Abstract
Spectral flow cytometry is a new technology that enables measurements of fluorescent spectra and light scattering properties in diverse cellular populations with high precision. Modern instruments allow simultaneous determination of up to 40+ fluorescent dyes with heavily overlapping emission spectra, discrimination of autofluorescent signals in the stained specimens, and detailed analysis of diverse autofluorescence of different cells—from mammalian to chlorophyll-containing cells like cyanobacteria. In this paper, we review the history, compare modern conventional and spectral flow cytometers, and discuss several applications of spectral flow cytometry.
You have full access to this open access chapter, Download protocol PDF
Similar content being viewed by others
Key words
- Spectral cytometry
- Flow cytometry
- Fluorescence spectra
- Aurora cytometer
- Sony spectral analyzer
- Autofluorescence
- Spectral unmixing
- Virtual filtering
1 Introduction
Flow cytometry began its development in the middle of the twentieth century and has established itself as one of the major functional methods widely used by scientists and clinicians. As it developed, flow cytometry in the twenty-first century diverges into the following directions: (1) Conventional flow cytometry and fluorescent activated cell sorting (FACS); (2) Imaging flow cytometry; (3) Spectral flow cytometry (spectral FCM).
Conventional cytometry allows studying the size, granularity, and several fluorescent signals of individual cells or particles at the rate of 1000 events per second. Imaging flow cytometry, a hybrid technology, which combines the principles of flow cytometry and microscopy, allows obtaining an image of each cell and thus collects galleries of images along with light scatter and fluorescent signals. However, its throughput is significantly less than conventional flow cytometry [1]. Spectral FCM, which is based on spectroscopy, made it possible to record the full spectrum of every single cell during measurements and now operates at a rate similar to conventional flow cytometry. Both imaging flow cytometry and spectral FCM allow sophisticated offline analysis of the specimens. Recent technical advances in multicolor cytometry were focused on detecting and analyzing cellular subpopulations with complex immunophenotypes participating in the immune response to diseases and/or vaccine response [2, 3]. Besides, significant progress in the decomposition of complex fluorescent spectra was introduced by Rosetti and co-authors [4], which could improve spectral unmixing and detection of autofluorescence. It will allow better separation of negative, dim, and positive populations using multicolor labeling.
2 Development of Spectral Flow Cytometry
Wade and colleagues made one of the first attempts to extract full emission spectra during flow cytometry analysis in 1979 [5]. They used a grating spectrograph and projected the spectrum of the fluorescent signal onto TV type Vidicon detector (objects: Anacyctis nidulans—chlorophyll and phycobilin, 600–750 nm; 3T3 fibroblasts from Balb/c mice, propidium iodide (PI) and fluorescamine staining) [5]. The recorded signal from individual cells has a very low signal-to-noise (S/N) ratio, and reasonable spectra were obtained only by averaging recordings from hundreds of cells (fibroblasts).
The next configuration of the spectral flow cytometer was based on a photomultiplier tube (PMT) as a detector and grating monochromator. This cytometer had a 10 nm bandwidth spectral resolution, and signal detection was performed when cells were running through the cuvette at a rate of hundreds of events per second [6]. The system eliminated problems of the noisy background using PMT with adjustable gain and offset as a detector (analyzed objects—fixed rat thymocytes, stained with Hoechst 33258) [6]. However, the spectrum measured was only between 400 and 600 nm, not including far-red and infrared (IR) wavelengths.
Buican [7] described in 1990 a “real-time FT spectrometer” that was an interferometer-based spectral detector using PMT with minimal time needed for the recording of the spectrum (only 3.2 μs). However, this instrument was never used as a commercially available cytometer. Subsequently, several more spectral systems were created in an attempt to obtain spectra from short measurements using conventional cytometers in 1990–2000. Thus, Gauci and co-authors described configuration with the prism and 512-element intensified photodiode array based on the FACS IV laser flow cytometer. They analyzed spectra obtained from Dictyostelium discoideum spores stained with Cy3, fluorescein isothiocyanate, R-phycoerythrin (R-PE), and calibration beads [8]. This system was relatively slow (operating at 62.5 Hz) and not sensitive enough to show individual spectra of the labeled cells [8]. Furthermore, Asbury et al. (1996) [9] were the first group able to obtain the fluorescent spectrum using a standard flow cytometer (Cytomation, USA) with a monochromator attached in front of PMT used as a detector of fluorescent signal. However, this was not a real spectral FCM yet. The monochromator was operating sequentially—for each wavelength (spectral point), 100 events were recorded. Then monochromator was shifted to the adjacent position, recording another 100 events and so on. The overall spectrum (400–800 nm) was built up from measurements made on 20,000 particles.
At the beginning of the multicolor analysis, the sensitivity of flow cytometers and confocal microscopes in the far-red and IR parts of the spectrum was limited by the low sensitivity of PMTs at wavelengths beyond 650 nm [10]. The use of avalanche photodiode detectors (APD) led to substantially better S/N performance over the PMT in the red and near-IR spectral regions. Changing conventional PMTs to APD and APD arrays [11, 12] made it possible to achieve reasonable S/N for multichannel detectors using short-time exposures even in near IR (wavelengths up to 800–900 nm) [13]. An alternative type of detector was used by Isailovic and colleagues [14]. Their instrument (single-cell fluorescence spectrometer) was based on ICCD (intensified charge couple device) detector and used a 5–20 ms exposure time, thus coming close to the real spectral FCM. Using this instrument, they demonstrated that measurement of individual spectra with a spectral resolution of 6.5 nm from fluorescently labeled E. coli expressing GFP and non-fluorescent apo-subunits of R-PE gives more accurate results compared to the measurement of bulk spectra.
