 Home > Biochemistry News > Biotechnology News > Diabetic hyperglycemia promotes primary tumor progression through glycation-induced extracellular stromal sclerosis

Diabetic hyperglycemia promotes primary tumor progression through glycation-induced extracellular stromal sclerosis

 Last Update: 2023-01-05
 Source: Internet
 Author: User

Tags

body life diet

collagen glucosamine chondroitin

Search more information of high quality chemicals, good prices and reliable suppliers, visit www.echemi.com

Summary

Diabetes is a complex metabolic disorder that has been linked to
an increased risk of breast cancer.
Despite this correlation, the interaction between tumor progression and diabetes, particularly with regard to extracellular stromal sclerosis, remains mechanistically unclear
.
Here, we built a mouse model that induced hyperglycemia
before breast tumorigenesis.
Using mouse models, in vitro systems, and patient samples, we show that hyperglycemia increases tumor growth, extracellular matrix hardness, glycosylation, and epithelial-mesenchymal transformation
of tumor cells.
When inhibiting glycosylation or mechanical transformation in diabetic mice, these indicators are reduced to levels
comparable to those of non-diabetic tumors.
In conclusion, our study describes a novel biomechanical mechanism that promotes breast tumor progression
by glycifying the extracellular matrix, diabetic hyperglycemia.
In addition, our study also demonstrates that glycosylation inhibition is a potential adjuvant treatment for diabetic cancer patients, as stromal sclerosis plays a key role
in both diseases.

Brief introduction

Diabetes mellitus is a common endocrine disorder characterized by deficient insulin secretion and/or hyperglycemia due to insulin resistance (1).
Clinical studies have shown that diabetes is associated with an increased risk of several cancers, including liver, colon, pancreatic, bladder, breast, and rectal cancers (2).
More specifically, there is evidence that breast cancer patients with a history of diabetes for more than 5 years have a higher mortality rate than non-diabetic patients (3–6).
Previous studies have shown that diabetic status promotes tumor progression through a variety of mechanisms, including hyperinsulinemia, hyperglycemia, and chronic inflammation (iii).
However, the role of the extracellular matrix (ECM) in the progression of diabetes has been overlooked
.

Reducing sugars, such as glucose, can cause increased cross-linking and ECM modification in collagen tissue through glycosylation (7–9).
Glycosylation is a non-enzymatic reaction of sugars with amino groups of proteins, lipids, and nucleic acids, resulting in crosslinking and the formation of advanced glycosylation end products (AGEs) (9).
Glycosylation can interfere with the normal function of proteins by disrupting conformation, altering enzyme activity, and interfering with ligand-receptor interactions (10,11).
Disruption of these protein levels can alter cell signaling and may affect tumor progression (5).
In addition to altering cell signaling, hyperglycemia-induced glycosylation has also shown crosslinking and hardening of the collagen matrix in vitro (7, 12, 13).
Since tumor extracellular stromal sclerosis promotes malignancy (14,15) that sugar can promote stromal sclerosis through glycosylation, we hypothesize that diabetic states promote tumor progression
through non-enzymatic glycosylation-induced extracellular stromal sclerosis.

Here, we find that hyperglycemia hardens the tumor extracellular matrix, promotes the progression of breast tumors, increases cell proliferation, and further moves tumor cells along the epithelial cells towards mesenchymal transformation (EMT
).
Of note, drug inhibition of glycosylation or cellular mechanical induction reduced tumor progression without affecting blood glucose levels compared to non-diabetic mice
.
Our findings reveal a new mechanism by which diabetes promotes breast tumor progression through glycosylation, and point to therapies targeting glycosylation as a mechanism
to delay tumor growth in diabetic patients.

Results A model of diabetic hyperglycemia in FVB/N-TG (MMTV-PYMT) 634Mul/J mice was established

In this study, FVB/N-Tg (MMTV-PyMT) 634Mul/J (PyMT) mice were used to develop spontaneous breast tumors
from 6 weeks after birth.
To induce hyperglycemia in PyMT mice, we feed mice a high-fat feed starting at 4 weeks of age and then continuously inject streptozotocin (STZ) at 70 mg/kg doses five times daily at 5 weeks of age (Figure 1A).
STZ induces pancreatic and renal tumors in rats, while STZ induces tumors at high doses (16–18).
Therefore, we next tested whether STZ treatment affected breast tumor progression independently of hyperglycemic induction
.
MET-1 cells extracted from PyMT mice were treated with STZ for 5 days to mimic the effect of
STZ on mice.
DNA damage was quantified with comet assays (19, 20).
There was no significant difference in DNA damage in STZ-treated or non-STZ-treated cells (Figure S2A).

Since necrotic and apoptotic cells contain a large amount of damaged DNA and strand breaks, to ensure the accuracy of comet assays, cell death
of control and STZ-treated cells was compared using terminal deoxynucleotidyl transferase-mediated deoxyuridinetriphosphate notched end labeling (TUNEL).
STZ treatment did not cause significant changes in cell death (Figure S2B).

In addition, as a control, a group of STZ-treated mice were given insulin treatment, using osmotic pumps to lower blood glucose to the normal range
.
To verify the induction of hyperglycemia, blood glucose levels
before and after fasting were measured weekly.
At 12 weeks of age, mice are euthanized and tissue specimens
are collected and analyzed.
Compared to non-diabetic PyMT mice, STZ-injected PyMT mice had significantly elevated blood glucose levels, while insulin treatment reduced blood glucose in STZ-treated mice to levels comparable to those in the non-diabetic group (Figures 1, B, and C).
The blood glucose of the STZ-injected mice rose above 400 mg/dl and remained steadily elevated
throughout the study.
The control group, mice injected with STZ and STZ+ insulin steadily gained weight (Figure 1D).
The glucose tolerance and insulin tolerance tests performed at week 6 showed elevated blood glucose levels in mice in the STZ injection group compared to the control group (Figures 1, E and F).
Staining the pancreas of diabetic mice with anti-insulin antibodies revealed a significant reduction in islet cells (the main cells targeted by the STZ) (Figures 1, G and H).
These results show that MMTV-PyMT mice successfully induce diabetic hyperglycemia at week 6 and can be reversed
by an insulin pump.

Figure 1Induction
of diabetes mellitus in MMTV-PyMT mice.

