Interleukin-6 Induced Acute Phenotypic Microenvironment Promote...（四）
Flow cytometry
Spleens were extracted after therapy and mechanically homogenized with gentle MACS Dissociator to obtain a single-cell splenocyte suspension. Erythrocytes were removed with 1× red blood lysis buffer (1.5 mM NH4Cl, 120 mM NaHCO3, 1 mM Na2-EDTA). Single-cell suspensions were stained with the following Abs: FITC-anti-mouse- CD3, PE-anti-mouse-CD4, APC-anti-mouse-CD8, PerCP-Cy5.5-anti-mouse-CD26, Alex fluor 488-antimouse- CD11c, PerCP-Cy5.5-anti-mouse-MHCII, PE-anti-mouse-CD86, PE-anti-mouse-Gr1 and FITC-anti-mouse-CD11b. Abs were obtained from eBioscience (San Diego,CA) and BioLegend (San Diego, CA). Samples were measured with FACSAria II (BD Biosciences, Mountain View, CA) and analyzed using FlowJo analysis software (Tree Star, Ashland, OR).
Results
Cryo-thermal therapy significantly improved cancer survival rate and reduced metastasis
In present study, we applied cryo-thermal therapy on spontaneous metastasis murine breast cancer xenograft model to examine its efficacy (Figure 1A; Materials and Methods). 21 days post tumor inoculation, metastases started to be observed in lungs (Figure 1D).
Figure 1. The schematic graph of cryo-thermal therapy using murine breast tumor model and the therapeutic efficacy. (A) Balb/c female mice were s.c. injected with 5×105 4T1 cells and received treatment after 3 weeks. (B) Kaplan-Meier survival curve. Survival curves were compared using log-rank tests. ***p<0.0001. (C) Tumor growth curves. (D) Cryo-thermal therapy could effectively inhibit or reduce tumor metastasis to lung. Data are shown as mean ± SD. **p < 0.01, ***p < 0.001 by two-way ANOVA with the Bonferroni correction. n=14 per group.
In this situation, 14 mice underwent cryo-thermal therapy and 14 without treatment as the control. Healthy mice were taken as “normal” group. In mice treated, significantly prolonged survival rate (64%) was achieved, however, all of the untreated mice died within 50 days after tumor inoculation (Figure 1B).Tumor growth concomitant with high mortality was observed in untreated group, indicating its poor prognosis, whereas in therapeutic group, tumor shrank immediately after therapy, scared over time and consequently healed without tumor recurrence (Figure 1C, Additional File 7: Figure S1). Moreover, dramatic increase of metastasis was observed in control mice accompanied by severe lung swelling. By contrast, metastasis was extensively limited or undetectable after therapy, and meanwhile lung’s volume was similar to that of healthy mice (Figure 1D). Taken together, these results suggested that local cryo-thermal therapy provided a promising strategy to hamper primary tumor progression, eradicate or attenuate the development of distant metastasis, and therefore contributed to a good outcome of survival.
Shotgun proteomics identification and quantification of serum glycosylated proteins
To comprehensively investigate the underlying tumor suppression mechanisms by cryo-thermal therapy, a time course study over 8 time points was carried out to compare the serum proteome changes between post- and non-treatment mice (Figure 2). Serum samples under each condition (n=5~8) were pooled and analyzed using iTRAQ-based quantitative proteomics technology focusing on N-linked glycoproteome [25]. By applying the hydrazide chemistry to specifically enrich N-glycosylated proteins, more than 90% of the identified peptides, corresponding to 96% glycoproteins under each condition contained the N-glycosylation signature of NXS/T motif, suggesting a high specificity of the enrichment in serum samples (Additional File 7: Figure S2A and B). In total, we identified 3002 unique peptides corresponding to 308 proteins at a false discovery rate (FDR) of 1% (Figure 3A left panel, Additional File 7: Figure S2D left panel, Additional File 3: Table S3). By projecting these identified proteins into the published Peptide-Atlas database [34], we found that their detectable dynamic range was largely enhanced (7 vs 4 orders of magnitude in crude plasma[35]), with the minimal detection level of ng/mL (Figure 3B). Of the identified proteins, 241 were annotated as N-glycoproteins, 231 of which were quantified (Figure 3A right panel, Figure S2D right panel, Additional File 4: Table S4). We analyzed the subcellular localization of the 231 quantified proteins and there were 118 (51%) and 59 (26%) proteins categorized as “extracellular” and “plasma membrane” respectively (Additional File 7: Figure S3A), which fit well with the properties of glycoproteins and is consistent with the composition of serum system. To better profile the proteome change in response to cryo-thermal therapy, protein ratios between treated and untreated mice were calculated and transformed to the logarithm scale which fitted a normal distribution (Figure S3B). Proteins with ratios outside the range of mean ± 2σ at one or more time points were considered as significant proteins in the following data mining. Overall, a total of 53 significant glycoproteins were determined across the entire time course, and 11 on average were obtained for each time point (Additional File 5: Table S5)
Acute phase response was the most enriched pathway in response to cryo-thermal therapy
After obtaining 53 proteins significantly changed during the time course, we first performed an enriched pathways analysis using Ingenuity Pathway Analysis (IPA) software to reveal the significantly affected canonical pathways. Acute phase response (p value = 3.64E-06) was the most enriched pathway. Interestingly, all of the significant proteins were elevated on the 2nd day post therapy (Additional File 7: Figure S4A). However, such up-regulated expression turned to be down-regulated (Figure S4B) on the last two days.
Quantitative time course mmeasurement of induced proteome changes
Next, to get a better visualization of a global proteome change over the time course, we performed an unsupervised hierarchical clustering according to the logarithm ratios on 34 significant proteins with no missing values. As shown in Figure 4A, hierarchical clustering distinguished the proteome to early and late response, in which before day 2, they appeared to be similar in expression patterns, however, at the other six time points, they showed a markedly distinct profile. Such differential expression revealed that the overall response to cryo-thermal therapy was time-characterized. In addition, proteins were hierarchically clustered into 4 groups, annotated with different biological functions. Besides the ‘acute phase response’ in cluster ‘c’, proteins contained in cluster ‘a’ were mostly up-regulated during the late time points. They are broadly associated with tumor progression. Proteins grouped in cluster “b” are mostly involved in carbohydrate and lipid metabolism. These proteins were commonly unchanged in the early time periods while elevated on the 28th day. Proteins clustered in ‘d’ were partially associated with leukocyte movement, and they are mostly down-regulated during the late stage. Overall, the integrated analysis of pathway and supervised clustering revealed that cryo-thermal therapy induced diverse biological responses and they were time-dependent. In terms of the timeframe, acute phase response was the first-end event, encompassing all of the biological responses. These findings implied that acute phase response played an important role in modulating tumor progression and carbohydrate/lipid metabolism (Figure 4B).