Since the beginning of the twenty-first century, various systems have achieved sufficient sensitivity for recording a spectrum of fluorescent signals from a single cell in a reasonably short time. The next step in the development of spectral flow cytometers became possible when computer speed accelerated and paralleled recording of multiple signals with high frequency was achieved on a standard PC. Rapid registration of fluorescent spectra was done using parallel data recording and digital processing. These instruments were based on multidetector arrays, where emission light is split and projected onto the grid of PMTs or APDs. A flow cytometer equipped with 32-channel Hamamatsu multi-anode PMT able to collect spectral information in not more than 5 μs was built in Purdue University Cytometry Laboratory and later patented by Purdue University [15, 16]. This instrument allowed a digitization rate of up to 75,000 complete a 32-channel spectra per second at 14 bits dynamic range for uniformly (in time) presented events. The system was based on an EPICS Elite cell sorter (Beckman Coulter, USA) equipped with argon (488 nm) and HeNe (633 nm) lasers [16]. This system achieved a speed of 3000 random events per second; however, the sensitivity was lower than that of conventional filter-based detectors.
A similar system based on a modified BD FACSCalibur cytometer equipped with argon-ion laser and 100 W mercury lamp was built by Goddard and co-authors [17] using a grating spectrograph and Hamamatsu CCD array with 80% quantum yield. The spectra analyzed by this instrument were in the range of 500–800 nm. This instrument allowed recording spectra with great linearity, making spectral subtraction to remove background signals from labeled specimens such as Rayleigh scattering, Raman light scatter, and even cell autofluorescence feasible. Also, the sensitivity of the instrument was significantly lower (10–30 times) than that of the conventional cytometer [17].
Alternative spectral cytometry systems used a charge-coupled device (CCD) camera as a detector to measure spectra from single cells and beads [17,18,19]. In 2012 Nolan’s group [20] developed spectral FCM instruments and data analysis algorithms suitable for everyday use. Their two systems were based on FACSCanto equipped with 405 and 488 nm lasers and using EM-CCD (electron-multiplying CCD) detector (11.3 nm resolution in the 500–800 nm range) and Coulter Elite cytometer using 785 nm laser for IR emission (at 3.23 nm resolution in 790–930 nm range). Their spectral flow cytometers used a holographic grating and EM-CCD detector for high-speed spectra detection. Customized software was developed for the spectral unmixing and production of spectra-derived parameters for individual cells.
Instrument calibration and data analysis were very complicated at these early stages of spectral FCM development (circa 2012) [21]. Instrument design was not standardized, requiring thorough spectral calibration for each instrument. Also, different instruments used different data formats, making cross-platform spectral analysis tricky. In the first spectral cytometers, spectral unmixing was performed through the least square unmixing algorithm or indirectly through principal component analysis [22]. Overall comparing the spectral data obtained by different instruments was practically impossible. So far, at that time, the advantages of spectral FCM over conventional multichannel flow cytometry were impossible to use in many applications. The next step was done when commercially available spectral cytometers with standardized parameters appeared.
The system patented by Purdue University was licensed by Sony Inc., which is producing the first-generation commercial spectral cytometry system (sometimes named hyperspectral cytometer)—the Sony SP6800 Spectral Analyzer was announced at the end of 2012 and came to the market in 2014. Also, in 2014 Cytek Biosciences (USA) developed and soon released its Aurora spectral flow cytometer. Nowadays, two companies are concerned with the production of commercial models of spectral cytometers: (1) Sony Biotechnology (spectral cell analyzers SA 3800, SP 6800, ID 7000); (2) Cytek Biosciences (Cytek Aurora and Northern Lights instruments). In summary, recent advances in hardware, detectors, and computer analysis algorithms resulted in commercially available spectral FCM instrumentation.
3 Current Spectral Cytometry Instruments
Modern Sony ID7000 instrument supports up to 7 lasers and can use up to 168 detectors (in 7 laser configuration) covering the spectral range from 360 to 920Â nm with ~10Â nm resolution. Specialized InGaAs PMTs are used for efficient capturing of the IR signals.
Aurora Cytek spectral cytometer measures fluorescence in up to 64 fluorescent channels (in the 5-lasers instrument—16UV + 16 V + 14B + 10YG + 8R) across the APD detector arrays (Fig. 1). Each channel uses a special bandpass filter with about 10–15 nm bandwidth, reflecting all wavelengths outside of its transmission band. The full spectral range is 400–900 nm. In both types of instruments, lasers excite the specimen sequentially.
Three laser Aurora Cytek instrument—optical setup. Fluorescence signal is delivered to the sets of detectors (V for violet excitation, B for blue excitation, and R for red excitation). Notice that SSC signal is measured for each laser, and the number of APD detectors is different. Laser beams are spatially separated at the conventional cytometer. Picture was modified from figures given at Aurora Cytek website (https://cytekbio.com>pages>aurora-cs)
4 Advances and Limitations of Spectral Flow Cytometry
A critical review of the latest advances and remaining problems in spectral FCM was published recently [23]. The essential aspects of spectral FCM are that instrument performance in the case of Cytek Aurora strongly depends on the characteristics of each filter (total—of 64 filters). For example, a thorough check uncovered two out-of-specification filters in the commercial instrument that precluded efficient separation of eFluor450 from BV421 and SB436 [23]. Other issues dealt with laser delay and titration of antibodies. In the case of spectral FCM, titration of antibodies is more complicated because of living and dead cells in the same tube. Authors suggest using live and dead cell markers along with a standard set of CD markers, making titration a multistep process. This process can be described as inversed to FMO (fluorescence minus one) controls used in conventional multicolor cytometry. The sequence of suggested tests for titration is the following: viability dye, major markers like CD45, lineage-specific markers (CD3, CD19, etc.), and finely more specific markers to identify small subpopulations of blood cells [23].