(A) Schematic diagram
of the MMTV-PyMT mouse model of diabetes.
(B) Rehearsal and (C) post-fasting blood glucose levels of MMTV-PyMT mice injected with STZ or sodium citrate buffer reference (Ctrl) 1 time daily for 5 consecutive days, or simultaneous injection of STZ and insulin (STZ + insulin).

(D) MMTV-PyMT mouse weight tracking, injection of 70 mg/kg sodium citrate (Ctrl) or STZ for 5 consecutive days, or simultaneous administration of STZ and insulin (STZ + insulin) treatment
.
(E) Changes in blood glucose levels over time after insulin injection in the control group and STZ group
.
(F) Changes in blood glucose levels over time in the Ctrl and STZ groups after glucose injection in the glucose tolerance test
.
(G) Typical image
of insulin immunohistochemical staining in the pancreas of MMTV-PyMT mice injected with sodium citrate (Ctrl) or STZ.
(H) Quantification of percentage of islet cells in control groups or STZ-injected mice (Ctrl, N=7; STZ firm, N=5; STZ+ Insulin, N=3).

The data were shown by mean ± scanning electron microscopy ***P<0.
001 and ****P<0.
0001.
NS, which were not significant
.

Open diabetic miliary in the viewer leads to tumor extracellular stromal sclerosis by glycosylation

Because hyperglycemia leads to increased glycosylation of the extracellular matrix (12,13), glycosylated tumor extracellular matrix (ECM) levels
were detected by immunohistochemistry (IHC) and western blotting.
Compared with non-diabetic mice, diabetic mice had an increased tumor age, while diabetic mice with insulin treatment significantly reduced tumor age, making it comparable to non-diabetic mice (Figure 2, A to D).
Because sugars crosslink collagen through a non-enzymatic reaction, resulting in a hardening of the matrix (7), we expanded our research to assess collagen concentration, a major ECM component
in breast tumors between diabetic and non-diabetic PyMT tumors.
Tumor sections are stained with picric red and imaged by quantitative polarization microscopy
.
There was no significant difference in the optical delay of collagen signaling observed between diabetic and non-diabetic tumors, indicating similar collagen concentrations between experimental groups (Figure 2E).
To assess whether hyperglycemia and the formation of AGEs associated with it hardens the tumor extracellular matrix, we performed an unconfined compression assay to detect the equilibrium modulus
of diabetic and non-diabetic tumors.
Tumors from diabetic mice have a higher equilibrium modulus compared to non-diabetic mice (Figure 2F).
Consistent with this result, atomic force microscopy (AFM) measurements showed that hyperglycemia increased tumor elastic modulus compared to control mice, while diabetic mice with insulin treatment significantly reduced tumor elastic modulus compared to control mice (Figure 2, G to I).
These data suggest that hyperglycemia-induced glycosylation stiffens breast tumors (Figure 2G) and increases tumor stiffness heterogeneity (Figures 2, H and I).

Figure 2 Hyperglycemia promotes glycosylation and increases ECM hardness
.

(A) IHC-stained images
of representative tumor slices collected from non-diabetic (Ctrl), diabetic (STZ), and insulin-treated diabetic mice (STZ+ insulin).
(B) Corresponding quantification of standardized age-positive area (Ctrl, N=5 and n=14; STZ companies, N=4 and n=17; insulin + insulin, N=3 and n=12).

(C) Representative Western blot protein bands of AGEs and GAPDH in diabetic (STZ) and non-diabetic (Ctrl) mouse tumors
.
(D) Corresponding quantification of GAPDH standardized age (N=4 and n=6).

(E) Quantitative study of collagen deposition in diabetic and non-diabetic tumors (N=3 and n=3; 30 measurements per case).

(F) Unconfined compression assay showing equilibrium modulus of diabetic or non-diabetic mouse tumors (Ctrl, N=4 and n=6; STZ Inc.
).
N=4 and n=5; 20 measurements in each case).

(G) Measurement of elastic modulus (N=3 and n=3; of non-diabetic (Ctrl), diabetes mellitus (STZ) and insulin-treated diabetic tumors (STZ + insulin) by atomic force microscopy 250 to 300 measurements).

(H) AFM measurements for non-diabetic (Ctrl), diabetes mellitus (STZ), and insulin-treated diabetic tumors (STZ + insulin) correspond to histograms
.
(i) Representative effort to extract tumors
from non-diabetic (Ctrl), diabetes mellitus (STZ), and insulin-treated diabetic mice (STZ + insulin).
Data are represented by mean ± SEM*P<0.
05***P<0.
001, and ****P<0.
0001.
a.
u.
, arbitrarily
.

Opening hyperglycemia in the viewer promotes tumor progression in PyMT mice

Since hyperglycemia-induced glycosylation increases the stiffness of PyMT tumors, and stiffness of ECM is thought to promote tumor malignancy (21, 22), we next investigated the relationship between progression of breast tumors and diabetic status in mice
.
Tumor volume
of non-diabetic, diabetic, or diabetic mice treated with insulin was measured weekly until 12 weeks of age.
Accelerated tumor growth in diabetic mice (Figure 3A) ultimately resulted in larger tumors (Figures 3, B, and C) compared
to the non-diabetic group.
At the same time, insulin therapy can significantly slow the tumor growth
of diabetic mice.
Histological analysis showed that diabetic tumors were less differentiated and more malignant than non-diabetic tumors (Figure 3D).
To determine whether the increase in tumor cell proliferation is the cause of increased tumor growth in diabetic mice, we quantitatively analyzed the percentage of tumor cells ⁺ nuclear immunofluorescence staining (Figure 3E) with Ki67 The results showed that hyperglycemic states significantly promoted cell proliferation, and insulin therapy inhibited cell proliferation (Figure 3F).

Figure 3: Hyperglycemia promotes tumor growth
by promoting cell proliferation.

(A) Weekly average tumor volume measurements of non-diabetic (Ctrl), diabetic (STZ) and diabetic mice treated with insulin (STZ + insulin), from tumors becoming palpable and large enough for caliper measurements (week 9) to study endpoints (Ctrl, N = 14; STZ companies, N =12; STZ + insulin, N=3).

