Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Oct;10(28):e2304020.
doi: 10.1002/advs.202304020. Epub 2023 Aug 6.

Intra-Operative Definition of Glioma Infiltrative Margins by Visualizing Immunosuppressive Tumor-Associated Macrophages

Chong Cao  1 , Hang Yin  1 , Biao Yang  1 , Qi Yue  1 , Guoqing Wu  2 , Meng Gu  1 , Yuwen Zhang  3 , Yang Fan  1 , Xiaoyan Dong  1 , Ting Wang  1 , Cong Wang  1 , Xiao Zhu  1 , Ying Mao  1 , Xiao-Yong Zhang  3 , Zuhai Lei  1 , Cong Li  1   4
Affiliations

Intra-Operative Definition of Glioma Infiltrative Margins by Visualizing Immunosuppressive Tumor-Associated Macrophages

Chong Cao et al. Adv Sci (Weinh). 2023 Oct.

Abstract

Accurate delineation of glioma infiltrative margins remains a challenge due to the low density of cancer cells in these regions. Here, a hierarchical imaging strategy to define glioma margins by locating the immunosuppressive tumor-associated macrophages (TAMs) is proposed. A pH ratiometric fluorescent probe CP2-M that targets immunosuppressive TAMs by binding to mannose receptor (CD206) is developed, and it subsequently senses the acidic phagosomal lumen, resulting in a remarkable fluorescence enhancement. With assistance of CP2-M, glioma xenografts in mouse models with a tumor-to-background ratio exceeding 3.0 for up to 6 h are successfully visualized. Furthermore, by intra-operatively mapping the pH distribution of exposed tissue after craniotomy, the glioma allograft in rat models is precisely excised. The overall survival of rat models significantly surpasses that achieved using clinically employed fluorescent probes. This work presents a novel strategy for locating glioma margins, thereby improving surgical outcomes for tumors with infiltrative characteristics.