5 Development of Spectral Unmixing Algorithms
The significant advantage of spectral measurements against conventional flow cytometry is its ability to make a detailed comparison of fluorescent spectra from individual cells (objects) in a heterogeneous population. Multiparametric cytometry often has bleed-through problems due to the overlapping spectra of fluorophores. To identify and characterize complex interactions of multiple cell types, it is necessary to analyze a significant number of fluorescent labels simultaneously. Fluorescence signals were initially analyzed as a linear combination of reference spectra with algorithms extracting the weight of individual spectra (linear unmixing) [24]. Identification of heavily overlapping spectra can be performed to a limited extent using the spectral compensation procedure, and instead, spectral unmixing was introduced. Spectral unmixing refers to a group of techniques that attempt to determine how much each fluorophore contributes to the observed emission spectrum. It was initially suggested for microscopy [25] and later applied in flow cytometry [21, 26]. Spectral unmixing in cytometry allows analysis of the simultaneous labeling of cells with several fluorophores and/or fluorescent proteins. Spectral unmixing methods have been developed extensively for the remote sensing analysis of hyperspectral data [27, 28]; however, some key differences make many unmixing algorithms unsuitable for spectral cytometry: (a) the number of fluorophores used for cellular staining is known a priori, though the number of autofluorescent signals can be unknown; (b) remote sensing spectral analysis is focused on blind unmixing of source signals while in spectral cytometry it is possible to use reference spectra to define emission spectral endmembers.
6 Spectral Unmixing Problems
The fluorophores originated from algal photosynthetic apparatus such as PE, APC, and PERCP have broad and overlapping spectra, and to some degree, can be excited by violet laser (405 nm excitation) [29]. Synthetic dyes such as Alexa Fluor and Cyan families are small organic fluorophores that do not exhibit much crossbeam excitation. Most spectral unmixing algorithms cannot separate a signal from background noise or autofluorescence. Autofluorescence is a common, undesired signal arising from endogenous fluorophores contained in the cells or extracellular matrix (i.e., NAD(P)H, flavine adenine nucleotide (FAD), lipids, collagen, elastin, and other common fibrous proteins, porphyrins) [30] often with wide emission spectra [31]. One of the major endogenous fluorophores inside cells is a mitochondrial NADH (Exc./Em. 350/460 nm) [32], declining with cellular injury. Cellular samples may contain different types of autofluorescent molecules, and it is challenging to predict their distribution since they can change in time (the cell is dying or becoming apoptotic). Spectral unmixing for subtracting autofluorescence is possible using the non-negative matrix factorization variant of spectral unmixing, which exploits spectra obtained at the different excitation wavelengths [4, 33].
7 Comparison of Spectral Unmixing and Spectral Compensation
Despite extensive development in cytometry, the compensation stays based upon the classical algorithms, using the single controls approach developed by Bagwell and Adams [34], with some recent developments [35, 36]. Two methods of separating fluorophore signals in multicolor cytometry were recently compared by Niewold and colleagues [35]. One of the major limitations of spectral compensation is the increased spread of compensated signals compared to the original ones that diminish the ratio between mean/median values of positive and negative populations [37]. Particularly it precludes discrimination between negative and dim populations.
For some highly overlapping fluorophores, spectral unmixing algorithms made it possible to resolve the two fluorescence signals where spectral compensation did not. Unmixing in spectral cytometers gives less spreading, which is important when using numerous (panels >16) fluorophores [35]. However, if the cytometer uses optical filters (Aurora spectral analyzer, Cytek, USA), the quality of these filters plays a crucial role in the spreading when unmixing similar spectra [3]. The commercially available filters might slightly deviate from the characteristics provided by the supplier and, thus, sometimes, do not adequately exclude the fluorescent emission of other fluorophores and/or autofluorescent molecules that overlap with the desired signal. In commercial spectral cytometers from SONY, instead of the optical filters, specialized prism-based optics are used to measure and separate emissions from different fluorophores [38].
Another advantage of spectral unmixing in spectral cytometry is better extracting of autofluorescence signal that could be treated as an additional fluorophore [36], while compensation cannot be applied to autofluorescence until its spectrum is recorded.
8 Comparison of Spectral Cytometry and Mass Cytometry
Flow cytometry allows analysis of up to 25–40 parameters at a rate of several thousands of events per second. On the other hand, mass cytometry, currently a competitor to spectral FCM, allows typing of various immune cells on panels from 14 to 42 parameters with minimal overlap between channels and without autofluorescence [39,40,41,42]. Despite these benefits, broader practical applications of mass cytometry are affected by limitations such as slow collection rates (300–500 events/s vs. several thousand events/s. with conventional cytometry) and total cost of experimentation/ownership [43].
9 Differences and Similarities Between Spectral and Conventional Flow Cytometry
The common feature of spectral and conventional cytometry is the observation of a single cell. The full spectrum of a single event can be detected under the action of hydrodynamic focusing, where the cell passes an interrogation point and is excited by a collinear or non-collinear laser system. Subsequently, the detection of the emission signal for these two systems is fundamentally different. Spectrum detection became possible because of a unique emitting optical system. This system uses prisms and gratings to disperse fluorescence light, while a conventional cytometer splits fluorescent signal using bandpass, short pass, and long pass filters (Fig. 2). Prisms as dispersive optics in spectral FCM propagate light in a non-linear manner, unlike gratings that propagate light into a detector in a linear manner. Moreover, spectral cytometry to detect the full spectrum uses an array of detectors such as CCDs and multianode PMTs, while in most conventional configurations, separate PMT is utilized in each forward scatter (FSC), side scatter (SSC), and fluorescence channel.
The differences and similarities between spectral and conventional cytometry. Conventional cytometry: optical part – dichroic mirrors and bandpass filters. Light collection – reflection, transmission, blocking. Detectors – photomultipliers (PMT). Spectral flow cytometry: optical part – grating or prisms. Light collection - dispersion. Detectors – multianode PMTs or CCD
Further development of the real-time spectral FCM allowing measurement of emission spectra in the flow cell with the frequency typical to that of the standard flow cytometer (about 10,000 events per second) as well as the use of the spectral detectors in fluorescent microscopy was stimulated by the development of numerous fluorescent proteins with similar spectra [44]. Emission spectra of these proteins overlap significantly and thus cannot be distinguished by conventional fluorescent microscopy or FCM using dichroic mirrors and even highly selective bandpass filters [38].
