detected and imaged to demonstrate the principle and feasibility of the method. However, it is possible to detect multiple targets on the same tissue section using combinations of enzymes and mass tags with separable m/z values. For example, multiple enzymes can be used in conjunction with their specific substrate species. We have identified several AP substrates as reported here, and it is reasonable to expect that enzymes such as β-galactosidase could be used in a similar fashion as AP with its corresponding substrate moiety.38 Alternatively, a sequential staining method can be adopted which allows for the use of the same enzyme for multiple mass tag deposition. Usually, a washing or inhibition step would be involved to deactivate the enzyme after the 1465

dx.doi.org/10.1021/ac402718f | Anal. Chem. 2014, 86, 1459−1467

Analytical Chemistry

Article

Figure 9. Average FBBB mass tag intensity detected in different FFPE cell lines for HER2.

HER2 in different cell lines (Figure 9), other confounding factors such as calibration and ion suppression still need to be carefully addressed.41 In addition, the tissue sample itself is highly heterogeneous in its biochemical (protein or lipid or nucleic acid) and physical properties (i.e., thickness and density). As a result, the tissue can impact the signal intensity of the detected mass tag in significant and unpredictable ways. In some cases, this very complexity of the tissue induces different ion detectability even with similar staining intensity by color, as observed in Figure 2. To overcome this variability, incorporation of an internal standard seems to be necessary to generate more reliable quantitative information;42 efforts in these directions are currently underway. In summary, we have demonstrated an MSI method which uses enzymatic deposition of mass tags, to achieve signal amplification and targeted detection of protein targets in FFPE tissues. The ability to detect a wide variety of proteins without the need of matrix helps to overcome previous limitations in MSI, i.e., low spatial resolution, detection of proteins with high expression levels, detection of proteins which are smaller than 50 KDa, and incompatibility with FFPE tissues. The method presented here could open new routes of cancer biomarker detection in clinically relevant settings.

Figure 7. Mass spectrometric image of FB-BB stained Ki-67in tonsil tissue: (a) structures of the azo compound deposited and the mass tag detected upon LDI; (b) heat map of ions with m/z = 482; (c) optical images of FB-BB staining; and (d) overall average MS spectrum of the imaged region.

■

ASSOCIATED CONTENT

S Supporting Information *

Mass spectrometry images of HER2 cell lines by detecting FBBB mass tag. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

Figure 8. Mass spectrometric image of FB-RR stained Ki-67 in tonsil tissue: (a) structures of the azo compound deposited and the mass tag detected upon LDI; (b) heat map of ions with m/z = 453; (c) optical images of FB-RR staining; and (d) overall average MS spectrum of the imaged region.

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We would like to thank Dr. Gary Kruppa and Dr. Shannon Cornett (Bruker Daltonics Inc., Billerica, MA, USA) for assistance in LDI-MSI. We would like to thank Adrian Murillo, Julia Ashworth-Sharpe, and Dr. Brian D. Kelly (Ventana Medical Systems, Inc.) for valuable discussions and assistance with IHC staining.

39

reaction. Although significantly more steps would have to be added to the staining process, the use of automated stainers would make the process practical and reproducible. The demonstration of this capability is beyond the scope of the current paper and will be reported in a separate publication.40 Another potential advantage of the mass spectrometry based detection method is the quantification capability. Although we have observed strong correlation between protein concentrations and the MSI intensities in several systems, such as

■

REFERENCES

(1) Hanahan, D.; Weinberg, R. A. Cell 2011, 144, 646−674.

