Application of genetically modified mouse (GEM) model in oncology research

Author: Dr. Yu Xiaofeng, senior vice president of tournament business model animal, Senior Scientist

cutting edge

The genetically modified mouse (GEM) model is a de novo tumor established under natural intact immune conditions relative to the tumor cell vaccination and immunodeficient mouse models commonly used in the past. Therefore, as a tool for oncology research, the GEM model can mimic the histopathological and molecular characteristics of human tumors, showing better genetic heterogeneity, which has the advantage of reflecting the tumor cells themselves and the tumor microenvironment. Interacting factors such as cells, including the ability to cause the primary tumor to begin to develop into a metastatic disease. The establishment and application of the GEM model has greatly promoted the development and progress of oncology research. At present, the GEM model has been successfully applied to verify potential tumor genes and drug targets, assess treatment effects, analyze the effects of tumor microenvironment, and evaluate drug resistance mechanisms. Moreover, combining clinical patient research and constructing a more reasonable and effective GEM model to optimize preclinical trials of tumor intervention will undoubtedly further promote the development of new strategies for tumor therapy and improve the success rate of its effective conversion into clinical application.

The main challenges facing current oncology research

Oncology research and tumor treatment still face many challenges in clinical practice. Among them, the formation of anti-tumor drug tolerance and tumor metastasis diseases are two important practical problems currently facing. Anti-tumor resistance is caused by the emergence of primary mutations in heterogeneous tumors, or by the massive growth of pre-treatment resistant clones, while existing monotherapy or chemical drug therapies targeting anti-tumor agents are available. Drug tolerance cannot be avoided. Moreover, after a clearly successful treatment, a small amount of drug-tolerant tumor cells can survive and become a major cell population after a certain period of time, eventually forming a recurrent disease that is different from the original tumor phenotype. Tumor metastatic disease is the cause of more than 90% of cancer-related deaths, because there are currently no effective treatments for these secondary tumors. In recent years, although some encouraging progress has been made in tumor immunotherapy for intervention in the immune system of cancer patients, the therapy has only been effective in some specific cases, and it has no practical clinical significance for most tumor patients.

The so-called successful cancer treatment often requires synergistic effects of various methods, such as surgery, radiation exposure, cytotoxic therapy, and comprehensive strategies such as immunotherapy. In order to design effective and reasonable comprehensive treatment measures and programs, the primary basis and premise is to understand the formation and development of tumors, metastasis, and the mechanism of interaction between tumor cells themselves and their microenvironment cells in response to treatment, so as to find The most effective treatment for different tumor types. To achieve this, researchers must rely on preclinical studies of animal models. Although in the past it relied on traditional preclinical mouse models (ie, through the establishment of transplanted human tumor cell lines or homologous mouse tumor cell line models) to obtain successful validation of new preclinical anticancer therapies, the vast majority of these new The therapy ended in failure in the clinical phase III trial.

In general, the traditional in vivo tumor mouse model does not perform well in predicting the effectiveness of new clinical therapies, thus highlighting the significance and value of finding improved preclinical models with better predictive power and effects. Recent advances and developments in genetic modification technology have made it possible to rapidly develop a GEM model that can more effectively simulate human tumors. The GEM model has a genetic composition, interaction between tumor cells and its tumor microenvironment, drug response and tolerance. Closer to the tumor patient. The emergence of a new generation of GEM models has also greatly promoted the transformation of new anti-tumor strategies into clinical applications, ultimately achieving the goal of improving the survival rate of cancer patients.

Advantages and disadvantages of traditional mouse models commonly used in oncology research

The mouse tumor transplantation model, which was first established in nude mice 50 years ago to transplant human/mouse tumor cell lines, has become a common mouse model for tumor research. This type of transplantation model can quickly test potential tumor and metastasis-related genes and become clinical. The main tool for pre-drug testing. For example, xenograft studies help to reveal the mechanism of colon cancer (CRC) tolerance to drugs (such as Vemurafenib) in mice, enabling the initiation of CRC patients in clinical trials against both mutant BRAF (eg V600E) and EGFR. The targeted combination therapy demonstrates the practical significance of this type of xenograft model in establishing a new combination therapy strategy.

Xenograft studies are also helpful in identifying specific gene expression profiles and studying the characteristics of organ-specific localization and metastasis. The application of this model confirms that the spread of breast cancer cells is present in the vicinity of blood vessels, which provides a possible countermeasure for regulating these spread breast cancer cells. Moreover, in vivo model studies using tumor cell line transplantation have provided many basic concepts for anti-tumor immunity, T cell tolerance mechanisms, and tumor immune escape pathways. These findings have laid the foundation for the ongoing breakthrough in tumor immunotherapy.