This principle of spectral FCM operation is used with commercial spectral cytometry companies but with some differences in optical layout. The Sony spectral analyzer separates the emitted light with a set of prisms before sending it to 32-channel PMT arrays. To capture the fluorescence spectrum, the Cytek Aurora system employs multiple APDs with a unique set of filters in front of each APD. The possibility of obtaining a full emission spectrum with commercial spectral analyzers allowed new combinations of fluorochromes, which due to the significant spectra overlap, are not used together in conventional cytometry. Moreover, spectral FCM allows using more fluorochromes per experiment. To address the existing gap in commercially available fluorochromes, new dyes are necessary, and this need started to be addressed [45]. Another advantage of Spectral FCM is extracting the autofluorescence (AF) of cells and using it as a separate parameter(s) [46], allowing better signal resolution and even a comparison of different autofluorescent parameters [47].
10 Applications of Spectral FCM
Major problems of conventional flow cytometry can be solved using the spectral FCM: (1) enhanced number of fluorescent parameters used in a single tube (hematology, minimal residual disease (MRD)); (2) subtraction of fluorescent signal with the improvement of S/N ratio and detailed analysis of autofluorescence signal for analysis of unlabeled cells. The enhanced number of fluorescent channels is critical for analyzing small biopsies such as bone marrow aspirates in MRD. Subtraction of autofluorescence is particularly helpful for the analysis of cells with a high level of autofluorescence, such as myocytes, macrophages, brain cells, and hepatocytes. Primary cells are heterogeneous, and each subpopulation may require assigning its autofluorescence as a separate fluorophore and performing additional spectral unmixing [3].
11 Current Applications: Multi-parametric Spectral Cytometry
Nevertheless, certain studies were already made at the early stage of spectral cytometry. In 2015 Futamura and co-workers [38] described an analysis of lymphocyte migration from the individual lymph node (within 24 h) and using photoconvertible protein, and 11-color labeling showed that CD69 low naive T cell subset was replaced in lymph node faster than CD69 high memory T-cell subsets [36]. Schmutz and co-authors (2016) [48], using a two-laser Sony SP6800 instrument (405 and 488 nm), demonstrated by detailed fluorescence-minus-one control (FMO) that while the staining index (SI) for individual dyes in spectral FCM was the same as in conventional FCM, spectral FCM gives much better discrimination of dyes with similar fluorescent properties. Spectral FCM allowed discrimination of dyes with the same peak fluorescence intensity when the overall spectra were different and dyes with similar spectra but shifted for 10–20 nm peaks using Kaluza software (YFP versus GFP; both proteins versus FITC) [48].
Besides, spectral FCM allowed discrimination of lymphocytes among the cells isolated from the tissues with high autofluorescence. Complete elimination of autofluorescent signal makes it possible to discriminate dye-positive and dye-negative cells using dyes with emission spectra close to the autofluorescent spectra for further analysis [48].
The Sony SP6800 Spectral Cell Analyzer instrument utilized a 32 multianode PMT (Hamamatsu), and spectrum separation is achieved through a complex prism-based monochromator. SONY Inc. demonstrated a prototype instrument and reported on hyperspectral technology during the ISAC Congress in Seattle in 2012 and announced the launch of the new hyperspectral flow cytometer product—an SP6800 Spectral Cell Analyzer—in 2012.
In some applications, the multiplexing by spectral tags may not require spectral unmixing. In this setting, it may be beneficial to classify the spectra directly instead of classification based on unmixed intensities. Many techniques may be utilized here, including unsupervised data reduction (using, for example, principal component analysis, independent component analysis, or factor analysis) or supervised techniques (such as neural networks or support vector machines).
Advantages of spectral cytometry such as a large number of studied parameters in one panel with better resolution due to the removal of the autofluorescence signal and a rate of several thousand events per second (Sony SP6800 10,000–20,000 events/second, Cytek Aurora 35,000 events/s), have led to an increase in the practical use of spectral cytometers in immunophenotyping. One of the first multicolor panels (nine colors) was created by Futamura and co-authors [38] at the presentation of the Spectral Analyzer SP 6800 to study the movement of KikGr protein after photoconversion in the inguinal lymph node cells. The remaining immune cells, after photoconversion, changed their emission from green to red (KikGrGreen-KikGrRed) while migrated cells stayed green. In this experiment, the emission spectra of fluorochromes and fluorescent proteins, which strongly overlapped with each other, were separated using spectral unmixing (EGFP/FITC/KikGr-Green, KikGr-Red/PE, KikGr-Green/Venus, EGGP/Venus, KikGr-Red/mKO2) [36]. It would be difficult to apply this panel in conventional flow cytometry, and with the spectral analyzer, it became possible to separate and eliminate the low and high levels of autofluorescence that were found in the mouse splenocytes with strong expression of F4/80 marker (major macrophage biomarker, APC labeled) [38]. Solomon and co-authors [49] used a 15-fluorochrome panel and spectral FCM to describe the aging of the bone marrow in mice.
The separation of lasers at the Cytek spectral flow cytometer allowed the creation of 30–40 multicolor cytometric panels. The 40-color panel OMIP-069 with Aurora for identifying T cells, B cells, NKT—like cells, monocytes, and dendritic cells was reported recently [50]. This panel is effective in the study of the immune response with low sample volume [50]. In this panel, with spectral cytometry, it became possible to use dyes that have a strong overlap of the emission signal between them (PE/FITC, PE-Alexa Fluor 700/PerCP-eFluor 710, BUV 496/eFluor 450, SuperBright 436). Using data acquired by a 3-laser 38-color Aurora (Cytek, USA) spectral cytometer and analyzed by Kaluza and FlowJo software, Chen and co-authors [51] demonstrated that SFC allows distinguishing subsets of myeloid cells when using one tube with 24-color staining more precise compared to the standard 3*8-color panel. By automated clustering, malignant cells from patients with minimal residual disease (MRD) were distinguished from rare normal mast cells and basophils. In the early study, Murphy and colleagues [52] conducted a similar study for typing human peripheral blood mononuclear cells (PBMCs), but separate panels have been developed for the determination of T cells (23 colors) and B cells (22 colors). Schmutz and co-authors [48] described a 19 colors panel for the separation of murine splenocytes into B-, T-, NK-, and dendritic cells.