Keywords: gliomas; hierarchical strategy; pH responsive probes; surgical navigations; tumor-associated macrophages.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
M2‐TAMs accumulate at the invasive margins of GBM. A) The heatmap shows the immune profiles in the TCGA‐glioma cohort. The upper panel demonstrates the expression of biomarkers involved in immunosuppression. The below panel demonstrates the densities of immune cells in glioma. The histology and WHO grade of glioma are annotated at the top of the heatmap. B) The relative percentage of immune cells in LGG (blue plot) versus GBM (red plot) was analyzed in the TCGA database. The scattered dot represents the relative percentage of immune cells in the tumor sample from each glioma patient. C) The percentages of M2‐TAMs in normal brain and glioma tissues from patients. D) The percentages of M2‐TAMs as a function of glioma grades. E) Kaplan‐Meier survival curves comparing overall survival between glioma patients with low (blue) and high (red) percentages of M2‐TAMs in the Rembrandt database. Tick marks indicate censoring. F) Receiver‐operating characteristic (ROC) curve for prediction of diagnosis in patients with glioma. When AUC > 0.7, gliomas were accurately diagnosed by the percentages of M2‐TAMs. G) Representative H&E, CD68, and CD206 immunohistochemistry images of marginal tumor region from the adjacent slices of patient glioma sample. Scale bar: 20 µm. T: tumor; N: normal tissue; LGG: low‐grade glioma; GBM: glioblastoma; ROC: Receiver operating characteristic; AUC: The area under curve; OS: overall survival. NS  =  non‐significant, *p < 0.05, ***p < 0.001, ****p < 0.0001. Data are given as the mean ± SD., the statistical significance of the survival curve was calculated using the log‐rank test in E, and a two‐tailed Student's t‐test was performed in C, D, and F. Statistical significance was considered for a p‐value of <0.05.
Figure 2
Figure 2
Synthesis and characterization of CP2‐M. A) Chemical structures of CP2‐M in its protonated and deprotonated forms. B) Chemical structures of control probes including CP2, CP2‐P modified with PEG2K, and CB‐M modified with PEG2K and mannose but without pH responsiveness. C) The HOMO and HOMO‐1 energy levels for CP2, based on density functional theory (DFT) calculations at the B3LYP/6–311 G (d,p) level. Absorption D) and emission spectra E,F) of CP2‐M as a function of pH. CP2‐M was excited at 740 and 670 nm, respectively. G) Plotting fluorescence intensity ratio (I740/I670 nm) as a function of pH. Inset: the linear relationship between I740/I670 ratio and pH value. H−K) Surface plasmon resonance (SPR) H,I) binding analysis and concentration‐response plots of PEG2K‐mannose and J,K) CP2‐M binding CD206 performed at concentrations between 0.78 µM and 12.5 µM with fitted affinity (1:1 steady‐state affinity model). Steady‐state binding curves from SPR indicate that the binding affinity depends on the concentration of I) PEG2K‐mannose and K) CP2‐M.
Figure 3
Figure 3
CP2‐M identifies M2‐Mφ with high specificity. A) Isolation, differentiation, and polarization of primary macrophages from mouse bone marrow. i) extraction of bone marrow, ii) differentiation into M0‐Mφ by M‐CSF, iii) polarization into M1‐Mφ via LPS/IFN‐γ, and iv) into M2‐Mφ via IL‐4. B) Confocal fluorescence microscope images of live M2‐MΦ after treatment with CP2, CP2‐P, or CP2‐M in the presence and absence of mannose (blockade control); C) Confocal fluorescence images of live C6 GBM cancer cells, M0‐Mφ, M1‐Mφ, and M2‐Mφ treated with CP2‐M (25.0 µM) for 1 h, scale bar: 30 µm; D,E) Quantified intracellular mean fluorescence intensities (MFI) of CP2‐M by flow cytometry statistical analysis corresponding panel B and panel C; F) Fluorescence images of macrophages treated with CP2‐M and Lysotracker, respectively. Yellow regions indicated the CP2‐M uptake in the lysosomes, scale bar: 10 µm; G) Illustration of CP2‐M specifically visualizing M2‐MΦ instead of M1‐MΦ or cancer cells via hierarchical strategy; H) pH map of live macrophages after treatment of CP2‐M (25.0 µM). pH maps were generated from the ratiometric fluorescence signal excited at 670 and 740 nm, λ em  =  780−830 nm. Scale bar: 10 µm. I) Average intracellular pH values measured in the phagosomes of M1‐Mφ and M2‐Mφ. Mφ: macrophages. Data with error bars are expressed as mean ± S.D. (n  =  3), and a two‐tailed unpaired t‐test is used for comparison between two sets of data in D, E, and I. Statistical significance was considered for a p‐value of < 0.05.
Figure 4
Figure 4
CP2‐M locates GBM xenograft with high specificity. A) Schematic of the PK/PD model investigating fluorescent probes, which includes a three‐compartment PK model along with an effect compartment for the target site of the probes. B) A visual predictive check from the PK model shows that the simulated probe concentrations have the same trend and variation as the observed data. The gray dots are the experimental data, the solid red line indicates the predicted 50 percentile, and the blue area represents the 90% prediction interval of the final model simulation. C) In vivo fluorescence images of mouse models at selected time points after intravenous injection of CP2, CP2‐P, CB‐M, or CP2‐M (with CP2 dose of 5.0 µmol kg−1). The probes were excited at 740 nm and emission between 780‒820 nm was collected. Tumors are indicated by D) white arrows. T/B ratio and E) mean fluorescence intensity at tumor site as a function of time post probe administration. F) Fluorescence images of heart (He), liver (Li), spleen (Sp), lung (Lu), kidney (Ki), and brain (Br) from mouse models sacrificed at 8 h post‐injection of CP2, CP2‐P, CB‐M or CP2‐M (5.0 µmol kg−1). G) Quantitative analysis of mean fluorescence intensity in the excised tumors and major organs. Data are given as the mean ± SD (n = 3 mice per group), and the two‐way analysis of variance (ANOVA) was performed in panel G. Statistical significance was considered for a p‐value of < 0.05.
Figure 5
Figure 5
CP2‐M defines glioma allograft in rat models after craniotomy. A) Schematic diagram of in vivo imaging intracranial glioblastoma xenograft via 740 nm excitation wavelength. B) Intra‐operative fluorescence images of intracranial tumor region at selected time points after intravenous injection of CP2, CP2‐P, CB‐M, or CP2‐M (5.0 µmol kg−1). Tumors are delineated with white dashed lines, scale bar: 2.0 mm. C) A visual predictive check from the final PD model shows that the simulated T/B ratios have the same trend and variation as the experimental data. D) Time‐dependent T/B ratios after a single administration of CP2‐M (5.0 µmol kg−1). The black dashed line indicates the ideal T/B value and the time slot between the two red lines is the recommended time window for operation.
Figure 6
Figure 6
CP2‐M locates GBM margins by targeting M2‐TAMs. A) Flow chart presents in vivo fluorescence imaging of GBM allograft in the same rat model by using three fluorescent probes subsequently. B) C) Representative white light, fluorescence images of tumor region post oral administration of 5‐ALA, D) intravenous injection of ICG, E) pH map delineated by intravenous administrated CP2‐M. The pH map of the tumor region was generated by determining the ratio of CP2‐M fluorescence intensities after excitation at 740 and 670 nm, respectively. The dotted lines indicate the tumor body. The color bar represents the pH magnitude. Scale bars, 2.0 mm. F) Representative H&E staining image of glioma xenograft (B) and the enlarged images of the region G) I, H) II, and I) III, scale bars: 2.0 mm and 50 µm (enlarged). J) Immunohistological staining of CD206+ TAMs in tumor and the surrounding region. Spatial distribution of CD206+TAMs in regions K) IV, L) V, and M) VI. N) Schematic illustration presenting spatial distribution patterns of M2‐TAMs and cancer cells in glioma. O) M2‐TAMs densities in different regions of the tumor. P) The maximal T/B ratio of the probes measured to visualize GBM allograft. Q) Flow cytometric analysis of intracellular fluorescence intensity of cancer cells and M2‐TAMs isolated from tumor tissue at 4 h post‐CP2‐M administration. scale bars: 2.0 mm and 50 µm (enlarged). Data with error bars are expressed as mean ± S.D. (n  =  3), and an unpaired t‐test was used for comparison between two sets of data in O, and P. Statistical significance was considered for a p value of <0.05.
Figure 7
Figure 7
CP2‐M guided surgery by delineating pH map of tumor bed. A) The procedure of CP2‐M guided surgery in rat models bearing C6 glioma allografts. B) Intracranial surgery with the guidance of a pH map delineated by CP2‐M. The surgery did not terminate until all the regions with pH values below 7.1 were excised, scale bars: 2.0 mm. Representative C) H&E staining and D) cell density of the excised tissues with different measured acidities. Scale bar: 20 µm. Flow cytometric analysis of immunostimulatory E,F) CD206+ CD86‐ macrophages (M2‐TAMs) among CD11b+ F4/80+ cells and G,H) Foxp3+ CD4+ Tregs among CD45+ CD3+ cells in the excised tumor tissues with different acidities. Data are shown as the mean ± s.d. (n = 3), and an unpaired t‐test is used for comparison between two sets of data in D, F, and H. Statistical significance was considered for a p‐value of < 0.05.
Figure 8
Figure 8
CP2‐M guided surgery suppresses tumor recurrence. A) Experimental timeline for monitoring tumor relapse after image‐guided surgery in rat models bearing intracranial GBM allograft. B) Representative T2W and contrast‐enhanced T1W MR images of rat models at selected days post‐ICG, 5‐ALA, or CP2‐M guided surgery. Tumor margins are indicated by dashed line. The mass at the tumor bed after CP2‐M guided surgery was verified as hydrocephalus instead of relapsed tissue. C) Representative H&E staining images of rat brain sections at the end of the experiments. The red dotted line indicates tumor recurrence after surgery, while the green dotted line represents the tumor excision bed that did not recur. The scale bars are 2.0 mm and 50 µm (enlarged). Relapsed tumor volume D) and Kaplan–Meier survival curves E) of rat models after surgical intervention (n = 5 rat models per group). F) Time spent on the rotarod by mouse models after surgery guided by ICG, 5‐ALA, or CP2‐M. G) Diagrams for the spontaneous (left) and Incorrect (right) alternation in the Y‐maze test. H) The spontaneous alternation performance of mouse models in the Y‐maze test. The percent of alternation in the Y‐maze was significantly above chance level (50%) in CP2‐M group but not in ICG and 5‐ALA group. I) Beam walking scores in the beam‐walking test after surgery (n = 4 mice per group). CE‐T1W: contrast‐enhanced T1‐weighted; MRI: magnetic resonance imaging. Data are given as the mean ± SD (n = 4 or 5 mice per group), the statistical significance of the survival curve was calculated using the log‐rank test, and two‐tailed Student's t‐test was performed in D, F, H, and I. Statistical significance was considered for a p‐value of <0.05.
See this image and copyright information in PMC