(B) Mean tumor volume (Ctrl, N=7 and n=22) of mice with diabetes treated with control (Ctrl), diabetes mellitus (STZ), and diabetic mice treated with insulin (STZ + insulin) at the study endpoint; STZ companies, N=13 and n=75; STZ + insulin, N=3 and n=12)
。 (C) Average tumor burden of control (Ctrl), diabetes mellitus (STZ), and diabetic mice treated with insulin (STZ + insulin) at the study endpoint (Ctrl, N = 7; STZ company, N =22; STZ + insulin, N=3).

(D) Tumor differentiation fraction (N=3 and n=3) of tumors in control group and STZ-treated mice.

(E) Representative images showing the colocalization
of Ki67 and nuclei within non-diabetic, diabetic (STZ) and insulin-treated diabetic tumors.
(F) Corresponding quantification of the percentage of Ki67 cells ⁺ nuclei (Ctrl, N=7=3 and n= 3,21 imaging fields; STZ companies, N=4 and n=6,67 imaging fields; STZ + insulin, N=3 and n=880 imaging fields).

Data were shown by mean ± SEM*P<0.
05***P<0.
001, and ****P o-phenylenediamine 0.
0001-0
.

Opening hyperglycemia in the viewer enables PyMT tumors to transition to the EMT phenotype

Stromal sclerosis-promoting tumor cell EMT (23) is associated with
increased tumor aggressiveness and worsening prognosis in patients.
Therefore, we next investigate whether hyperglycemia-induced glycosylation affects tumor progression
by upregulating EMT.
Analysis of epithelial (E-cadherin) and mesenchymal (vimentin) markers (Figure 4A) showed a higher ratio of wave protein expression to E-cadherin expression in diabetic tumors compared to the control group (Figure 4B).
In addition, in insulin-treated diabetic tumors, this rate was reduced to a level
comparable to that of control tumors.
Together, this suggests that diabetic hyperglycemia is leading to a shift
to EMT.
Since increased expression of fibronectin is associated with early phenotype of breast cancer and is known to cause alterations in cell morphology and expression of EMT markers (24) immunohistochemical staining analyzes fibronectin expression
in tumors in non-diabetic, diabetic, and diabetic groups.
The content of fibronectin in diabetic mouse tumors was higher than in non-diabetic mice (Figures 4, C and D).
Insulin treatment reduced blood glucose levels in diabetic mice and significantly reduced the content
of fibronectin in diabetic tumors.
In addition to fibronectin, diabetic mouse tumors contain more tumor growth factor β (TGF-β) than non-diabetic mice (Figure 4 to Figure E).
Since TGF-β is a key factor in modulating EMT, our results suggest that diabetic tumors metastasize further to an aggressive mesenchymal phenotype on the EMT spectrum compared to non-diabetic tumors, and that insulin therapy can prevent this transition
.

Figure 4 Diabetes promotes EMT
.

(A) Representative images of immunofluorescence staining showing E-cadherin and wave protein
in non-diabetic (STZ), diabetic (STZ), and insulin-treated diabetic tumors (STZ + insulin).
(B) Corresponding expression rates of E-cadherin and waveform protein (Ctrl and STZ, N=3 and n=5, 37 fields per tumor segment; STZ + insulin, N=3 and n=8; 80 visual fields per tumor section).

(C) Representative images showing the expression of fibrobinding protein (FN) in non-diabetic (Ctrl)-extracted tumors, and diabetic tumors
treated with (STZ+STZ) or without (STZ) insulin.
(D) Corresponding quantification of fibronectin-positive areas within tumors (Ctrl, N=4 and n=13; STZ Company, N = 5 and n=17; STZ + insulin, N=3 and n=12).

(E) Representative images
of TGF-β stained tumor sections taken from control groups (Ctrl) or diabetic mice (STZ + insulin) or mice treated with no (STZ) insulin.
(F) Corresponding quantification of TGF-β-positive regions within tumors (Ctrl, N=3 and n=13; STZ Company, N =5 and n=15; STZ + insulin, N=3 and n=12).

(G) protein bands show expression of fibrobinding protein and TGF-β in diabetic (STZ) and non-diabetic (Ctrl) tumors
.
GAPDH as a loading control
.
(H and I) corresponding quantification (N=3 and n=6) of fibrobinding protein (H) and transforming growth factor-β(I) expression
。 Data are represented by mean ± SEM*P<0.
05**P<0.
01, and ****P<0.
0001
.

Turn on glycosylation inhibition in the viewer to reduce ECM hardness

Noting that glycosylation is a multi-step chemical reaction that can increase the stiffness of the extracellular matrix, we tested whether inhibition of glycosylation by aminoguanidine (AG) or the age crosslinking disruptor alagebrium (ALT711) improves the observed phenotype of diabetic mice (Figure 5A) (8).
Measure mouse blood glucose weekly to verify that glycosylation inhibitors do not affect blood glucose throughout the experiment (Figure 5B).
To determine whether glycosylation inhibition affects AGE concentration in STZ-injected mice, IHC staining
is performed.
Age concentrations were reduced in STZ-injected mice treated with either inhibitor compared to STZ-injected diabetic control mice (Figures 5, C and D).
When assessing collagen concentration, there was no significant difference in mean light delay between STZ-injected mice and uninhibitor-treated mice (Figure 5E).
The overall mechanical stiffness
is evaluated using an unconfined compression test.
Both AG and ALT711 treatment reduced the hardness of tumors in STZ-injected mice compared to those of non-diabetic mice in the control group (Figure 5F).
To compare microscale hardness, AFM measured tumor samples
with and without glycosylation inhibition.
The average modulus of elasticity of diabetic mice treated with both inhibitors was reduced compared to diabetic control mice (STZ only) (Figure 5G).
Compared to the untreated diabetes control, the elastic modulus values of diabetic tumors treated with AG- or ALT711 shifted overall to lower values, but without an upper value (Figure 5H).
These experiments showed that the increase in tumor stiffness in diabetic mice was due to increased glycosylation, which was successfully eliminated to the non-diabetic control tumor level by AG or ALT711 age-suppressing treatment, thereby reducing the stiffness of the tumor extracellular matrix, although the mice still exhibited hyperglycemia
.

Figure 5: Glycosylation inhibition reduces the hardness
of the ECM.