A new generation of SONY spectral cytometers—ID7000 also has a combination of separate lasers (for sequential excitation). It allows the use of multicolor panels, such as a 28-colors panel for immune-profiling of COVID-19 patients [53]. Two highly autofluorescent fetal liver stromal subsets were clearly discriminated using spectral unmixing with autofluorescence assigned as an independent parameter [47]. The use of other multicolor panels for immunophenotyping with a spectral cytometer is summarized in Table 1.
12 Two Major Types of Spectral FCM Analysis: Virtual Filtering and Spectral Unmixing
Spectral unmixing is the most used and considered to be the most powerful approach, but it requires a thorough recording of autofluorescent controls from heterogeneous cellular populations. Sophisticated spectral unmixing with commercially available software allows robust separation from 4+ to 20+ fluorochromes. Another less powerful but more universal approach is virtual filtering. It was initially demonstrated in the phytoplankton study [71, 72]. Spectral cytometry allowed effective selection of “filtering off” autofluorescent part of spectra, which may overlap with fluorescent signals in the multiparametric analysis of multiple taxa of algae [71]. It mimics the interchange of hardware filters in the PMT channels in a standard flow cytometer. In conventional cytometry, changing optical filters means manipulation with hardware, and some optical bandpass filters may not be available on the market. With SFC, we can make a large selection of virtual filters after the sample is recorded [72].
The use of multiple fluorescent conjugates and dyes/pigments significantly affected cytometric analysis facilitating multivariate analysis, dimensionality reduction algorithms based on stochastic neighbor embedding (SNE), unsupervised cluster analysis, and cell-subset identification programs such as SPADE, CITRUS, FlowSOM, CellCNN, and viSNE [73,74,75,76,77]. An alternative to clustering algorithms is principal component analysis (PCA), which is widely used in other areas of biology. Recently, Ogishi and co-authors [78] introduced iMUBAC (integration of multi-batch cytometry datasets) using unsupervised cell-type identification across multiple batches.
13 Conclusions
Currently, the spectral cytometer becomes a superior alternative to the conventional cytometer since it allows the acquisition of fluorescent dyes and proteins without the limitations of hardware optics and detectors. It leads to reducing the complexity of multi-color panel design and allows easy acquisition of more than 20 colors with good discrimination of bright, dim, and negative cellular subpopulations. The latest multi-laser (up to seven lasers) commercially available spectral cytometer ID7000 (SONY) allows the detection and analysis of up to 40 fluorescent parameters. Spectral FCM or full-spectrum cytometry can subtract autofluorescence from signals generated by dyes without increasing spread, besides, it allows acquire autofluorescence as separate parameter(s). Spectral FSM allows detailed analysis of the autofluorescence that might be especially useful for analyzing phytoplankton where a strong autofluorescent signal from chlorophyll precludes using fluorescently labeled dyes/antibodies and for highly autofluorescent cells (macrophages, myeloid progenitors, infected cells, etc.). Available libraries of emission spectra of the numerous standard fluorophores make single-stained controls unnecessary. The limitations of the spectral deconvolution approach in Spectral FCM are related to the use of tandem dyes or the inability to use ratiometric probes. The new generation of multi-laser Spectral FSM instruments initiates a breakthrough in cytometric analysis and the replacement of conventional cytometers. Full-spectrum cell sorters and co-registering spectra with images of cells can be foreseen in the near future.
References
Imaging flow cytometry: methods and protocols (2016) Barteneva NS, Vorobjev IA (eds) Methods in molecular biology, vol 1389. Humana, New York, 295pp
Wang W, Su B, Pang L et al (2020) High-dimensional immune profiling by mass cytometry revealed immunosuppression and dysfunction of immunity in COVID-19 patients. Cell Mol Immunol 17:650–652. https://doi.org/10.1038/s41423-020-0447-2
Maucourant C, Filipovic I, Ponzetta A, Aleman S, Cornilett M et al (2020) Natural killer cell immunophenotypes related to COVID-19 disease severity. Sci Immunol 5:eabd6832. https://doi.org/10.1126/sciimmunol.abd68
Rossetti BJ, Wilbert SA, Welch JLM, Borisy GG, Nagy JG (2020) Semi-blind sparse affine spectral unmixing of autofluorescence-contaminated micrographs. Bioinformatics 36:910–917. https://doi.org/10.1093/bioinformatics/btz674
Wade CG, Rhyne RH, Woodruff WH et al (1979) Spectra of cells in flow cytometry using a vidicon detector. J Histochem Cytochem 27:1049–1052. https://doi.org/10.1177/27.6.110874
Steen HB, Stokke T (1986) Fluorescence spectra of cells stained with a DNA-specific dye, measured by flow cytometry. Cytometry 7:104–106. https://doi.org/10.1002/cyto.990070117
Buican, TN (1990) Real-time Fourier transform spectrometry for fluorescence imaging and flow cytometry. Proceedings of the SPIE, bioimaging and two-dimensional spectroscopy 1205:126–133. https://doi.org/10.1117/12.17787
