CoChat

Organs on chip approach: a tool to evaluate cancer -immune cells interactions

Open access
Biselli, Elena, Agliari, Elena, Barra, Adriano, Bertani, Francesca Romana, Gerardino, Annamaria, De Ninno, Adele, Mencattini, Arianna, Di Giuseppe, Davide, Mattei, Fabrizio, Schiavoni, Giovanna, Lucarini, Valeria, Vacchelli, Erika, Kroemer, Guido, Di Natale, Corrado, Martinelli, Eugenio, Businaro, Luca · 2017 · Scientific Reports

The DOI matches its Crossref record; 37 of the 37 references it cites were found in a registry or an index.

Abstract

In this paper we discuss the applicability of numerical descriptors and statistical physics concepts to characterize complex biological systems observed at microscopic level through organ on chip approach. To this end, we employ data collected on a microfluidic platform in which leukocytes can move through suitably built channels toward their target. Leukocyte behavior is recorded by standard time lapse imaging. In particular, we analyze three groups of human peripheral blood mononuclear cells (PBMC): heterozygous mutants (in which only one copy of the FPR1 gene is normal), homozygous mutants (in which both alleles encoding FPR1 are loss-of-function variants) and cells from ‘wild type’ donors (with normal expression of FPR1). We characterize the migration of these cells providing a quantitative confirmation of the essential role of FPR1 in cancer chemotherapy response. Indeed wild type PBMC perform biased random walks toward chemotherapy-treated cancer cells establishing persistent interactions with them. Conversely, heterozygous mutants present a weaker bias in their motion and homozygous mutants perform rather uncorrelated random walks, both failing to engage with their targets. We next focus on wild type cells and study the interactions of leukocytes with cancerous cells developing a novel heuristic procedure, inspired by Lyapunov stability in dynamical systems.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41598-017-13070-3