(A) Schematic diagram of the glycosylation reaction showing how glycosylation inhibitors block this reaction
.
(B) Blood glucose of diabetic mice treated with or without glycosylation inhibitors (STZ), N=12; STZ+AG, N=10; STZ+ALT711, N=12）
。 (C) Standardized positive area, N=4 and n=8; STZ+AG company, N=3 and n=7;STZ+ALT711,N=4 and n=7).

(D) IHC-stained images of representative tumor slices collected from diabetic mice treated with AG or ALT711 (E) Quantification of collagen deposition for diabetic and non-diabetic tumors (STZ), N=5 and n=7; 349 measurements were taken in each case; AG+STZ companies, N=2 and n=5; 126 measurements per case; STZ+ALT711, N=4 and n=8; 284 measurements).

(F) Unconfined compression assays showing the use of glycosylation inhibitors AG or ALT711 (STZ, N=6 and n=6, including 40 measurements; STZ+AG, N=5 and n=7, including 20 measurements; STZ+ALT711, N=3 and n=6, including 20 measurements).

(G) Measurement of the elastic modulus of diabetic tumors with or without glycation inhibitors using atomic force microscopy (N=3 and n=3430 to 589 measurements).

(H) Corresponding histogram
of AFM measurement of diabetic tumors treated with or without glycosylation inhibitors.
Data are shown as mean ± SEM with *P<0.
05 and ****P<0.
0001
.

Opening glycosylation in the viewer promotes the progression of primary tumors by increasing ECM hardness

Since ECM stiffness is a well-known malignancy (21, 22) inhibition of glycosylation by AG or ALT711 treatment reduces ECM hardness in STZ-injected mice, we assess whether glycosylation inhibition also alters tumor progression
by increasing ECM hardness.
Over time, the tumor volume of diabetic mice treated with AG or ALT711 was significantly lower than that of control tumors treated with diabetic STZ alone (Figure 6A).
At the study endpoint, the mean tumor volume of STZ-injected mice was significantly reduced compared to control mice injected with STZ alone (Figure 6B).
In addition, the average total tumor burden of diabetic mice treated with glycosylation inhibitors was significantly reduced compared to the control group alone (Figure 6C).
Glycosylation-inhibited diabetic tumors had significantly fewer Ki67-positive cells than diabetic control tumors (Figures 6, D and E).
These data suggest that the inhibition of glycosylation reduces the development of primary tumors and inhibits cell proliferation
.

Figure 6: Glycosylation inhibition reduces primary tumor progression
.

(A) Weekly tumor volume measurements (N=6 to 12) of diabetic mice treated with AG or ALT711 from week 9 to study endpoint.

(B) Average tumor volume at study endpoints (N = 6 to 12).

(C) Average tumor burden per mouse at study endpoint (N = 9 to 18).

(D) Quantitative analysis of percentage of Ki67 cells ⁺ nuclei (N=3 to 5 n=3 to 8, 32 to 64 imaging fields).

(E) Representative images showing cells with Ki67⁺ nuclei
.
(F) Comparison of MET-1 cells with normal or reduced RAGE expression on a matrix with 200~1000 pa hardness (N=3 and n=45).

Well, okay
.
1 vector
.
(G) Quantitative study of MET-1 cell proliferation in collagen matrix (N=3 and n=45).

(H) Typical images
of E-cadherin and vimentin in tumors.
(i) Corresponding quantification of E-cadherin and waveform protein expression rates (N=3 to 5n=4 to 7, 25 to 64 fields).

(J) Representative image
of TGF-β stained tumor sections.
(K) Corresponding quantification of TGF-β-positive area within the tumor (N=3 to 4n=8 to 7).

(i) corresponding quantification of fibrillonectin positive area (N=3 to 4n=7 to 8).

(m) Representative image
of fibronectin within the tumor.
Data are represented by mean ± SEM *P<0.
05**P< 0.
01****P<0.
001, and ****P < 0.
0001
.

Open in the viewer

To confirm that the inhibitory effect of glycosylation inhibitors on tumor progression is not due to potential side effects on tumor cells, but due to potential side effects on the tumor extracellular matrix, MET-1 cells isolated from MMTV-PyMT mice are treated with different concentrations of AG or ALT711 for 72 h
.
Quantify the percentage of cells with Ki67 to compare the proliferation of these cells ⁺ nuclei
.
Neither AG nor ALT711 treatment significantly disrupted cell proliferation (Panels S2, C, and D).

We also verified by comparing Ki67 that different treatment times had no significant effect on cell proliferation ⁺ nucleus density of cells treated with AG or ALT711 for 24 or 72 h at a concentration of 1000 mm (Figure S2E).

Noting that glycosylation, in addition to hardening the matrix, activates AGEs and their receptor (AGE-RAGE) signaling (7, 12, 13), we also observed the proliferation of MET-1 cells with normal or reduced RAGE expression on a supple or rigid matrix.

Quantification of cell density with Ki67 ⁺ nuclei showed that although short hairpin RNA (shRNA; Santa Cruz, sc-36375-SH) inhibition of RAGE expression significantly reduced cell proliferation, but cell proliferation increased with increasing matrix hardness (Figure 6F).
A similar effect was observed in cells embedded in a three-dimensional compliant or rigid collagen matrix, where hardness is controlled by glycosylation with varying amounts of glucose (Figure 6G).
These data suggest that glycosylation-mediated stromal sclerosis promotes cell proliferation regardless of activation of age-age
interactions.
EMT marker analysis showed that the EMT ratio of STZ tumors treated with glycosylation inhibitors was lower than in the control group treated with STZ (Figure 6, H and I).
Similarly, in diabetic tumors treated with inhibitors, levels of TGF-β- and fibronectin-positive regions were significantly reduced (Figure 6, J to M).
These results suggest that glycosylation inhibits the metastasis of the EMT spectrum of diabetic tumors to the epithelial phenotype
.
Overall, these results suggest that inhibition of glycosylation by AG or ALT711 treatment reduces tumor aggressiveness
in diabetic hyperglycemic mice.