Gauci MR, Vesey G, Narai J et al (1996) Observation of single-cell fluorescence spectra in laser flow cytometry. Cytometry 25:388–393. https://doi.org/10.1002/(sici)1097-0320(19961201)25:4<388::aid-cyto11>3.0.co;2-r
Asbury CL, Esposito R, Farmer C, van den Engh G (1996) Fluorescence spectra of DNA dyes measured in a flow cytometer. Cytometry 24:234–242. https://doi.org/10.1002/(sici)1097-0320(19960701)24:3<234::aid-cyto6>3.0.co;2-h
Lawrence WG, Varadi G, Entine G et al (2008) Enhanced red and near infrared detection in flow cytometry using avalanche photodiodes. Cytometry A 73A:767–776. https://doi.org/10.1002/cyto.a.20595
Zhao S, Wu X, Chen Y, et al (2011) High gain avalanche photodiode (APD) arrays in flow cytometer opitical system. 2011 international conference on multimedia technology, pp 2151–2153. https://doi.org/10.1109/icmt.2011.6002457
Yamamoto M (2017) Photon detection: current status. In: Single cell analysis. Springer, Singapore, pp 227–242. https://doi.org/10.1007/978-981-10-4499-1_10
Nolan JP, Condello D, Duggan E, Naivar M, Novo D. (2013) Visible and near infrared fluorescence spectral cytometry. Cytometry Part A; 83A:253–264. https://doi.org/10.1002/cyto.a.22241
Isailovic D, Li H-W, Phillips GJ, Yeung ES (2005) High-throughput single-cell fluorescence spectroscopy. Appl Spectrosc1 59:221–226. https://doi.org/10.1366/0003702053085124
Robinson JP, Rajwa B, Gregori G et al (2005) Multispectral cytometry of single bio-particles using a 32-channel detector. Proc SPIE 5692:359–365. https://doi.org/10.1117/12.591365
Gregori G, Patsekin V, Rajwa B, Jones J, Ragheb K, Holdman C, Robinson JP (2012) Hyperspectral cytometry at the single-cell level using a 32-channel photodetector. Cytometry A 81A:35–44. https://doi.org/10.1002/cyto.a.21120
Goddard G, Martin JC, Naivar M et al (2006) Single particle high resolution spectral analysis flow cytometry. Cytometry A 69A:842–851. https://doi.org/10.1002/cyto.a.20320
Watson DA, Brown LO, Gaskill DF et al (2008) A flow cytometer for the measurement of Raman spectra. Cytometry A 73A:119–128. https://doi.org/10.1002/cyto.a.20520
Nolan JP, Sebba DS (2011) Surface-enhanced Raman scattering (SERS) cytometry. Methods Cell Biol:515–532. https://doi.org/10.1016/b978-0-12-374912-3.00020-1
Nolan JP, Condello D, Duggan E et al (2012) Visible and near infrared fluorescence spectral flow cytometry. Cytometry A 83A:253–264. https://doi.org/10.1002/cyto.a.22241
Nolan JP, Condello D (2013) Spectral flow cytometry. Curr Protoc Cytom 63:1.27.1–1.27.13. https://doi.org/10.1002/0471142956.cy0127s63
Sanders CK, Mourant JR (2013) Advantages of full spectrum flow cytometry. J Biomed Opt 18:037004. https://doi.org/10.1117/1.jbo.18.3.037004
McCausland M, Lin Y-D, Nevers T et al (2021) With great power comes great responsibility: high-dimensional spectral flow cytometry to support clinical trials. Bioanalysis 13:1597–1616. https://doi.org/10.4155/bio-2021-0201
de Juan A, Tauler R (2021) Multivariate curve resolution: 50 years addressing the mixture analysis problem – a review. Anal Chim Acta 1145:59–78. https://doi.org/10.1016/j.aca.2020.10.051
Zimmermann T, Rietdorf J, Pepperkok R (2003) Spectral imaging and its applications in live cell microscopy. FEBS Lett 546:87–92. https://doi.org/10.1016/S0014-5793(03)00521-0
Garini Y, Young IT, McNamara G (2006) Spectral imaging: principles and applications. Cytometry A 69A:735–747. https://doi.org/10.1002/cyto.a.20311
Benachir D, Deville Y, Hosseini S, Karoui MS (2020) Blind unmixing of hyperspectral remote sensing data: a new geometrical method based on a two-source sparsity constraint. Remote Sens 12:3198. https://doi.org/10.3390/rs12193198
Wei J, Wang X (2020) An overview on linear unmixing of hyperspectral data. Math Probl Eng 2020:1–12. https://doi.org/10.1155/2020/3735403
Kronick MN (1986) The use of phycobiliproteins as fluorescent labels in immunoassay. J Immunol Methods 92:1–13. https://doi.org/10.1016/0022-1759(86)90496-5
Monici M (2005) Cell and tissue autofluorescence research and diagnostic applications. Biotechnol Annu Rev:227–256. https://doi.org/10.1016/s1387-2656(05)11007-2
Croce AC, Bottiroli G (2014) Autofluorescence spectroscopy and imaging: a tool for biomedical research and diagnosis. Eur J Histochem 58:2461. https://doi.org/10.4081/ejh.2014.2461
Chance B (2004) Mitochondrial NADH redox state, monitoring discovery and deployment in tissue. Methods Enzymol:361–370. https://doi.org/10.1016/s0076-6879(04)85020-1
Rakotomanga P, Soussen C, Khairallah G, Amoroux M, Zaytsev S et al (2019) Source separation approach for the analysis of spatially resolved multiply excited autofluorescence spectra during optical clearing of ex vivo skin. Opt Express 10:3410–1424. https://doi.org/10.1364/BOE.10.003410
Bagwell CB, Adams EG (1993) Fluorescence spectral overlap compensation for any number of flow cytometry parameters. Ann N Y Acad Sci 677:167–184. https://doi.org/10.1111/j.1749-6632.1993.tb38775.x
Niewold P, Ashhurst TM, Smith AL, King NJ (2020) Evaluating spectral cytometry for immune profiling in viral disease. Cytometry A 97:1165–1179. https://doi.org/10.1002/cyto.a.24211
Roca CP, Burton OT, Gergelits V et al (2021) AutoSpill is a principled framework that simplifies the analysis of multichromatic flow cytometry data. Nat Commun 12:1–16. https://doi.org/10.1038/s41467-021-23126-8