Cites

37 of 37 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 The immune-oncology chip hosting the experiments and track examples. Panel A) shows a general scheme of the device, composed by six reservoirs for cell loading and culture medium replacement and four chambers (or compartments) for cell culture. Panel B) presents a detailed view of the four chambers. The left chamber is dedicated to TCs culture, while PBMC, initially loaded in the right chamber, passively move in the two central chambers where they stop until sensing the chemo-attractant signals from treated TCs, thus starting to migrate towards them through the micro-channels array connecting the two culture chambers. A picture of the whole device is presented in panel C). In the lower part of the figure, we show examples of the typical responses performed by PBMCs in the three different settings. In each panel the reference system has the origin on the bottom left. The same reference system will be used throughout the data analysis. Panel D) shows the trajectories of pure FPR1 CC cells, able to migrate towards cancerous cells, which are loaded in the reservoir (not shown) at the bottom right of the channel; panel E) reports the trajectories of FPR1 CA cells, characterized by a weaker drift in the motion towards the target; panel F) shows the prevalently erratic behaviors of FPR1 AA cells. In all the three cases the black spots represent TCs.
Figure 1 The immune-oncology chip hosting the experiments and track examples. Panel A) shows a general scheme of the device, composed by six reservoirs for cell loading and culture medium replacement and four chambers (or compartments) for cell culture. Panel B) presents a detailed view of the four chambers. The left chamber is dedicated to TCs culture, while PBMC, initially loaded in the right chamber, passively move in the two central chambers where they stop until sensing the chemo-attractant signals from treated TCs, thus starting to migrate towards them through the micro-channels array connecting the two culture chambers. A picture of the whole device is presented in panel C). In the lower part of the figure, we show examples of the typical responses performed by PBMCs in the three different settings. In each panel the reference system has the origin on the bottom left. The same reference system will be used throughout the data analysis. Panel D) shows the trajectories of pure FPR1 CC cells, able to migrate towards cancerous cells, which are loaded in the reservoir (not shown) at the bottom right of the channel; panel E) reports the trajectories of FPR1 CA cells, characterized by a weaker drift in the motion towards the target; panel F) shows the prevalently erratic behaviors of FPR1 AA cells. In all the three cases the black spots represent TCs.CC BY · PubMed Central
Figure 2 Distributions of step length for CC cells. For each CC cell we evaluate the extent of the motion performed during the time lapse between two consecutive measurements. This can be quantified in terms of displacement along the x and the y direction and the related step lengths are referred to as Δx and Δy. By merging experimental data from the whole set of tracks we build the distributions P(Δx) and P(Δy) shown in these panels in a semi-logarithmic scale, where experimental data are reported with standard errors. Neglecting the tails of the distributions (for which the statistics is rather poor), we can fit these distributions according to exponential and Gaussian functions, as reported in the panels together with the related fit coefficients. The motion resulting from these distributions displays a drift along positive x and negative y.
Figure 2 Distributions of step length for CC cells. For each CC cell we evaluate the extent of the motion performed during the time lapse between two consecutive measurements. This can be quantified in terms of displacement along the x and the y direction and the related step lengths are referred to as Δx and Δy. By merging experimental data from the whole set of tracks we build the distributions P(Δx) and P(Δy) shown in these panels in a semi-logarithmic scale, where experimental data are reported with standard errors. Neglecting the tails of the distributions (for which the statistics is rather poor), we can fit these distributions according to exponential and Gaussian functions, as reported in the panels together with the related fit coefficients. The motion resulting from these distributions displays a drift along positive x and negative y.CC BY · PubMed Central
Figure 3 Scatter plots for step length along different directions. For each panel, each data point represents the step length along x and y direction performed by a cell during the time lapse between two consecutive measurements. Left panel: CC cells; central panel: CA cells; right panel: AA cells. Note that the clouds of data lack any internal structure of polarization.
Figure 3 Scatter plots for step length along different directions. For each panel, each data point represents the step length along x and y direction performed by a cell during the time lapse between two consecutive measurements. Left panel: CC cells; central panel: CA cells; right panel: AA cells. Note that the clouds of data lack any internal structure of polarization.CC BY · PubMed Central