Inhibition of FAK-mediated cellular mechanical sensing slows the progression of diabetic tumors without affecting ECM stiffness

To confirm that glycosylation promotes tumor progression by strengthening the tumor extracellular matrix (ECM), which in turn activates cellular mechanical sensing signals, STZ-injected mice were treated with the adhesion kinase (FAK) inhibitor PF573228 for 7 weeks until the end of
the study.
PF573228, a small molecule inhibitor of FAK phosphorylated at Tyr^{, 397}, is currently undergoing preclinical trials (25).
The inhibitory efficiency of PF573228 was confirmed by immunofluorescence staining, showing the concentration of phosphorylated FAK in diabetic tumors with or without FAK inhibitors (panel S2F).

Over time, the tumor volume of diabetic mice treated with FAK activation inhibitors (FAKi) was significantly lower than that of the diabetic control group (Figure 7A).
At the study endpoint, the average tumor volume of STZ-injected mice given FAKi was significantly lower than that of control mice injected with STZ-only (Figure 7B).
In addition, the average total tumor burden of FAKi-treated diabetic mice was significantly lower than that of the diabetic control group alone (Figure 7C).
Among diabetic tumors inhibited by FAK activation, Ki67-positive cells were significantly fewer than in diabetic control tumors (Figure 7, D and E).
These data suggest that inhibition of cellular mechanical induction can reduce the occurrence of primary tumors and inhibit cell proliferation
.
EMT marker analysis showed that the EMT ratio of STZ tumors treated with FAKi was lower than in the control group treated with STZ (Figure 6, F and G).
Similarly, levels of TGF-β- and fibronectin-positive regions in FAKi-treated diabetic tumors were significantly reduced (Figure 7, H to K).
These results suggest that inhibition of FAK activation can prevent the metastasis
of diabetic tumors along the EMT spectrum to the mesenchymal phenotype.

Figure 7: Inhibition of FAK reduces tumor progression without affecting the hardness
of the ECM.

(A) Weekly tumor volume measurement (N=3) for diabetic mice treated with or without FAK inhibitors.

(B) Average tumor volume at study endpoints (N=3 and n=6 to 10).

(C) Average tumor burden per mouse at study endpoint (N=3).

(D) Representative images showing Ki67 and cell nuclei within the tumor colocalization
.
(E) Quantitative analysis of percentage of Ki67 cells ⁺ nuclei (N=3 and n= 8,40 to 80 imaging fields).

(F) Typical images
of E-cadherin and vimentin in tumors.
(G) Corresponding quantitative studies of E-cadherin and waveform protein expression rates (N=3 and n=8, 40 to 80 imaging fields).

(H) Representative image
of TGF-β stained tumor sections.
(i) Corresponding quantification of TGF-β-positive area within the tumor (N=3 and n=6 to 8).

(J) Typical image
of fibronectin expression within tumors.
(K) Corresponding quantitative analysis of fibrectin-positive area in tumor tissue (N=3 and n=6 to 8).

(i) Effect of FAKi on blood glucose in diabetic mice (N=3).

(m) IHC-stained image
of representative tumor sections.
(N) Corresponding quantification of normalized age-positive area (N=3 and n=8).

(O) Modulus of elasticity of FAKi in combination with FAKi for diabetic tumors (N=3 and n=4584 to 882 measurements).

(P) AFM histogram
of the corresponding measurement.
Data are represented by mean ± SEM*P<0.
05**P< 0.
01****P<0.
001, and ****P < 0.
0001
.

Open in the viewer

We also confirmed that FAK activation inhibition had no effect
on blood glucose, glycosylation, and tumor extracellular matrix hardness in diabetic mice.
Mice blood glucose is measured weekly to verify that FAKi does not affect blood glucose throughout the experiment (Figure 7L).
To determine whether FAKi treatment affects age concentrations in STZ-injected mice, IHC staining
is performed.
There was no significant difference in age concentrations between STZ-injected mice and STZ-injected mice receiving inhibitors (Figure 7, M and N).
To compare the hardness of the tumor extracellular matrix, we performed AFM measurements
on tumor samples with and without FAK activation inhibition.
The distribution of mean modulus of elasticity and modulus of elasticity of tumors in diabetic mice treated with FAKi was similar to that of diabetic control mice (STZ only) (Figure 7, O and P).
These experiments showed that the inhibition of FAK activity did not affect the glycosylation reaction and hardness
of the tumor extracellular matrix.
In summary, these results suggest that inhibition of cellular mechanosensia can reduce tumor aggressiveness in diabetic hyperglycemic mice without affecting the hardness
of the tumor extracellular matrix.
These data suggest that cellular mechanosensory involvement in diabetic hyperglycemia-mediated glycosylation promotes breast tumor progression
.

The extracellular matrix stimulates tumor cell proliferation and promotes breast cancer cell proliferation

Our in vivo data suggest that hyperglycemia-mediated glycosylation increases the hardness of the extracellular matrix and promotes the progression
of breast tumors.
To validate this finding clinically, we quantified age concentrations within the stroma of diabetic and non-diabetic human breast tumors using IHC staining (Figure 8, A and B) and measured matrix stiffness with atomic force microscopy (Figures 8, C and D).
Early breast tumors isolated from diabetic patients (stage I/II) have higher age concentrations and higher ECM stiffness
than patients without diabetes.
In addition, consistent with our in vivo data, there were significantly more Ki67-positive cells in tumors in diabetics than in non-diabetic patients (Figure 8, E and F).
Consistent with this finding, tumors with diabetes had significantly higher EMT ratios, TGF-β, and fibronectin levels (Figure 8, G to L).
Together, these data suggest that tumors in diabetic breast cancer patients have more glycosylation, increase extracellular matrix stiffness, increase cell proliferation, upregulate EMT, support our in vivo mouse data, showing that diabetic hyperglycemia-mediated glycosylation promotes breast cancer progression
through stromal sclerosis.

Figure 8 Diabetes activates glycosylation, promotes cell proliferation, and alters EMT
within tumors in breast cancer patients.

(A) Representative images
of tumor slices extracted from breast cancer patients diagnosed with diabetes or non-diabetic breast cancer.
(B) Corresponding quantification of normalized age-positive area (N=4).

(C) Measurement of elastic modulus of tumors in diabetic and non-diabetic patients by atomic force microscopy (N=4478 to 482 measurements).

(D) Corresponding histogram
of AFM measurements for diabetic or non-diabetic tumors.
(E) Representative images showing Ki67 colocalization with nuclei within tumors
extracted from breast cancer patients diagnosed with diabetes or non-diabetic.
(F) Quantitative analysis of percentage of Ki67 cells ⁺ nuclei (N=4, including 40 imaging fields).