References

    1. Gao M., Fu Y., Zhou W., Gui G., Lal B., Li Y., Xia S., Ji H., Eberhart C. G., Laterra J., Ying M., Cancer Res. 2021, 81, 3580. - PMC - PubMed
    1. Luo J., Liu P., Lu C., Bian W., Su D., Zhu C., Xie S., Pan Y., Li N., Cui W., Pei D.‐S., Yang X., Brain 2021, 144, 615. - PMC - PubMed
    1. Agliardi G., Liuzzi A. R., Hotblack A., de Feo D., Núñez N., Stowe C. L., Friebel E., Nannini F., Rindlisbacher L., Roberts T. A., Ramasawmy R., Williams I. P., Siow B. M., Lythgoe M. F., Kalber T. L., Quezada S. A., Pule M. A., Tugues S., Straathof K., Becher B., Nat. Commun. 2021, 12, 444. - PMC - PubMed
    1. Jung E., Osswald M., Ratliff M., Dogan H., Xie R., Weil S., Hoffmann D. C., Kurz F. T., Kessler T., Heiland S., von Deimling A., Sahm F., Wick W., Winkler F., Nat. Commun. 2021, 12, 1014. - PMC - PubMed
    1. Kubben P. L., ter Meulen K. J., Schijns O. E., ter Laak‐Poort M. P., van Overbeeke J. J., van Santbrink H., Lancet Oncol. 2011, 12, 1062. - PubMed

Publication types

LinkOut - more resources