Figure 4 Polar histograms of the turning angle described by cells during their motion. For each cell we evaluate the turning angle Δθ during the time lapse between two consecutive measurements and we analyze separately data pertaining to CC cells (left panel), to CA cell (central panel) and to AA cells (right panel). For each set, the raw values of the turning angle are grouped in bins, according to their numeric range. Notice that: in the CC case and in the CA case cells tend to maintain the same direction also if the frequency of steps preserving the direction of the motion is 2.5 times smaller for CA and the distribution is broader; in the AA case the distribution has almost zero mean and cells do not have a preferred direction.
Figure 4 Polar histograms of the turning angle described by cells during their motion. For each cell we evaluate the turning angle Δθ during the time lapse between two consecutive measurements and we analyze separately data pertaining to CC cells (left panel), to CA cell (central panel) and to AA cells (right panel). For each set, the raw values of the turning angle are grouped in bins, according to their numeric range. Notice that: in the CC case and in the CA case cells tend to maintain the same direction also if the frequency of steps preserving the direction of the motion is 2.5 times smaller for CA and the distribution is broader; in the AA case the distribution has almost zero mean and cells do not have a preferred direction.CC BY · PubMed Central
Figure 5 Histograms of the magnitude of the velocity and of its Cartesian and polar components for the PBMC distance from TC. For each TC shown in Fig. 1, we select the trajectories of the PBMC that reach a distance from TC shorter than the physical interaction range, evaluated as the sum of PBMC and TC mean diameters, and we calculate the instantaneous velocity v(t) for all of them. The selected setting is novel but has been chosen in line with standard situations found in the literature. Panels A) and B) show the histograms of the magnitude of the velocity of each trajectory of the PBMC that move towards the selected TCs at the time step t = 0 (upper panel) and at the time step t = 30 (lower panel). As the trajectories are studied backward (such that when their collapse to the tumor cell, it is t = 0, while their initial positions, far from the tumor, belong to timeframe 30s), note that for t = 0 the mean velocity is peaked at 0, while for t = 30 the peak is shifted versus higher speeds (≈2.0 ± 0.8 μ m/min). Panels C)-F) show the histograms of the Cartesian components of the velocity at the time step t = 0 (upper panels) and at the time step t = 30 (lower panels). Note that, especially for t = 30, the distribution appears not symmetric, since vy does not change over time because the drift of PBMC motions towards the reservoir has a stronger bias along x rather than along y. Panels G)-L) show the histograms of the polar components of the velocity at the time step t = 0 (upper panels) and at the time step t = 30 (lower panels). Note that the angular velocity is always peaked at zero, as expected for quite smooth trajectories which point toward the tumor position, while the radial velocity is much spread at long times, when the interaction with the tumor is still weak. All the histograms are normalized by scaling between 0 and 1.
Figure 5 Histograms of the magnitude of the velocity and of its Cartesian and polar components for the PBMC distance from TC. For each TC shown in Fig. 1, we select the trajectories of the PBMC that reach a distance from TC shorter than the physical interaction range, evaluated as the sum of PBMC and TC mean diameters, and we calculate the instantaneous velocity v(t) for all of them. The selected setting is novel but has been chosen in line with standard situations found in the literature. Panels A) and B) show the histograms of the magnitude of the velocity of each trajectory of the PBMC that move towards the selected TCs at the time step t = 0 (upper panel) and at the time step t = 30 (lower panel). As the trajectories are studied backward (such that when their collapse to the tumor cell, it is t = 0, while their initial positions, far from the tumor, belong to timeframe 30s), note that for t = 0 the mean velocity is peaked at 0, while for t = 30 the peak is shifted versus higher speeds (≈2.0 ± 0.8 μ m/min). Panels C)-F) show the histograms of the Cartesian components of the velocity at the time step t = 0 (upper panels) and at the time step t = 30 (lower panels). Note that, especially for t = 30, the distribution appears not symmetric, since vy does not change over time because the drift of PBMC motions towards the reservoir has a stronger bias along x rather than along y. Panels G)-L) show the histograms of the polar components of the velocity at the time step t = 0 (upper panels) and at the time step t = 30 (lower panels). Note that the angular velocity is always peaked at zero, as expected for quite smooth trajectories which point toward the tumor position, while the radial velocity is much spread at long times, when the interaction with the tumor is still weak. All the histograms are normalized by scaling between 0 and 1.CC BY · PubMed Central