(G) Immunofluorescence staining showing typical images of E-cadherin and waveform proteins in tumors of diabetic or non-diabetic patients (N = 4, including 40 imaging fields).

(H) Quantification of EMT ratios (N=4, including 40 imaging fields).

(i) Representative image
of TGF-β stained tumor sections extracted from diabetic and non-diabetic patients.
(J) Corresponding quantification of TGF-β-positive area within the tumor (N=4).

(K) Typical image
of fibronectin expression within tumors.
(i) Corresponding quantitative analysis of fibrectin-positive area in tumor tissue (N=4).

Data are represented by mean ± SEM*P<0.
05**P<0.
01, and ****P<0.
0001
.

Open the discussion in the viewer

Here, we found that diabetic hyperglycemia promotes breast tumor progression
through glycosylation-mediated extracellular stromal sclerosis.
The drug inhibition of glycosylation is sufficient to reduce tumor progression
in diabetic mice independently of hyperglycemic states.
We established a novel mouse model of diabetic hyperglycemia in female MMTV-PyMT mice by STZ injection and high-fat diet prior to spontaneous breast tumorigenesis
.
Using this model, we discovered and described a previously unknown role of non-enzymatic glycosylation in promoting tumorigenesis under hyperglycemic conditions
.
Compared to non-diabetic tumors, diabetic mouse tumors were stiffer, more proliferative, and advanced in the EMT spectrum, all of which were associated with worsening prognosis (26–28).
By using AGE inhibitors, we mechanistically demonstrated that AGE-mediated cross-linking uniquely promotes tumorigenesis
in hyperglycemic mice.
This mouse model shows great potential
in further exploring the synergy between diabetes and cancer progression.
Our findings highlight the importance of considering the interplay between diabetes and cancer, both of which are among the top 10 causes of death in the United States (29).

The effects of glycosylation and AGEs on chronic hyperglycemia are profound and have been linked to many diabetes-related complications, including retinopathy, neuropathy, nephropathy, and cardiomyopathy (30).
While other studies have focused on the chemical signaling role of RAGEs, RAGEs, in cancer progression, we have also focused on mechanistic non-enzymatic glycosylation, where AGEs are biologically active byproducts of this reaction (31, 32).
To analyze the effects of mechanical crosslinking and AGE signaling, we treated with two inhibitors, AG and alagebrium, which affect different steps
of the non-enzymatic glycosylation reaction.
Since AG inhibits the formation of AGE, this inhibitor simultaneously blocks the chemical and mechanical effects of non-enzymatic glycosylation (33).
Alagebrium, as a crosslinking breaking agent for AGEs, allows AGEs to form but removes crosslinks, effectively allowing chemical signaling while counteracting the mechanical effects of non-enzymatic glycosylation (34).
We found that the effect of AG and alagebrium treated mice with diabetes was similar across all indicators, and our findings suggest that age-mediated ECM sclerosis plays a key role in the increased aggressiveness of breast tumors in hyperglycemic mice, which cannot be compensated
for by AGE-RAGE signaling alone.
An important question that we have not addressed in this work is whether the enhancement of tumor cell proliferation we observed in hyperglycemic mice is primarily driven directly by increased ECM stiffness, or is caused by increased tumor growth pressure due to ECM stiffness, which also induces protocarcinoma signaling in vitro and in vivo (35, 36) .
Further research to elucidate the exact mechanism of cellular mechanical sensing will help to discover more targeted treatments that will benefit
patients with these highly prevalent comorbidities.

Age is associated with many diseases, including diabetes, cancer, atherosclerosis, rheumatoid arthritis, bone fragility fractures, kidney failure, Alzheimer's disease, and premature aging (37, 38).
While efforts have been made to develop and test a variety of anti-aging drugs, clinical trials have yielded mixed results (38, 39).
Due to our study and the large amount of experimental and clinical evidence that the prognosis of breast cancer patients with diabetes deteriorates, this subset of diabetic cancer patients may reap unique benefits from adding age-suppressing therapy to traditional treatment strategies (40).
A commonly used drug, metformin, which lowers blood sugar, is a first-line treatment for type 2 diabetes and is also thought to suppress age (41).
Diabetes patients who used metformin had approximately one-third lower cancer incidence and mortality compared with those who used other antidiabetic drugs (42).
In addition to diabetes, metformin has recently shown incredible promise as an adjunct therapy and standard of care treatment (42).
Given our findings and the benefits of metformin in diabetes and cancer treatment, the synergistic effects of breast cancer and diabetes should be further investigated and the potential complementary effects
of other age-suppressing drugs tested.

Materials and methods: antibodies and reagents

The antibodies used for staining in Western blot and IHC experiments are as follows: anti-aging antibody (ab23722, Abcam, Cambridge, UK), antifibronectin antibody (SC-9068 and SC-6953, Santa Cruz Biotechnology, Dallas, Texas, USA), anti-transforming growth factor β antibody (ab92486, Abcam).
and antiglyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibodies (Poly6314; BioLegend, San Diego, California, USA).

Primary and secondary antibodies for immunofluorescence staining include anti-wave protein antibodies (ab92547, Abcam), anti-polyomavirus medium T antigen antibodies (PyMT; ab15085, Abcam), anti-E-cadherin antibody (7870, Santa Cruz Biotechnology), anti-Ki67 antibody (14-5698-82, Thermo Fisher Scientific, Waltham, MA, USA), anti-Alex Fisher antibody, Alex Fisher antibody, Alex 047, Alex Fisher antibody
.
All other reagents used here are purchased from Thermo Fisher Scientific Technologies
.

MMTV-PyMT transgenic mouse study

FVB/N-Tg (MMTV-PyVT) 634Mul/J (MMTV-PyMT) mice
were used in this experiment.
All mice are housed
according to the protocol approved by the Vanderbilt University Animal Care and Use Committee (protocol number: M1700029-01).
From 4 weeks of age, female MMTV PyMT mice with a background of FVB strain are fed with D12492 high-fat feed (NC0004611, Research Diete, New Brunswick, NJ) (Jackson Laboratory, Bal Harbor, New Jersey, USA).