However, since tumor cell lines contain many mutations at the beginning and additional mutations are generated during long-term in vitro culture, such vaccination models are difficult to truly reflect the morphological and genetic heterogeneity of human tumor cells, thereby The reliability of this type of model as a predictive effect of clinical application is reduced. In addition, in order to prevent possible rejection, the xenograft model of tumor cell lines is based on immunodeficient mice, which also limits its application in the fields of immune system and therapeutic response in tumor development.

Unlike cell line transplantation models, patient-derived tumor xenografts (PDTX, or PDX) models constructed by transplanting fresh human tumor biopsies into immunodeficient mice have more advantages, PDX mouse models. The tumor retains features of tumor tissue molecules, genetics, and histological heterogeneity from tumor patients (even after passage through several generations in mice). Therefore, the PDX model has also become a useful tool for personalized medicine and preclinical drug screening.

At present, the research of PDX model has been applied to potential clinical drug prediction experiments on a large scale. Some researchers have constructed about 1000 PDX models from different types of mutations and applied these different PDX models to screen different drugs in mice to find the correlation between drugs and tumor genotypes to achieve experimental repeatability and Unification of clinical interpretability.

Unfortunately, the dissatisfaction of the PDX model in studying certain types of tumors, such as estrogen receptor-positive breast cancer and prostate cancer, has become a major obstacle to its application. Moreover, the PDX model must be built on immunodeficient mice that lack the natural anti-tumor and tumor-promoting activities mediated by the acquired immune system.

The researchers also understand that although the PDX model lacks a functional acquired immune system, the model can provide valuable clinical data. A corresponding improvement in progress is to reconstruct a humanized mouse model of the human immune system by transplanting human CD34-positive hematopoietic stem cells or precursor cells with significant success. Although the reconstruction of human immune cells from certain specific lineages in mice is still challenging, it can promote the development and maturation of human bone marrow cells in mice by introducing human-related cytokines, chemokines, and growth factors. Effect.

To construct immunodeficient mice optimized to support human HLA-restricted T cell development, a humanized mouse model was established using genetic modification techniques to introduce human HLA molecules into the corresponding regions of knockout mouse MHC types I and II. The humanized mouse model can be used as a useful tool for preclinical evaluation of immunotherapy. However, when the source of the human hematopoietic cell donor (such as by cord blood or fetal liver) is limited, the higher construction cost in the actual operation also naturally becomes The disadvantages of the practical application of the model.

Application of continuous improvement of GEM model to study primary tumor

After the successful establishment of mouse pronuclear injection transgenic technology in the 1970s, in the early 1980s, the method was used to introduce the cloned oncogene into the mouse genome for the first time, and the so-called oncomice was successfully prepared. The oncogenic mouse is the first tumor GEM model that specifically expresses the oncogene v-HRas using a mammary gland specific promoter (MMTV), and the oncogenic mouse was successfully constructed for the first time, confirming that the mouse forms a primary breast tumor, and This has greatly inspired the field of cancer research, because the results of this study for the first time truly prove the hypothesis that oncogenes can express tumors in normal cells. In 1992, along with the breakthrough development of mouse embryonic stem cell (ES) gene targeting technology, the successful construction of tumor suppressor gene (TSG) knockout mice also proved the important role of this kind of TSG gene in tumorigenesis.

Although carcinogenic mice and TSG knockout mice provide a very valuable theoretical basis, these two models also have their limitations. Since the mouse obtained by transgenic technology is to express the transferred foreign gene to all cells in a specific tissue, the TSG knockout mouse is a related gene in all cells in the body. However, the formation process of a real tumor is a diffuse tumor phenomenon caused by the accumulation of a certain genetic variation in a single cell under the premise of the health of the entire tissue and organ in the individual. In order to meet the actual conditions of the tumor formation process, it is more necessary to design and construct a more reasonable or complex mouse model, such as the conditional inactivation of tumor suppressor genes in somatic cells, or the activation (mutation) of oncogenes. GEM model. The basic principle of the construction of the conditional GEM model is to add the loxP recombination site to the two ends of the modified gene. In the presence of a specific Cre recombinase, the DNA between the ends of the loxP can be knocked out under certain conditions. The purpose of inactivating the gene. The first successful example of the application of this model is a mouse colorectal cancer model constructed using the Cre-loxP system-mediated somatic cell inactivation of the Apc gene. Adenoviral vector is used to express Cre recombinase specifically in intestinal epithelial cells, and tissue-specific knockout of APC gene causes rapid formation of scattered colorectal adenoma in mice, which is characterized by familial colon adenomatous polyposis (FAP) patients. There are many similarities. Therefore, by discovering mutants of specific cancer genes, researchers can construct mouse models that are more capable of mimicking the similar characteristics of tumor patients in histology, molecular science, and clinically.