Figure 6 Upper panel: representative picture of the landscape generated by a tumor cell (i.e., the attractor) and possible dynamics of the immune cell. PBMC, attracted by the target TC, collapses in the minimum of the linearization range inside the interaction range. In the case of a stable trajectory (associated to the red sphere), any small perturbation at time 0 will not cause the escape of the particle from the minimum at time 1. On the contrary, metastable trajectories (associated to the blue sphere) are sensitive to small perturbations and the particle can easily escape from the local equilibrium. Central panel: plots of different possible immune cells dynamics. The panels resume the three different possible outcomes of the PBMC convergence to the TC attractor (stable, unstable and metastable dynamics), where the eigenvectors, labeled by λ, are the natural base built as a linear combination of λ1 and λ2. Bottom panel: real tracks of immune cells attracted by one tumor cell. The position of the tumor cell corresponds to the center of the black circle. The radius of the dashed circle represents the physical interaction range given by the sum of the mean diameter of PBMC (∼9 μm) and TC (∼20 μm). The radius of the solid circle represents the interaction range of the selected TC, (∼50 μm). The colored lines represent the tracks of the immune cells that approach the tumor cell considered. Each color corresponds to a different immune cell.
Figure 6 Upper panel: representative picture of the landscape generated by a tumor cell (i.e., the attractor) and possible dynamics of the immune cell. PBMC, attracted by the target TC, collapses in the minimum of the linearization range inside the interaction range. In the case of a stable trajectory (associated to the red sphere), any small perturbation at time 0 will not cause the escape of the particle from the minimum at time 1. On the contrary, metastable trajectories (associated to the blue sphere) are sensitive to small perturbations and the particle can easily escape from the local equilibrium. Central panel: plots of different possible immune cells dynamics. The panels resume the three different possible outcomes of the PBMC convergence to the TC attractor (stable, unstable and metastable dynamics), where the eigenvectors, labeled by λ, are the natural base built as a linear combination of λ1 and λ2. Bottom panel: real tracks of immune cells attracted by one tumor cell. The position of the tumor cell corresponds to the center of the black circle. The radius of the dashed circle represents the physical interaction range given by the sum of the mean diameter of PBMC (∼9 μm) and TC (∼20 μm). The radius of the solid circle represents the interaction range of the selected TC, (∼50 μm). The colored lines represent the tracks of the immune cells that approach the tumor cell considered. Each color corresponds to a different immune cell.CC BY · PubMed Central
Figure 7 Analysis and comparison of the stable and metastable cases. In the left column (panels A1 and B1) bare Cartesian trajectories of PBMC are shown while moving towards the TC (black dot). The black continuous circle defines the interaction range while the inner dashed circle defines the linearization range. The interaction range has been defined considering the polar plots presented in the central column of the image (panels A2 and B2), which show, frame by frame, the distance r of the selected PBMC from the TC plotted as a function of the angle θ. Cell’s motion starts in the top right part of the plot (blue spots) and it is mainly noisy up to r ∼ 50 μm where a change in the direction is perceived, setting the boundaries of the interaction range. In this range (green spots) r decreases linearly with θ when the PBMC points towards the TC, and, only in the metastable case, it also increases linearly with θ when the PBMC leaves the TC after the interaction. In all the cases, the linear behaviors are fitted with the functions r=p1θ+p2 (red lines, panels A2 and B2), whose parameters are p1=62.67,p2=−144.61 for the stable convergence (panel A2), p1=−31.17,p2=146.55 in approaching the TC and p1=12.42,p2=−24.39 in leaving the TC for the metastable convergence (panel B2). The linearization range is set by taking the PBMC experimental data of (vr,vθ) versus (r,θ) close to (r 0 ,θ 0) that show a linear behavior (see Figs S3 and S4). The linear regime is within ∼20 μm in the stable case and ∼49 μm in the metastable case. The right column (panels A3a,b and B3a,b) describes the temporal evolution of the Lyapunov coefficients λ1 and λ2 computed within the linearization range. Standard errors extracted from the best fits of the phase space analysis are calculated for each point. Notice that: in the stable case (A3a,b) for t → 0 they both converge to negative values thus confirming a stable match between PBMC and TC; in the metastable case (B3a,b) their values have almost always opposite sign (that they switch when they leave the fixed point) framing such a behavior as a saddle with regard to the prescribed TC as a candidate target cell.
