Elsevier

Talanta

Volume 221, 1 January 2021, 121442
Talanta

Multivariate analysis of mean Raman spectra of erythrocytes for a fast analysis of the biochemical signature of ageing

https://doi.org/10.1016/j.talanta.2020.121442Get rights and content

Highlights

  • Raman mapping is exploited to collect several single-cell mean spectra of dehydrated red blood cells (RBC).

  • PCA is used to rapidly identify subtle spectral differences related to conformational and biochemical changes due to ageing.

  • A two-step ageing process is demonstrated, associated to different roles of proteins and lipids.

  • Relevant spectral effects ascribed to reduction in Hb oxygenation and membrane fluidity due to ageing are also evidenced.

  • The method allows to simply and rapidly classify RBC based on age and morphology, thus assessing the quality of blood.

Abstract

Ageing of red blood cells (RBC) is a physiological process, fundamental to ensure a proper blood homeostasis that, in vivo, balances the production of new cells and the removal of senescent erythrocytes. A detailed characterization at the cellular level of the progression of the ageing phenomenon can reveal biological, biophysical and biochemical fingerprints for diseases related to misbalances of the cell turnover and for blood pathologies. We applied Principal Components Analysis (PCA) to mean Raman spectra of single cells at different ageing times to rapidly highlight subtle spectral differences associated with conformational and biochemical modifications. Our results demonstrate a two-step ageing process characterized by a first phase in which proteins plays a relevant role, followed by a further cellular evolution driven by alterations in the membrane lipid contribution. Moreover, we used the same approach to directly analyse relevant spectral effects associated to reduction in Haemoglobin oxygenation level and membrane fluidity induced by the ageing. The method is robust and effective, allowing to classify easily the studied cells based on their age and morphology, and consequently to evaluate the biological quality of a blood sample.

Introduction

In the decades since 1930, when Raman and Landsteiner together collected their Nobel prize for, respectively, the discovery of the expressly-named Raman effect and of human blood groups, many investigations mixing the two findings have been carried out. Consequently, an increasing number of related analytical and diagnostic applications have been proposed and successfully demonstrated. Raman spectroscopy has now proved to be a powerful tool both for the study of whole blood and of its cellular components [[1], [2], [3], [4]]. The first Raman studies on the structure and vibrational properties of isolated Haemoglobin (Hb) dates back to 1970s [5] and were mainly dedicated to the investigation of different signatures of its oxygenated and deoxygenated states. These results have been fundamental in successive analyses of intact and alive erythrocytes [6], which are substantially constituted of Hb and enzymes surrounded by a cellular membrane containing many transmembrane proteins that anchor the supporting cytoskeleton, in a very peculiar architecture called membrane-skeleton [7]. Raman studies on RBCs membrane alone have also been performed by means of artificially isolated “ghosts” and have mainly been devoted to investigate their composition (lipid and protein contribution and conformation) and their fluidity properties [8]. All the information acquired on oxy/deoxy states of Hb and on membrane flexibility proved important to study the cell ageing and to deepen the knowledge on degradation and damaging processes of erythrocytes [9]. More in general, the mechanical, structural and biochemical modifications at the basis of the ageing process of RBCs are important in the study of specific pathologies [10,11] and to analyse the effects of blood storage for transfusion [12,13].

Very recently, many important papers have been published demonstrating the potentiality of Raman spectroscopy as a non-invasive tool for stored red blood cell analysis focusing on mean effects of haemolysis [14,15], variation of Hb oxygenation [16,17] and age-related changes in supernatant composition, with specific reference to lactate accumulation [18]. In fact, these studies have been performed on dried drops of supernatant [14,18], dry-fixed smears [16] and also directly through sealed transfusion bags, in order to preserve storage sterility and to classify blood quality and its correlation with the time of storage in blood bank [15,17].

In this regard, the ability of Raman microscopy to perform rapid and label-free characterization and spatial mapping of samples appears particularly attractive and already allowed to gain valuable insight into the differences between “old” and “young” erythrocytes at the single cell level [19,20]. As an example, it was demonstrated that in aged RBC, Hb tends to aggregate, partially loses its oxygenation ability and speed and is not homogeneously distributed across the cell, but it is mainly bound to the plasma membrane [19]. This uneven distribution, possibly contributes to reduce the cell's elasticity and deformability. It was also suggested that this behaviour of the Hb, triggered by the ageing process and correlated with the increase of cell density, can produce structural changes and higher exposition of tyrosine residue [13], leading to an increase in the globin-to-heme separation and resulting in a final breakdown of the globin [12]. The ageing pattern also produces serious alterations at the level of the plasma membrane. The total content of sialic acid of old RBCs is expected to be lower than in young cells [[21], [22], [23]], while alterations in the membrane structure and function have been observed and hypothesized to be preceded by effects on the protein components [12]. However, to date, the correlation between modifications of cell properties and ageing is not completely clear and most of the reported Raman results are based on the identification of small intensity differences between selected bands.