Nguyen R, Perfetto S, Mahnke YD et al (2013) Quantifying spillover spreading for comparing instrument performance and aiding in multicolor panel design. Cytometry A 83A:306–315. https://doi.org/10.1002/cyto.a.22251
Futamura K, Sekino M, Hata A et al (2015) Novel full-spectral flow cytometry with multiple spectrally-adjacent fluorescent proteins and fluorochromes and visualization of in vivo cellular movement. Cytometry A 87:830–842. https://doi.org/10.1002/cyto.a.22725
Baumgart S, Peddinghaus A, Schulte-Wrede U et al (2016) OMIP-034: comprehensive immune phenotyping of human peripheral leukocytes by mass cytometry for monitoring immunomodulatory therapies. Cytometry A 91:34–38. https://doi.org/10.1002/cyto.a.22894
Jaracz-Ros A, Hémon P, Krzysiek R et al (2018) OMIP-048 MC: quantification of calcium sensors and channels expression in lymphocyte subsets by mass cytometry. Cytometry A 93:681–684. https://doi.org/10.1002/cyto.a.23504
Brodie TM, Tosevski V, Medová M (2018) OMIP-045: characterizing human head and neck tumors and cancer cell lines with mass cytometry. Cytometry A 93:406–410. https://doi.org/10.1002/cyto.a.23336
Dusoswa SA, Verhoeff J, Garcia-Vallejo JJ (2019) OMIP-054: broad immune phenotyping of innate and adaptive leukocytes in the brain, spleen, and bone marrow of an orthotopic murine glioblastoma model by mass cytometry. Cytometry A 95:422–426. https://doi.org/10.1002/cyto.a.23725
Iyer A, Hamers AAJ, Pillai AB (2022) CyTOF® for the masses. Front Immunol 13:815828. https://doi.org/10.3389/fimmu.2022.815828
Chudakov DM, Matz MV, Lukyanov S, Lukyanov KA (2010) Fluorescent proteins and their applications in imaging living cells and tissues. Physiol Rev 90:1103–1163. https://doi.org/10.1152/physrev.00038.2009
Seong Y, Nguyen DX, Wu Y, Thakur A, Harding F, Nguyen TA (2022) Novel PE and APC tandems: additional near-infrared fluorochromes for use in spectral flow cytometry. Cytometry A:1–11. Epub 2 Feb 2022. https://doi.org/10.1002/cyto.a.24537
Riggs JR, Medina EM, Perrenoud LJ, Bonilla DL, Clambey ET, van Dyk LF, Berg LJ (2021) Optimized detection of acute MHV68 infection with a reporter system identifies large peritoneal macrophages as a dominant target of primary infection. Front Microbiol 12:656979. https://doi.org/10.3389/fmicb.2021.656979
Peixoto MM, Soares-da-Silva F, Schmutz S, Mailhe M-P, Novault S, Cumano A, Ait-Mansour C (2022) Identification of fetal liver stromal subsets in spectral cytometry using the parameter autofluorescence. Cytometry A. Epub 2 May 2022. https://doi.org/10.1002/cyto.a.24567
Schmutz S, Valente M, Cumano A, Novault S (2016) Spectral cytometry has unique properties allowing multicolor analysis of cell suspensions isolated from solid tissues. PLoS One 11:e0159961. https://doi.org/10.1371/journal.pone.0159961
Solomon M, DeLay M, Reynaud D (2020) Phenotypic analysis of the mouse hematopoietic hierarchy using spectral cytometry: from stem cell subsets to early progenitor compartments. Cytometry A 97:1057–1065. https://doi.org/10.1002/cyto.a.24041
Park LM, Lannigan J, Jaimes MC (2020) Omip-069: forty-color full spectrum flow cytometry panel for deep immunophenotyping of major cell subsets in human peripheral blood. Cytometry A 97:1044–1051. https://doi.org/10.1002/cyto.a.24213
Chen M, Wang H, Fu M et al (2020) One tube 24 color full spectral flow cytometry and multi-dimensional software to study the maturation pattern and antigen expression of the myeloid. Blood 136:13–14. https://doi.org/10.1182/blood-2020-140600
Murphy KA, Bhamidipati K, Rubin SJS et al (2019) Immunomodulatory receptors are differentially expressed in B and T cell subsets relevant to autoimmune disease. Clin Immunol 209:108276. https://doi.org/10.1016/j.clim.2019.108276
Chen MC, Lai KS, Chien KL, Teng ST, Lyn YR et al (2021) Impact of placenta-derived mesenchymal stem cells treatment on patients with severe lung injury caused by COVID-19 pneumonia: clinical and immunological aspect. Res Square. Preprint. https://doi.org/10.21203/rs.3.rs-1013382/v1
Park LM, Lannigan J, Jaimes MC (2020) Forty-color full spectrum flow cytometry panel for deep immunophenotyping of major cell subsets in human peripheral blood. Cytometry A 97A:1044–1051. https://doi.org/10.1002/cyto.a.24213
Lippitsch A, Chukovetskyi Y, Baal N et al (2017) Unique high and homogenous surface expression of the transferrin receptor CD71 on murine plasmacytoid dendritic cells in different tissues. Cell Immunol 316:41–52. https://doi.org/10.1016/j.cellimm.2017.03.005
Rousseau M, Goh HMS, Holec S et al (2016) Bladder catheterization increases susceptibility to infection that can be prevented by prophylactic antibiotic treatment. JCI Insight 1:e88178. https://doi.org/10.1172/jci.insight.88178
Affandi AJ, Grabowska J, Olesek K et al (2020) Selective tumor antigen vaccine delivery to human CD169+antigen-presenting cells using ganglioside-liposomes. Proc Natl Acad Sci U S A 117:27528–27539. https://doi.org/10.1073/pnas.2006186117
Amor C, Feucht J, Leibold J et al (2020) Senolytic car T cells reverse senescence-associated pathologies. Nature 583:127–132. https://doi.org/10.1038/s41586-020-2403-9
Turner JS, Zhou JQ, Han J et al (2020) Human germinal centres engage memory and naive B cells after influenza vaccination. Nature 586:127–132. https://doi.org/10.1038/s41586-020-2711-0
