| Purpose:Proteasome inhibitor bortezomib, as a novel class of antitumor drugs targeting multiple molecular pathways, has been approved by FDA for the treatment of multiple myeloma and mantle cell lymphoma. It is still in its phase of clinical trial for treating solid tumor. It is key to clarify the critical antitumor biological mechanism, recognize the early response evaluation index, select the patient responsive to bortezomib treatment for a successful tumor target therapy. Preferential cytotoxicity of bortezomib was found toward hypoxic tumor cells and hypoxic endothelial cells in vitro. Meanwhile, tumor hypoxia response was significantly suppressed by bortezomib in both patients with metastatic colorectal cancer and multiple kinds of xenograft models. In this study, we were aimed to1) investigate the role of pretreatment tumor hypoxic microenvironment in the antitumor effect of bortezomib and its effect on tumor microcirculation at histological level,2) and explore the feasibility of using DCE MRI to indirectly evaluate the antitumor biological effect of bortezomib through monitoring the changes in tumor microcirculation.Methods and materials:Established human colorectal carcinoma HT29cell containing dual hypoxia response reporter fusion gene herpes simplex virus type1thymidine kinase and enhanced green fluorescence protein (HSV1-TKeGFP), under the control of nine copies of hypoxia response element (9HRE), designated as HT29-9HRE, and its xenograft model were used.1) In vitro:to verify the suppression of hypoxia response by bortezomib in HT29-9HRE cell. Under normoxic (21%O2) or hypoxic (0.2%O2) condition, the expression of hypoxia response reporter gene eGFP was detected using flow cytometry, the function of TK by the uptake of its substrate [14C]2’-Deoxy-2’-fluoro-β-D-arabinofuranosyl-5-iodouracil (14C-FIAU) using a Wallac1410liquid scintillation counter, the expression of hypoxia-inducible factor-1α (HIF-1α) and its downstream gene product carbonic anhydrase9 (CA9) by western blot, vascular endothelial growth factor (VEGF) by enzyme-linked immunosorbent assay. Each experiment was repeated at least thrice.2) In vivo:to clarify the relationship between pretreatment tumor hypoxia microenvironment and the antitumor effect of bortezomib. Nude mice bearing subcutaneous HT29-9HRE xenografts on the right hind limbs were randomly assigned to bortezomib (2mg/kg) or drug vehicle control,8-10mice per group. The mice were i.v. given the first exogenous hypoxia marker pimonidazole2h before treatment, the second hypoxia marker EF520h post treatment, blood perfusion marker Hoechst3334224h post treatment followed by sacrificed1min later. The tumor was immediately dissected, embedded, frozen, fixed, stained with routine or fluorescent immunohistochemical staining and examined by fluorescent microscope. The following analyses were done:i) to evaluate the antitumor effect of bortezomib in HT29-9HRE xenograft and its effect on tumor blood flow at microscopic level-qualitatively and quantitatively analyze the overall apoptosis percentage and the distribution of apoptosis marker cleaved caspase3in tumor cells and endothelial cells at24h after bortezomib treatment, and observe the histological characteristics of microvasculature with apoptotic endothelial cells and the effect of apoptosis in endothelial cells on vessel function, i.e. tumor blood flow; ii) to analyze the distribution of tumor hypoxia, its dynamic change, hypoxia response, and the local microcirculation in HT29-9HRE xenografts-the entire tumor microscopic images were subdivided into a batch of1x1mm2micro-regions. The pretreatment hypoxia level in these regions were arbitrarily classified into three grades, mild (<10%), moderate (>10%), and severe (>20%), based on the staining of the first hypoxia marker pimonidazole. The dynamic changes in tumor hypoxia, the hypoxia response, and the microcirculation in these micro-regions were evaluated by comparing the distributions of the second hypoxia marker EF5(with22h time intervals between two hypoxia markers), hypoxia response product CA9, and perfusion marker Hoechst33342with that of pimonidazole; iii) to evaluate the role of pretreatment tumor hypoxia microenvironment in the pro-apoptosis effect of bortezomib and the relationship with its inhibition effect on tumor hypoxia response-quantitatively analyze the apoptosis percentages in total, in tumor cells, and in endothelial