Here, we applied Raman imaging to collect a large number of mean spectra of single dried cells at different ageing times. The mean cell spectra were, then, collectively analysed by means of Principal Component Analysis (PCA). This approach allowed to highlight rapidly the main spectral differences associated with conformational and biochemical modifications induced by the ageing process. In particular, we identified a time threshold between two different phases of the process: a first period in which proteins play a relevant role, followed by a second phase driven by alterations in the membrane lipid contribution. Finally, we used the same approach to analyse the relevant spectral effects associated with the reduction in cell oxygenation level and membrane fluidity induced by the ageing. The method and the protocols here defined, which have been applied to RBCs from two different donors to underline their robustness and effectiveness, were studied using the Quadratic Discriminant Analysis (QDA). This provided a quantitative classification of the studied cells on the basis of their age and morphology, which can be used for a rapid evaluation of the quality of blood samples.

Section snippets

Sample preparations

The blood samples have been obtained, for the purpose of scientific research only, from healthy donors after written informed consensus, as required by Italian “National Bioethics Committee”. All the preparation methods and sample manipulations have been performed in accordance to the relevant guidelines and regulations approved by the “Research Ethics and Bioethics Committee” of the Italian CNR.

Blood samples were obtained using venipuncture into Vacutainers (Becton-Dickinson, Franklin Lakes,

Results and discussion

The Raman microscope allowed to collect several optical images and spectral maps on the smears prepared, in the presence of plasma, at different ageing times. Fig. 1 presents an optical image and the correlated spectral map of four different cell morphologies (respectively, from top left to bottom down in the image, spiculed, biconcave, flat and crenated). During the ageing pathway the abundance of these four morphologic phenotypes changes, with each form that is predominant at a different time

Conclusions

We applied Principal Components Analysis (PCA) to mean vibrational spectra of single erythrocytes obtained by Raman mapping to rapidly evidence specific spectral differences associated to the progression of cell ageing. This is a fundamental process for the erythrocytes that, both in vivo and in vitro, is accompanied by the development of severe morphological alterations that are driven by, or associated to, chemical modifications of the cell architecture.

In this framework, our results

Author contribution

E. L. realized imaging experiments, performed multivariate analysis of spectroscopic data and contributed to data analysis and to writing the paper (Investigation, Formal Analysis, Writing); S. D., G. L. and M. G. prepared samples and contributed to data analysis and to writing the paper (Resources, Methodology, Formal Analysis, Writing); V. M. designed research, performed experiments, contributed to data analysis and to writing the paper (Conceptualization, Validation, Investigation, Formal

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (42)

  • W. Schmidt et al.

    Red cell age effects on metabolism and oxygen affinity in humans

    Respir. Physiol.

    (1987)
  • C.G. Atkins et al.

    Raman spectroscopy of blood and blood components

    Appl. Spectrosc.

    (2017)
  • A.J. Berger et al.

    Multicomponent blood analysis by near-infrared Raman spectroscopy

    Appl. Optic.

    (1999)
  • S. Raj et al.

    Mechanochemistry of single red blood cells monitored using Raman tweezers

    Biomed. Optic Express

    (2012)
  • E. Missov et al.

    Circulating cardiac troponin I in severe congestive heart failure

    Circulation

    (1997)
  • R. Petry et al.

    Raman spectroscopy--a prospective tool in the life sciences

    ChemPhysChem : a European journal of chemical physics and physical chemistry

    (2003)
  • S. Suresh

    Mechanical response of human red blood cells in health and disease: some structue-property-function relationships

    J. Metr. Res.

    (2006)
  • Y.X. Huang et al.

    Human red blood cell aging: correlative changes in surface charge and cell properties

    J. Cell Mol. Med.

    (2011)
  • R. Gautam et al.

    Characterization of storage-induced red blood cell hemolysis using Raman spectroscopy

    Lab. Med.

    (2018)
  • R. Gautam et al.

    Non-invasive analysis of stored red blood cells using diffuse resonance Raman spectroscopy

    Analyst

    (2018)
  • C.G. Atkins et al.

    Using Raman spectroscopy to assess hemoglobin oxygenation in red blood cell concentrate: an objective proxy for morphological index to gauge the quality of stored blood?

    Analyst

    (2017)
  • Cited by (5)

    • Anchoring red blood cell with tetrahedral DNA nanostructure: Electrochemical biosensor for the sensitive signage of circulating tumor DNA

      2023, Talanta
      Citation Excerpt :

      Therefore, the red blood cells were introduced. As we know that the red blood cells widely existed in whole blood, their low cytotoxicity and biocompatibility can make red blood cells as suitable candidate for the part of an electrochemical biosensor [20,21], which can detect ctDNA in blood without introducing foreign substances, and reduce the occurrence of false positive results. Therefore, the combination of red blood cells with TDN in electrochemical biosensors is suitable for detecting ctDNA in blood.

    • Trends in biomedical analysis of red blood cells – Raman spectroscopy against other spectroscopic, microscopic and classical techniques

      2022, TrAC - Trends in Analytical Chemistry
      Citation Excerpt :

      Inclusion of such procedure into Raman spectroscopy studies was reported for the first time in 1975 by Lippert et al., who managed to estimate the configuration of membrane proteins (mainly α-helices), as well as configuration of hydrophobic side chains of phospholipids (all-trans) [191]. More recently, such procedure was applied to investigate RBC membrane alterations connected with aging, hypercholesterolemia and atherosclerosis reflected by variations in lipid-related fractions, i.e., the amount of phospholipids, cholesterol esters, esterified lipids, total lipid content and lipid unsaturation level [1,115,193,194]. Another approach to obtain information about RBC membrane and membrane-bounded Hb remains optical trapping, e.g., using vortex beam [195].

    View full text