Mudd PA, Crawford JC, Turner JS et al (2020) Distinct inflammatory profiles distinguish COVID-19 from influenza with limited contributions from cytokine storm. Sci Adv 6:eabe3024. https://doi.org/10.1126/sciadv.abe3024
Fox A, Dutt TS, Karger B, Rojas M, Obregon-Henao A et al (2020) CYTO-feature engineering: a pipeline for flow cytometry analysis to uncover immune populations and associations with disease. Sci Rep 10:7651. https://doi.org/10.1038/s41598-020-64516-0
Ferrer-Font L, Mayer JU, Old S et al (2020) High-dimensional data analysis algorithms yield comparable results for mass cytometry and spectral flow cytometry data. Cytometry A 97:824–831. https://doi.org/10.1002/cyto.a.24016
Boss AL, Brooks AES, Chamley LW, James JL (2020) Influence of culture media on the derivation and phenotype of fetal-derived placental mesenchymal stem/stromal cells across gestation. Placenta 101:66–74. https://doi.org/10.1016/j.placenta.2020.09.002
Henderson J, Havranek O, Ma MCJ, Herman V, Kupcova K, Chrbolkova T et al (2021) Detecting Fӧrster resonance energy transfer in living cells by conventional and spectral cytometry. Cytometry A:1–17. https://doi.org/10.1002/cyto.a.24472
Boss AL. Damani T, Chamley LW, James JL, Brooks AES (2021) The origins of placental mesenchymal stromal: full spectrum flow cytometry reveals mesencgymal heterogeneity in first trimester placentae, and phenotypic convergence in culture. BioRxiv. Preprint. https://doi.org/10.1101/2021.12.21.473551
Fernandez MA, Alzayat H, Jaimes MC, Kharraz Y, Requena G, Mendez P (2022) High-dimensional immunophenotyping with 37-colors panel using full-spectrum cytometry. Methods Mol Biol 2386:43–60. https://doi.org/10.1007/978-1-0716-1771-7_4
Suda M, Shimizu I, Katsuumi G et al (2021) Senolytic vaccination improves normal and pathological age-related phenotypes and increases lifespan in progeroid mice. Nat Aging 1:1117–1126. https://doi.org/10.1038/s43587-021-00151-2
Costa B, Becker J, Krammer T, Mulenge F, Duran V et al (2022) HCMV exploits STING signaling and counteracts IFN and ISG induction to facilitate dendritic cell infection. Res Square. Preprint, https://doi.org/10.21203/rs.3.rs-953016/v1
Telford WG, Shcherbakova DM, Buschke D et al (2015) Multiparametric flow cytometry using near-infrared fluorescent proteins engineered from bacterial phytochromes. PLoS One 10:e0122342. https://doi.org/10.1371/journal.pone.0122342
Wanner N, Barnhart J, Apostolakis N, Zlojutro V, Asosingh K (2022) Using the autofluorescence finder on the Sony ID7000TM spectral cell analyzer to identify and unmix multiple highly autofluorescent murine lung populations. Front Bioeng Biotechnol 10:827987. https://doi.org/10.3389/fbioe.2022.827987. eCollection 2022
Dashkova V, Segev E, Malashenkov D et al (2016) Microalgal cytometric analysis in the presence of endogenous autofluorescent pigments. Algal Res 19:370–380. https://doi.org/10.1016/j.algal.2016.05.013
Barteneva NS, Dashkova V, Vorobjev I (2019) Probing complexity of microalgae mixtures with novel spectral flow cytometry approach and “virtual filtering”. BioRxiv. Preprint. https://doi.org/10.1101/516146
Qiu P, Simonds EF, Bendall SC, Gibbs KD Jr, Bruggner RV et al (2011) Extracting a cellular hierarchy from high-dimensional cytometry data with SPADE. Nat Biotechnol 29:886–891. https://doi.org/10.1038/nbt.1991
Amir ED, Davis KL, Tadmor MD, Simonds EF, Levine JH et al (2013) viSNE enables visualization of high-dimensional single-cell data and reveals phenotypic heterogeneity of leukemia. Nat Biotechnol 31:545–552. https://doi.org/10.1038/nbt.2594
Van Gassen S, Callebaut B, Van Helden MJ, Lambrecht BN, Demeester P, Dhaene T, Sayes Y (2015) FlowSOM: using self-organizing maps for visualization and interpretation of cytometry data. Cytometry A 87:636–645. https://doi.org/10.1002/cyto.a.22625
Nowicka M, Krieg C, Weber LM, Hartmann FJ, Guglietta S et al (2017) CyTOF workflow: differential discovery in high-throughput high-dimensional cytometry datasets. F1000Res 6:748. https://doi.org/10.12688/f1000research.11622.2
Bruggner RV, Bodenmiller B, Dill DL, Tibshirani RJ, Nolan GP (2014) Automated identification of stratifying signatures in cellular subpopulations. Proc Natl Acad Sci USA 111:E2770–E2777
Ogishi M, Yang R, Gruber C, Zhang SJ, Pelham S, Spaan AN et al (2020) Multi-batch cytometry data integration for optimal immunophenotyping. J Immunol 206:206–213. https://doi.org/10.1101/2020.07.14.202432
Acknowledgments
Work was supported by the Ministry of Health of the Republic of Kazakhstan under the program-targeted funding of the Ageing and Healthy Lifespan research program (IRN: 51760/ПЦФ-МЗ РК-19) and AP08857554 (Ministry of Education and Science, Kazakhstan) to IAV. NSB was funded by CRP 16482715 and SSH2020028 grants from Nazarbayev University, and AP14872088 MES grant (Kazakhstan).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Copyright information
© 2023 The Author(s)
About this protocol
Cite this protocol
Vorobjev, I.A., Kussanova, A., Barteneva, N.S. (2023). Development of Spectral Imaging Cytometry. In: Barteneva, N.S., Vorobjev, I.A. (eds) Spectral and Imaging Cytometry. Methods in Molecular Biology, vol 2635. Humana, New York, NY. https://doi.org/10.1007/978-1-0716-3020-4_1
Download citation
DOI: https://doi.org/10.1007/978-1-0716-3020-4_1
Published:
Publisher Name: Humana, New York, NY
Print ISBN: 978-1-0716-3019-8
Online ISBN: 978-1-0716-3020-4
eBook Packages: Springer Protocols