Figure 7 Analysis and comparison of the stable and metastable cases. In the left column (panels A1 and B1) bare Cartesian trajectories of PBMC are shown while moving towards the TC (black dot). The black continuous circle defines the interaction range while the inner dashed circle defines the linearization range. The interaction range has been defined considering the polar plots presented in the central column of the image (panels A2 and B2), which show, frame by frame, the distance r of the selected PBMC from the TC plotted as a function of the angle θ. Cell’s motion starts in the top right part of the plot (blue spots) and it is mainly noisy up to r ∼ 50 μm where a change in the direction is perceived, setting the boundaries of the interaction range. In this range (green spots) r decreases linearly with θ when the PBMC points towards the TC, and, only in the metastable case, it also increases linearly with θ when the PBMC leaves the TC after the interaction. In all the cases, the linear behaviors are fitted with the functions r=p1θ+p2 (red lines, panels A2 and B2), whose parameters are p1=62.67,p2=−144.61 for the stable convergence (panel A2), p1=−31.17,p2=146.55 in approaching the TC and p1=12.42,p2=−24.39 in leaving the TC for the metastable convergence (panel B2). The linearization range is set by taking the PBMC experimental data of (vr,vθ) versus (r,θ) close to (r 0 ,θ 0) that show a linear behavior (see Figs S3 and S4). The linear regime is within ∼20 μm in the stable case and ∼49 μm in the metastable case. The right column (panels A3a,b and B3a,b) describes the temporal evolution of the Lyapunov coefficients λ1 and λ2 computed within the linearization range. Standard errors extracted from the best fits of the phase space analysis are calculated for each point. Notice that: in the stable case (A3a,b) for t → 0 they both converge to negative values thus confirming a stable match between PBMC and TC; in the metastable case (B3a,b) their values have almost always opposite sign (that they switch when they leave the fixed point) framing such a behavior as a saddle with regard to the prescribed TC as a candidate target cell.CC BY · PubMed Central
Figure 8 Left panel: plot of the Lyapunov coefficients (related to the immune cells attracted by the selected tumor cell). Left panel: each point (λ1, λ2) represents the Lyapunov coefficients extracted from the phase space analysis of each immune cells attracted by the tumor cell analyzed in this section. Both negative λ1 and λ2 identify the space of stability convergence, while both positive λ1 and λ2 identify the space of unstability convergence. When λ1<0<λ2, the candidate stationary point is in the metastable space. Standard errors extracted from the best fits of the phase space analysis are calculated for each point. Right panel: effective chemoattractant potentials generated by the tumor cells. Representative landscape of the chemoattractant potentials generated by three tumor cells as revealed by our approach. The minimum of each potential well corresponds to the position of a tumor cell in the physical space of the microfluidic chip (i.e., the plane of the spatial coordinates (x,y)). Z(rel) accounts for the depth of the potential well (normalized to the largest value), determined by its local equilibrium under the (standard) parabolic approximation: the larger the local equilibrium, the deeper the well. The local equilibrium is isolated when the attractor does not interact with the others (see potential well A). Conversely, when two basins partly overlap (see potential wells B and C), interference among their signal may result in metastability.
Figure 8 Left panel: plot of the Lyapunov coefficients (related to the immune cells attracted by the selected tumor cell). Left panel: each point (λ1, λ2) represents the Lyapunov coefficients extracted from the phase space analysis of each immune cells attracted by the tumor cell analyzed in this section. Both negative λ1 and λ2 identify the space of stability convergence, while both positive λ1 and λ2 identify the space of unstability convergence. When λ1<0<λ2, the candidate stationary point is in the metastable space. Standard errors extracted from the best fits of the phase space analysis are calculated for each point. Right panel: effective chemoattractant potentials generated by the tumor cells. Representative landscape of the chemoattractant potentials generated by three tumor cells as revealed by our approach. The minimum of each potential well corresponds to the position of a tumor cell in the physical space of the microfluidic chip (i.e., the plane of the spatial coordinates (x,y)). Z(rel) accounts for the depth of the potential well (normalized to the largest value), determined by its local equilibrium under the (standard) parabolic approximation: the larger the local equilibrium, the deeper the well. The local equilibrium is isolated when the attractor does not interact with the others (see potential well A). Conversely, when two basins partly overlap (see potential wells B and C), interference among their signal may result in metastability.CC BY · PubMed Central
Figure 9 Schematic representation of the proposed tool. From left to right: background identification, cell localization, cell discrimination, individual cell tracking and track merging.
Figure 9 Schematic representation of the proposed tool. From left to right: background identification, cell localization, cell discrimination, individual cell tracking and track merging.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 37 references