cells in each micro-regions with variable pretreatment hypoxia level, and clarify the relationship among pro-apoptosis, tumor pretreatment hypoxia, and hypoxia response by comparing the distribution of apoptosis with the expression of CA9.3) Dynamic contrast enhanced magnetic resonance imaging (DCE MRI):to noninvasively monitor the dynamic changes in tumor blood flow with bortezomib treatment. Nude mice bearing subcutaneous HT29-9HRE xenografts on the right flank were performed tumor local baseline DCE MRI using a home-built, solenoidal1H MR coil. And then they were randomly assigned into single dose treatment (2mg/kg), split dose treatment (1.5mg/kg,24h interval) and drug vehicle control groups,6-8mice per group. Repeated DCE MRI were performed at24h and48h post first treatment, to evaluate the sequential effect on tumor microcirculation by continuous treatment of bortezomib and the subsequent effect by single dose treatment of bortezomib, respectively. Time-signal intensity curve for each voxel was obtained from DCE MRI images. The Hoffman model was used to estimate Akep values, the product of amplitude of the curve (A) and the exchange rate (kep), of individual voxels from the dynamic build up curves, and Akep maps were generated for the corresponding tumor slices. The median value of Akep of the whole tumor represents the average tumor blood flow, and the Akep map shows the distribution of blow flow inside the tumor.4). Combining with the ex vivo experiment, to explore the feasibility of using DCE MRI monitor tumor blood flow to indirectly reflect the antitumor biological effects of bortezomib. After the last DCE MRI examination, nude mice were given hypoxia marker pimonidazole injection, followed by perfusion marker Hoechst33342injected2h later and sacrificed1min later. The abovementioned histological experiment was repeated again, and further evaluated the distribution of tumor blood flow, the changes in tumor hypoxia, and the status of hypoxia response for each group. By comparing these ex vivo histological results with the last DCE MRI results, it could be decided whether DCE MRI has the ability to predict the antitumor biological effects of bortezomib.Results:1) In vitro experiments showed that bortezomib accumulates HIF-1a in a non-functional form and blocks the effective HIF mediated hypoxia response in HT29-9HRE cells. With bortezomib treatment, the degradation of HIF-1a was blocked under normoxic condition. But the expression of hypoxia response reporter gene product eGFP and the function of TK in hypoxic HT29-9HRE cells were reduced to the level of those maintained under normoxic condition; the hypoxia response downstream transcriptional target gene products CA9and VEGF were also decreased to normoxic level. All these results suggested that bortezomib suppresses hypoxia response and reduces its protection ability from hypoxic burden in this HT29-9HRE cell.2) In vivo experiments:i) The overall apoptosis percentage was significantly increased and the blood perfusion was dramatically decreased in HT29-9HRE xenografts with bortezomib treatment. The positive cleaved caspase3percentages were5.26%±0.98%and0.58%±0.13%(P<0.001) in treated group and drug vehicle control group, respectively. More apoptosis signals were found in endothelial cells (3.28%±0.63%) than those in tumor cells (1.98%±0.41%, P<0.05). Morphological examination shows that the micro vasculatures with apparent apoptosis in endothelial cells were usually in dot or short line shape, without a regular lumen formation and blood perfusion. The overall tumor blood perfusion was also dramatically decreased, with the Hoechst33342positive percentage of3.57%±0.83%in treated group vs.18.72%±2.59%in control group (P<0.001), respectively, ii) The characteristics of tumor hypoxia, dynamic change, blood flow, and hypoxia response in untreated HT29-9HRE xenografts:the distribution of tumor hypoxia is heterogeneous. In mild and moderate hypoxic micro regions, micro vessels were usually with moderate perfusion function, less de novo hypoxic cells were developed during22h time intervals, and the tumor cells have the fully hypoxia response ability to bear hypoxic burden. In severe hypoxic micro regions, there was almost no micro vessels with perfusion function, and more de novo hypoxic cells developed during this time interval with somewhat damaged hypoxia response, iii) After bortezomib treatment, apoptosis signals