To induce diabetes, mice were given 5 consecutive doses of STZ (miliporesigma) intraperitoneally at a dose of 70 mg/kg body weight
.
STZ is a small molecule that has been shown to selectively destroy islet cells and induce diabetic hyperglycemia in mice (43).
Control mice are injected with citrate buffer [0.
1 mm (pH4.
5); MilliporeSigma] as a solvent control
.
Before starting STZ injection, mice are measured weekly for body weight, tumor volume, and glucose levels until the mice are euthanized
.
The tumor volume of MMTV-PyMT transgenic mice is calculated as follows: π× length ²× width/12(44).
Mice with blood glucose levels above 400 mg dl^{? 1} Considered diabetic.

Further quantitative studies
were performed on the presence of islet cells in IHC.

For STZ-treated mice receiving insulin, treatment with insulin glargine (milliporesiga) is performed by a subcutaneous insulin pump (ALZET microosmotic pump with 0.
5 U insulin per day per mouse, model 2006) for 7 weeks
.
The day after diabetic hyperglycemia is diagnosed (week 6), an osmotic insulin pump is implanted (45).
For STZ-injected mice receiving AG, mice are treated
with 3 mg/kg body weight AG (MilliporeSigma) in drinking water prior to the study endpoint.
For STZ-injected mice treated with alagebrium (ALT711), 1 mg/kg of body weight of alagebrium (medchexpress, Monmouth Junction, NJ, USA) was administered intraperitoneally daily until the end of
the study.
For STZ-treated mice receiving PF573229, 50 μM PF573228 (MedChemExpress) is injected subcutaneously daily until the end of the study (46).

7 weeks after STZ injection, mice are humanely euthanized by _CO2 and autopsied
.
Breast tumor specimens are collected and snap-frozen in 70% ethanol dry ice slurry or fixed
with 4% (v/v) paraformaldehyde phosphate buffer (PBS).
This is followed by immunofluorescence and immunohistochemical staining in 5 μm thick sections and AFM
at 20 μm.
Fixed tissue sections are 5 μm thick and picrosirius stained for collagen quantification
.
Tumor aggressiveness was further assessed with hematoxylin and eosin staining, and tumors were then graded
by a blind veterinary pathologist.
Female mice are resistant to STZ-induced diabetic hyperglycemia (47).
Noting the difficulty of inducing diabetic hyperglycemia in female mice, we also tested one of several other commonly used methods STZ injection (43, 48, 49) on our MMTV-PyMT mice Our results show that these STZ injection regimens either do not raise blood glucose at all (panels S1, A to F) or raise blood sugar at a delayed rate, so diabetes levels are not reached until week 8 when the tumor has begun to progress (Panels S1, G and H).

(50).

Glucose and insulin tolerance measurement

Glucose tolerance and insulin tolerance are
performed within 1 week after STZ injection.
8 h before the start of the experiment, mice fast
.
In the glucose tolerance test, mice are intraperitoneally injected with 1 g/kg body weight with sterile 10% glucose (miliporesigma).

In the insulin resistance test, insulin (Indianapolis Erie, USA) is injected
intraperitoneally at 2 U/kg body weight.
Blood
is collected before and 15, 30, 60, 90, and 120 minutes before and after dosing.

cell culture

MET-1 cells were extracted from MMTV-PyMT mice and cultured in Dulbecco modified Eagle medium supplemented with 10% (v/v) fetal bovine serum (FBS) and 1% (v/v) penicillin streptomycin (20).
Keep cells at 37 °C in humidified air with 5% (v/v) _CO2 in the air
.
Cells were tested for mycoplasma contamination and experiments were performed
under mycoplasma negative conditions.
For glycosylation inhibition studies, cells are seeded on collagen-coated coverslips and treated with AG or ALT711 at a concentration of 10, 100, or 1,000 μM for 24 or 72 h
.
In cell proliferation assays, cells with normal or reduced RAGE expression are seeded onto collagen-coated polyacrylamide matrices or embedded in
collagen matrices with different glucose glycation.
In comet assays and TUNEL assays, MET-1 cells are seeded on collagen-coated coverslips and treated with STZ (0.
875 mg/ml) for 5 days
.
3.
7% (v/v) paraformaldehyde was fixed in PBS for 10min
at room temperature.
Cells are kept in
PBS at 4 °C prior to downstream experiments.

immunofluorescence

Tumor tissue sections (+/?) were extracted from MMTV-PyMT mice or MET-1 cells RAGE shRNA) was treated with AG or ALT711 and fixed in PBS with 4% (v/v) paraformaldehyde for 10 min
at room temperature.
After fixation, samples are washed with PBS and infiltrated
in PBS with 1% (v/v) Triton X-100 (Thermo Fisher Scientific).
After infiltration, tissue sections are closed with 10% (v/v) FBS, 5% (v/v) donkey serum, and 5% (v/v) goat serum for 2 h
at room temperature.
It was then diluted 10% (v/v) FBS, 5% (v/ v) Donkey serum, and 5% (v/v) goat serum overnight
at 4 °C.
After washing with PBS with 0.
02% Tween 20 added, then diluted 1:100 in PBS with secondary antibodies (including goat anti-rat Alexa Fluo 647 antibody (A21247, Thermo Fisher Scientific) and donkey anti-rabbit Alexa Fluor 568 antibody (A10042, Thermo Fisher Scientific) diluted 1:100 in PBS with 10% (v/ v) FBS, 5% donkey serum, 5% goat serum protected from light for 1 h
at room temperature.
Immunofluorescence images were obtained by using a ZEISS LSM800 confocal microscope, a 40×1.
1 numerical aperture (NA) immersion objective and 405, 568, and 640 excitation laser lines to obtain immunofluorescence images z-overlay imaging
.
Protein colocalization analysis as described above (51).
Briefly, the corresponding filtered images are superimposed and the overlapping area
is measured.