Research works printed but not linked: 2

  1. Agliari, E., Annibale, A., Barra, A., Coolen, A. C. C. & Tantari, D. Immune networks: Multitasking capabilities at medium load. J. Phys. A 46, 335 (2013).
  2. Khalil, H. K. & Grizzle, J. W. Nonlinear systems. (Prentice hall, New Jersey, 1996).

Cited by

See all 95

Metadata

Authors
Biselli, E., Agliari, E., Barra, A., Bertani, F. R., Gerardino, A., De Ninno, A., Mencattini, A., Di Giuseppe, D., Mattei, F., Schiavoni, G., Lucarini, V., Vacchelli, E., Kroemer, G., Di Natale, C., Martinelli, E., & Businaro, L.
Affiliations
  • Technical University of Munich (DE)
  • Sapienza University of Rome (IT)
  • National Research Council (IT)
  • University of Rome Tor Vergata (IT)
  • Istituto Superiore di Sanità (IT)
  • Inserm (FR)
  • Karolinska University Hospital (SE)
Corresponding author
Martinelli, Eugenio
Venue
Scientific Reports
Published
2017-10-06
Volume
7
Issue
1
Pages
12737-12737
ISSN
2045-2322 (ISSN-L)
OpenAlex
W2762204590
Cited by
91
References
36
Citations by year
2026: 6 · 2025: 3 · 2024: 9 · 2023: 12 · 2022: 8 · 2021: 12 · 2020: 16 · 2019: 17 · 2018: 8
FWCI
8.45
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Cineca Institutional Research Information System (Tor Vergata University) (submittedVersion, other-oa)
  • IRIS Research product catalog (Sapienza University of Rome) (submittedVersion, other-oa)
  • HAL (submittedVersion, cc-by)
  • AIRUS (submittedVersion)
Funders
  • Association pour la Recherche sur le Cancer
  • Istituto Nazionale di Alta Matematica "Francesco Severi" (Agliari2016)
  • E-Rare
  • European Commission
  • Deutsche Forschungsgemeinschaft
  • Agence Nationale de la Recherche (E-Rare-2)
  • Fondation Leducq
  • Fondation pour la Recherche Médicale
  • Ministero della Salute (RF-2011-02347120)
  • Ministero dell’Istruzione, dell’Università e della Ricerca (MIUR-PRIN2012, PRIN2012, 20125NMMLA)
  • Ligue Contre le Cancer
  • Institut Universitaire de France
  • Institut National Du Cancer
  • Gruppo Nazionale per la Fisica Matematica
  • Labex Immuno-Oncology
Topics
  • Physical Sciences › Mathematics › Modeling and Simulation › Mathematical Biology Tumor Growth
  • Physical Sciences › Engineering › Biomedical Engineering › 3D Printing in Biomedical Research
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cell Biology › Cellular Mechanics and Interactions
Keywords
organ-on-a-chip, tumor-immune interactions, leukocyte migration, FPR1, peripheral blood mononuclear cells, biased random walk, Lyapunov stability, time-lapse imaging
MeSH terms
Cell Communication, Cell Movement, Humans, Leukocytes, Motion, Neoplasms, Cell Line, Tumor, Lab-On-A-Chip Devices

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Organs on chip approach: a tool to evaluate cancer -immune cells interactions. https://app.cochat.ai/papers/10.1038/s41598-017-13070-3
The paper
Biselli, E., Agliari, E., Barra, A., Bertani, F. R., Gerardino, A., De Ninno, A., Mencattini, A., Di Giuseppe, D., Mattei, F., Schiavoni, G., Lucarini, V., Vacchelli, E., Kroemer, G., Di Natale, C., Martinelli, E., & Businaro, L. (2017). Organs on chip approach: a tool to evaluate cancer -immune cells interactions. Scientific Reports. https://doi.org/10.1038/s41598-017-13070-3

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.