Field-Flow Fractionation in Molecular Biology and Biotechnology

Abstract

Field-flow fractionation (FFF) is a family of single-phase separative techniques exploited to gently separate and characterize nano- and microsystems in suspension. These techniques cover an extremely wide dynamic range and are able to separate analytes in an interval between a few nm to 100 µm size-wise (over 15 orders of magnitude mass-wise). They are flexible in terms of mobile phase and can separate the analytes in native conditions, preserving their original structures/properties as much as possible. Molecular biology is the branch of biology that studies the molecular basis of biological activity, while biotechnology deals with the technological applications of biology. The areas where biotechnologies are required include industrial, agri-food, environmental, and pharmaceutical. Many species of biological interest belong to the operational range of FFF techniques, and their application to the analysis of such samples has steadily grown in the last 30 years. This work aims to summarize the main features, milestones, and results provided by the application of FFF in the field of molecular biology and biotechnology, with a focus on the years from 2000 to 2022. After a theoretical background overview of FFF and its methodologies, the results are reported based on the nature of the samples analyzed.

Keywords: asymmetrical flow field-flow fractionation (AF4); bio-nanoparticles; cell sorting; laser scattering; molecular biology; native separation; pharmaceutics; separation science.

Figure 1

Schematic representation of an FFF platform. The separative channel is a schematization of the typical FFF channel. The dotted box reports the possible fields that have been theoretically studied to be exploited for separation in FFF. Only the ones in bold are, nowadays, significantly used.

Figure 2

Schematization of the evolution modes in FFF. (A). Normal mode (B). Steric mode (C). Hyperlayer mode. Image adapted from [5].

Figure 3

Main applications of FFF-MS online coupling.

Figure 4

Schematization of a coupled offline FFF platform.

Figure 5

AF4-UV-MALS analysis of undiluted Victoza (6 mg/mL liraglutide) and 10-times diluted Victoza (dotted peaks). Different injection volumes were tested. All analyses showed a single peak in the UV chromatogram and a uniform molar mass of ~22 kDa across the peak. Different colors refer to different injection amounts. Image adapted from [89].

Figure 6

Application of an HF5-based platform as a tool for evaluating the selective interactions of components (heme) in a complex matrix (whole serum). (A). The fractogram shows the profiles of separated serum components at two wavelengths: 280 nm (protein specific) and 405 nm (heme-specific). (B). The 405 nm signal, registered after spiking serum with heme, simulating hemolysis, suggests the preferential binding of heme to HSA, Hd, and Hx. Binding to IgA and IgM has also been observed. Image adapted from [98].

Figure 7

A comparative study of the selective aptamer (A80R and A40)–lysozyme (egg white) interactions exploiting an AF4-UV platform. (A). AF4-UV fractograms of A80R and A40 and their mixtures with lysozyme. The dashed line represents the retention time at which that signal intensity is recorded to evaluate the signal decrease in free aptamer, correlated to the formation of a complex. (B). Percentage of bound aptamer (expressed as % (I_{apt alone}−  I_{apt mix})/I_{apt alone}) for each aptamer at different lysozyme concentrations. (C). Percentage of bound aptamer to BSA (interfering agent) at different BSA concentrations. Overall, boot A80R shows better binding properties to the lysozyme, and it is less affected by the presence of BSA compared to A40. Image adapted from [122] and published with permission.

Figure 8

An example of the outstanding potential of offline AF4-nUHPLC to the lipidomic analysis of lipoproteins, which allowed the recognition of 363 lipidic species. PCA (A) and Vulcano (A′) analysis allowed a fast preliminary identification of lipids to differentiate the samples. (B,B′) represent absolute and percent composition of the lipids of eight different classes in the MCI, AD, and control samples. More than twofold differences in abundance were observed for certain lipidic classes when comparing the LDL/VLDL of the AD and MCI to the control ones. (C) Represents the correlation between the abundance of three prominent lipid species in HDLs and LDLs/VLDLs with certain diseases. (C′) Represents the ROC curves of the four candidate markers: two in MCI in combination with MMSE scores (solid line) superimposed with those of MMSE scores alone (dotted line) [148].

Figure 9

(a) SdFFF sorting scheme of glioblastoma cell lines: F1 = subpopulations with specific differentiation characteristics; F2 = cancer stem cells (CSCs) and high-frequency crossover (HFC) values. (b) Functional characterization of F1, F2, total peak collected (TP), and unfractionated cells (crude). *** p < 0.0001 (ANOVA). (A) Cell cycle analysis by DNA content measurement: CSCs (F2) are found in the G1 phase (quiescent cells). (B) Soft agar assay for colony formation examination: the F2 subpopulation consists of a population of cells enriched in CSCs, whereas the F1 subpopulation is enriched in differentiated cells. * p < 0.05; ** p < 0.001 (Student’s t test). Images adapted from [242].

Figure 10

(a) A schematic representation of Celector^®: cells are injected into the inlet of the separation channel filled with mobile phase and eluted through it, acquiring different velocities related to their physical properties. A camera connected to the imaging software that plots the number of counted cells vs. time (fractogram) visualizes cells. Finally, cells are collected at the outlet and divided into different tubes according to the sample’s fractogram. Images are adapted from [249] and published with permission. (b) Representative images of a successful isolation protocol of amniotic epithelial cells (AECs) (Type 1) and an unsuccessful (Type 2) protocol. The profile represents the number of cells versus time of analysis (A,C) and collected subpopulations F1 and F2; live images of eluting cells (B,D); cell distribution between F1 and F2 based on the calculation of the area under the curve (AUC) expressed as a percentage compared to the total area of the profile (E); distribution was also expressed as a number of counted cells by the software for each fraction of all samples analyzed (F). (t-test: p < 0.05 *.) Images are adapted from [249,250] and published with permission.

Figure 11

An example of an HF5 platform coupled offline with a MALDI/TOF detection system used to separate and characterize different cell subtypes. (A). Characteristic MALDI/TOF MS peaks of E. coli (§, green) and B. subtilis (#, red). (B). A MALDI/TOF MS spectrum of a mixture of the two species without HF5 separation. Five characteristic peaks were identified for boot species. (C). A fractogram of the same mixture separated through HF5. Two fractions were collected (namely 1 and 2). (D,D′) A MALDI/TOF MS spectrum of the collected fractions after a concentration process. The comparison with the characteristic peaks for both species allowed the identification of Fraction 1 as B. subtilis and Fraction 2 as E. coli. A higher number of characteristic peaks was observed for both species after HF5 separation. Images are adapted from [267].

Figure 12

An indicative number of documents (review, research articles) published in the XXI century related to field-flow fractionation. The numbers are derived from www.sciencedirect.com (accessed on 15 September 2022) using “FFF” and “Field Flow Fractionation” as keywords.