were found preferentially located in moderate and severe pretreatment hypoxic micro regions. The cleaved caspase3positive percentage in these regions were7.54%±1.31%and7.92%±1.12%, respectively, compared to that in mild hypoxic micro regions (1.08%±0.32%, P<0.001). The activation of apoptosis was found in both tumor cells and endothelial cells in pretreatment moderate and severe hypoxia region, with significantly increased tumor hypoxia, and the expression of protective hypoxia response target gene product CA9was significantly inhibited in both original and newly developed hypoxic cells. In pretreatment mild hypoxic regions, the apoptosis signals were only found in a few endothelial cells in dot or short line shape, without the increases in tumor hypoxia and interferences with hypoxia response as plenty CA9expression in tumor cells.3) DCE MRI examinations showed that after single dose bortezomib treatment, the tumor blood flow were significantly decreased in both central and peripheral tumor region at24h post treatment, and partially recovered in the central region at48h post treatment. The median Akep values of pre,24h post, and48h post treatment were0.072±0.017,0.018±0.010(P<0.05), and0.068±0.018/min, respectively. In split dose of bortezomib treatment group, the tumor blood flow were continuously decreased in both central and peripheral regions at24h and48h post treatment. The median Akep values of pre,24h post, and48h post treatment were0.072±0.017,0.037±0.013, and0.022±0.010/min (P<0.05), respectively.4) Comparing with the last DCE MRI results, the ex vivo histological experiment validated the effect of bortezomib on tumor blood flow is synchronized with its inhibitions effects on tumor hypoxia response. The distribution of tumor blood flow at50h post single dose of bortezomib treatment were observed in both central and peripheral tumor regions, suggesting that the effect of bortezomib on tumor blood flow has decreased or vanished. The extent of tumor hypoxia were reduced without changes in hypoxic intensity (staining intensity). The expression of protective hypoxia response target gene product CA9was also recovered approximately to the level of untreated control group. In split dose of bortezomib treatment group, the tumor blood flow as detected by Hoechst33342positive signal was significantly decreased, with only few blue fluorescent signal seen in the peripheral tumor regions. The staining intensity of hypoxia marker pimonidazole was significantly increased, suggesting the deteriorated hypoxia status due to the significant decreases of tumor blood flow. The expression of CA9was also dramatically suppressed in these regions. All these results suggested that it is feasible to use DCE MRI to indirectly predict the antitumor effects of bortezomib by noninvasively monitoring the changes in tumor blood flow.Conclusions:Pretreatment tumor hypoxia microenvironment plays a critical role in antitumor effects of bortezomib. The endothelial cells and tumor cells under pretreatment moderate and severe hypoxic micro regions were the major apoptotic populations induced by bortezomib treatment, which closely correlated with its inhibition effects on tumor hypoxia response. Especially, tumor endothelial cells in immature, poorly developed vasculatures were more sensitive to bortezomib, and easily induced apoptosis, leading to the collapse of local micro vessels, devoid of local blood perfusion, and further deteriorated local hypoxia. DCE MRI could effectively monitor the changes in tumor blood flow, thereby indirectly reflect the antitumor biological effect of bortezomib.Significance:This study revealed the important role of pretreatment tumor hypoxia microenvironment in bortezomib antitumor effect, and provided a substantial foundation for improving bortezomib therapeutic effect in clinical treatment by fully taking advantage of hypoxia characteristic of solid tumors. This study also verified that it is feasible to using DCE MRI to noninvasively and early monitor the effect of bortezomib on tumor blood perfusion, thereby indirectly reflect its antitumor biological effect. DCE MRI could be a potential clinical tool to select the patients responsive to bortezomib during early treatment and guide to establish the rational scheme combining bortezomib with other therapeutic methods such as routine radiotherapy or chemotherapy targets normoxic tumor cells. |
PDF Full Text Request