1466

dx.doi.org/10.1021/ac402718f | Anal. Chem. 2014, 86, 1459−1467

Analytical Chemistry

Article

(2) Brennan, D. J.; O’Connor, D. P.; Rexhepaj, E.; Ponten, F.; Gallagher, W. M. Nat. Rev. Cancer 2010, 10, 605−617. (3) Ely, S. Transl. Res. 2009, 154, 303−308. (4) Haines, D. M.; West, K. H. Vet. Immunol. Immunopathol. 2005, 108, 151−156. (5) Aebersold, R.; Mann, M. Nature 2003, 422, 198−207. (6) Caprioli, R. M.; Farmer, T. B.; Gile, J. Anal. Chem. 1997, 69, 4751−4760. (7) Schwamborn, K.; Caprioli, R. M. Nat. Rev. Cancer 2010, 10, 639− 646. (8) Chughtai, K.; Heeren, R. M. A. Chem. Rev. 2010, 110, 3237− 3277. (9) Hardesty, W. M.; Caprioli, R. M. Anal. Bioanal. Chem. 2008, 391, 899−903. (10) Lemaire, R.; Desmons, A.; Tabet, J. C.; Day, R.; Salzet, M.; Fournier, I. J. Proteome Res. 2007, 6, 1295−1305. (11) Cornett, D. S.; Frappier, S. L.; Caprioli, R. M. Anal. Chem. 2008, 80, 5648−5653. (12) Sjövall, P.; Lausmaa, J.; Johansson, B. Anal. Chem. 2004, 76, 4271−4278. (13) Stoecklia, M.; Staaba, D.; Schweitzer, A. Int. J. Mass Spectrom. 2007, 260, 195−202. (14) qSchwamborn, K.; Caprioli, R. M. Mol. Oncol. 2010, 4, 529− 538. (15) Cornett, D. S.; Reyzer, M. L.; Chaurand, P.; Caprioli, R. M. Nat. Methods 2007, 4, 828−833. (16) Seeley, E. H.; Caprioli, R. M. Trends Biotechnol. 2011, 29, 136− 143. (17) Seeley, E. H.; Caprioli, R. M. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 18126−18131. (18) van Remoortere, A.; van Zeijl, R. J. M.; van den Oever, N.; Franck, J.; Longuespee, R.; Wisztorski, M.; Salzet, M.; Deelder, A. M.; Fournier, I.; McDonnell, L. A. J. Am. Soc. Mass Spectrom. 2010, 21, 1922−1929. (19) Garden, R. W.; Sweedler, J. V. Anal. Chem. 2000, 72, 30−36. (20) Armstrong, D. W.; Zhang, L.; He, L.; Gross, M. L. Anal. Chem. 2001, 73, 3679−3686. (21) Yang, J.; Caprioli, R. M. Anal. Chem. 2011, 83, 5728−5734. (22) Franck, J.; Arafah, K.; Elayed, M.; Bonnel, D.; Vergara, D.; Jacquet, A.; Vinatier, D.; Wisztorski, M.; Day, R.; Fournier, I.; Salzet, M. Mol. Cell. Proteomics 2009, 8, 2023−2033. (23) Cornett, D. S.; Mobley, J. A.; Dias, E. C.; Andersson, M.; Arteaga, C. L.; Sanders, M. E.; Caprioli, R. M. Mol. Cell. Proteomics 2006, 5, 1975−1983. (24) Thiery, G.; Shchepinov, M. S.; Southern, E. M.; Audebourg, A.; Audard, V.; Terris, B.; Gut, I. G. Rapid Commun. Mass Spectrom. 2007, 21, 823−829. (25) Thiery, G.; Anselmi, E.; Audebourg, A.; Darii, E.; Abarbri, M.; Terris, B.; Tabet, J.-C.; Gut, I. G. Proteomics 2008, 8, 3725−3734. (26) Lemaire, R.; Stauber, J.; Wisztorski, M.; Van Camp, C.; Desmons, A.; Deschamps, M.; Proess, G.; Rudlof, I.; Woods, A. S.; Day, R.; Salzet, M.; Fournier, I. J. Proteome Res. 2007, 6, 2057−2067. (27) Shchepinov, M. S.; Korshun, V. A. Chem. Soc. Rev. 2003, 32, 170−180. (28) Yang, J.; Chaurand, P.; Norris, J. L.; Porter, N. A.; Caprioli, R. M. Anal. Chem. 2012, 84, 3689−3695. (29) Rhodes, A.; Jasani, B.; Couturier, J.; McKinley, M. J.; Morgan, J. M.; Dodson, A. R.; Navabi, H.; Miller, K. D.; Balaton, A. J. Am. J. Clin. Pathol. 2002, 117, 81−89. (30) Shi, S. R.; Cote, R. J.; Taylor, C. R. J. Histochem. Cytochem. 1997, 45, 327−343. (31) Thiery, G.; Mernaugh, R. L.; Yan, H.; Spraggins, J. M.; Yang, J.; Parl, F. F.; Caprioli, R. M. J. Am. Soc. Mass Spectrom. 2012, 23, 1689− 1696. (32) Mullink, H.; Vos, W.; Jiwa, W.; Horstman, A.; van der Valk, P.; Walboomers, J. M.; Meijer, C. J. J. Histochem. Cytochem. 1992, 40, 495−504. (33) Hacker, G. W.; Grimelius, L.; Danscher, G.; Bernatzky, G.; Muss, W.; Adam, H.; Thurner, J. J. Histotechnol. 1988, 9, 213−221.

(34) Bieniarz, C.; Farrell, M. U. S. Patent 7642064, 2010. (35) Hainfeld, J. F. U. S. Patent 6670113, 2003. (36) Sherrod, S. D.; Diaz, A. J.; Russell, W. K.; Cremer, P. S.; Russell, D. A. Anal. Chem. 2008, 80, 6796−6799. (37) Rusnak, D. W.; Alligood, K. J.; Mullin, R. J.; Spehar, G. M.; Arenas-Elliott, C.; Martin, A. M.; Degenhardt, Y.; Rudolph, S. K.; Haws, T. F., Jr.; Hudson-Curtis, B. L.; Gilmer, T. M. Cell Proliferation 2007, 40, 580−594. (38) Holzmann, B.; Johnson, J. P. J. Immunol. Methods 1983, 60, 359−367. (39) Pirici, D.; Mogoanta, L.; Kumar-Singh, S.; Pirici, I.; Margaritescu, C.; Simionescu, C.; Stanescu, R. J. Histochem. Cytochem. 2009, 57, 567−575. (40) Manuscript in preparation. (41) Heeren, R. M.; Smith, D. F.; Stauber, J.; Kükrer-Kaletas, B.; MacAleese, L. J. Am. Soc. Mass Spectrom. 2009, 20, 1006−1014. (42) Pirman, D. A.; Kiss, A.; Heeren, R. M.; Yost, R. A. Anal. Chem. 2013, 85, 1090−1096.

1467

dx.doi.org/10.1021/ac402718f | Anal. Chem. 2014, 86, 1459−1467