Western blot analysis

Quick-frozen tumors extracted from mice are ground with pre-cold (?).
80 °C) mortar and pestle
.
The crushed tumors are dissolved in Laemmli buffer and centrifuged at 14,000 g for 10 min
at 4 °C.
Collect the supernatant and use it for western blotting
.
Measure protein concentration using a DC detection kit (Bio-Rad), perform gel electrophoresis [8% (w/v) acrylamide gel], and transfer to a polyvinylidene fluoride membrane as described above (52).
The cell membrane
is then closed with 5% bovine serum albumin (miliporesigma) in tris-buffered saline (TBS).
Then incubate overnight with primary antibody at 4 °C and wash
with TBS with 0.
1% Tween 20.
After washing, stain with secondary antibodies for 1 h
at room temperature.
Prepare primary antibodies
in 5% bovine serum albumin diluted 1:1000.
Prepare horseradish peroxidase secondary antibody
by 1:2000 dilution.
Using the ImageQuant LAS-4000 system, the films were imaged
using either West Pico or West Dura (Thermo Fisher Scientific) according to their respective protocols.
Quantitative analysis
with ImageJ [National Institutes of Health (NIH)].
Protein expression levels of interest are expressed
as the ratio of the protein of interest to GAPDH.

Atomic force microscopy

Extracellular matrix stiffness
of tumors in PyMT mice or breast cancer patients was measured by contact atomic force microscopy (MFP-3D, refuge study).
During the measurement, sections are made at a thickness of 20 μm and cultured
in PBS with protease inhibitor cocktail (Thermo Fisher Scientific) added.
Silicon nitride cantilever with a nominal spring constant of 0.
06 N m^{? 1} Spherical borosilicate glass beads with a diameter of 5 μm (Novascan, Boone, IA, USA)
are used.
Probe spring constants
are measured before each treatment.
Tissue displacement maps were obtained by two force maps spaced at 10 μm intervals and two force plots spaced 10 μm apart¹ Until the maximum setting force of 3nn
is reached.
To calculate the modulus of elasticity for each indentation, the force-displacement curve is fitted to a Hertz model
assuming a Poisson's ratio of 0.
5.

Immunohistochemistry

MMTV-PyMT mouse tumor tissue sections were fixed in PBS with 4% (v/v) paraformaldehyde for 10 min
at room temperature.
After overnight incubation with primary antibodies at 4 °C, endogenous peroxidase in the samples is oxidized at 0.
3% _H2O2 for 15 min
at room temperature.
Then dilute in TBS at 1:400 with secondary antibody, horseradish peroxidase-bound anti-rabbit-goat antibody (611-103-122, Rockland Immunomicals, Limerick, PA, USA) at room temperature and incubate for 1 h
at room temperature.
Samples were then cultured with 3,3'-diaminobenzidine (DAB) (cell signaling technology) and counterstained
with Meyer hematoxylicin.
After dehydration and mounting, tissue sections are digitized
with the Leica SCN400 slide scanner.
The signal strength and percentage of positive area of the signal were quantified
using the Leica digital image center.

Comet experiment

The comet test was conducted according to the scheme described earlier (19).
MET-1 cells suspended in 37 °C PBS for 5 days with STZ were treated with STZ for 1% low-gel-temperature agarose (Sigma-Aldrich; Type VII, catalog number A-4018) and immediately pipette onto a slide with a 1% agarose layer
.
After agarose geling, first soak the slides in an alkaline solubilization solution at 4 °C overnight and then rinse for 20 min
at room temperature.
After rinsing three times, transfer the slides to an electrophoresis cell for 20 min at a voltage of 0.
6 V/cm
.
Rinse with 20 ml of sodium iodide solution under the microscope and rinse 1.
7 ml of glass slides with 20 ml of sodium iodide solution for Z-overlay imaging
.
The tail moment representing DNA damage was quantified using the Comet Analysis IV software (Instem, Staffordshire
, UK).

Notation

The TUNEL assay was performed using the Click-iT Plus TUNEL Assay Kit (C10617, Thermo Fisher Scientific Technologies) and following the
manufacturer's instructions.

DNA treated cells are used as a positive control
for detection.
Using the ZEISS LSM700 confocal microscope, imaging is performed using a 20×/1.
1Na immersion objective and z-overlay
.

Collagen deposition test

Hydrate and stain fixed tumor sections collected from PyMT mice with the picric red staining
kit (24901-250, Polysciences Inc.
) according to the manufacturer's instructions.
After dehydration and installation, the sections were examined quantitatively by polarization microscopy using an inverted axis microscope equipped with a rotatable linear polarizer and a circular polarizer with a ZEISS Axiocam 506 color camera (52).
Optical delay was quantified using pre-established methods (52).
Simply put, the average delay rate of a region of interest is randomly selected from each image to quantify the collagen content
in the tumor section.

Unconfined compression test

As described earlier, thaw Snap frozen tumors from PyMT mice in PBS immediately prior to performing the mechanical assay with a protease inhibitor cocktail (21).
The mechanical properties of the tumor were measured with the taelectroforce model 3100 (TA instrument
).
The tissue is subjected to 15% strain and 3% gradual displacement over five steps
.
Each step consists of 1 mm indentation and 15 minutes of relaxation time
.
A custom MATLAB script is used to calculate the equilibrium modulus based on the slope of the stress-strain curve generated by the
porous elastic model.

Breast tumor analysis

To investigate the relationship between glycosylation and tumor progression, tumor specimens from patients with diabetic or non-diabetic breast cancer were obtained from a network of partner human organizations according to the protocol approved by the Institutional Review Board (IRB 201314 Vanderbilt University Medical Center (IRBs), see Table S1 for corresponding clinical information
.
Take frozen breast tumors with a section thickness of 8 μm for immunofluorescence staining and immunochemical staining, and 20 μm sections for AFM
.

Statistical analysis

Statistical analysis
using GraphPad Prism 8.
0a (GraphPad software, Lahora, California, USA).
The data here is expressed
as an average ±SEM.
The two-tailed t-trial was used to compare
the two groups.
When more than two groups were analyzed, parametric one-tailed or two-tailed one-way analysis of variance (ANOVA) was used where appropriate, followed by the Mann-Whitney test or post-event Tukey test
.
P<0.
05 is considered statistically significant
.

This article is an English version of an article which is originally in the Chinese language on echemi.com and is provided for information purposes only. This website makes no representation or warranty of any kind, either expressed or implied, as to the accuracy, completeness ownership or reliability of the article or any translations thereof. If you have any concerns or complaints relating to the article, please send an email, providing a detailed description of the concern or complaint, to service@echemi.com. A staff member will contact you within 5 working days. Once verified, infringing content will be removed immediately.