With the Cre-ERT fusion protein induction system, researchers can modify the relevant target genes in somatic cells at specific times with specific tissues, merging the mutant hormone binding region of the estrogen receptor with Cre recombinase to establish an inducible The regulated Cre recombinase expression system, when in the presence of an estrogen analog (such as Tamoxifen), causes Cre recombinase activity to be activated after translation, exerting its role in recognizing the loxP site, and achieving the purpose of inducible cleavage of the target gene. . Therefore, the conditional genetic modification of LoxP mice (ie, loxP sites on both ends of the target gene DNA), by adding the inducer Tamoxifen at the selected time, controls the specific expression of Cre-ERT to achieve the target gene. Time and space and the purpose of specific modifications on the area.

Although the Cre-loxP system can be used for the modification of more than one gene, since this process occurs simultaneously, it is difficult to completely simulate the multi-step formation process of the tumor, and the mutation is a feature of the gradual accumulation formation process. Recently, researchers have used a dual-enzyme system that can function independently (such as Flp-FRT/Cre-loxP, or Cre-loxP/Dre-rox) to establish a modification of the target gene expression. The practical significance of the successful application of this technical method is as follows: 1. Independent research on spontaneous and non-spontaneous pathways and processes of tumor cells; 2. Simulating human multi-step cancer formation process, with continuous induction mutation; 3. Unique genetic evaluation of tumor treatment targets.

Improvement of Tumor GEM Model Construction Strategy and Technology

The GEM model has proven to be an effective tool for oncology research because of its unique advantages, but researchers have been working hard to continuously improve and improve it. Due to the long period of the construction and development process, the workload is high, and the cost is high, especially for the modification of multiple allelic genetic loci, this mouse model is constructed to construct a new mutant mouse model with genetic characteristics. It is more time consuming and requires a longer mating breeding process. This has also become a major factor limiting the widespread practical application of genetically capable GEM models. In recent years, due to the widespread popularization and application of tumor genome sequence research technology and the rapid growth of newly discovered cancer-related gene mutations, it is more necessary to establish a rapid and novel mouse model development strategy to achieve rapid verification of potential oncogenes in vivo. And the purpose of establishing a non-reproductive genetically modified GEM model of known patient-related mutations. At present, the progress in this area mainly includes: ES cell-based tumor model; application of CRISPR/Cas9 technology genome programming; and improved tumor model of tumor patient related sites.

1. ES cell-based tumor model

In order to further accelerate the development of a novel human tumor GEM model, it has become a research strategy to directly perform oncology research on mouse embryonic stem cells (ESC) after genetic modification as non-genetically modified (eg, chimeric) mice. The recently reported GEM-ESC strategy is to establish an ES cell-based tumor GEM model. This kind of model is based on the original genetic modification, and rapidly constructs a novel genetically modified mouse model. For example, using the GEM-ESC strategy, the MET protooncogene was directly introduced into the ESC of the mouse breast tumor model of K14cre-Brca1-Trp53 (KBIP) to construct a novel breast cancer model of KBIP-MET. The results of the study showed that compared with KBIP mice, the breast cancer formed by this KBIP-MET mouse has more transformation characteristics, which is more likely to form carcinosarcoma. In response to tumor drugs, breast cancer in the KBIP mouse model is sensitive to clinical RARP inhibitors (such as Olaparib), while transformed breast tumors in the KBIP-MET mouse model show tolerance to the inhibitor.

2. Genomic programming using CRISPR/Cas9 technology

In the past decade, with the rapid development of new genome programming techniques (such as ZFNs and TALENs), the CRISPR/Cas9 genome programming system appeared in 2013. This kind of genome editing technology has become the development of PCR technology in the past few years. Revolutionary and most influential technology updates. The CRISPR/Cas9 system was first discovered in prokaryotes against the immune system established against foreign invasive genetic material and was quickly used for gene editing of various species. With a single guide RNA (sgRNAs), Cas9 nucleases can specifically target any gene locus in the genome for gene knockout. By applying Cas9-induced DNA fragmentation and single-stranded nucleotide/donor DNA, the system can also be genetically modified for specific gene mutations, or for specific insertion of loxP/FRT recombination sites.

The CRISPR/Cas9 technology system demonstrates the ability to efficiently edit gene-targeting strategies at different sites in the genome, making it the best choice for rapid development of tumor mouse models. All genetic mutations currently observed in human tumor patients can be rapidly constructed by genetic modification methods, including conditional gene knockouts, point mutations, translocations, and the like. In addition, researchers have used the CRISPR/Cas9 technology to perform somatic (non-reproductive hereditary) editing of mouse oncogenes and TSGs. Because of the efforts and success of this research strategy, the system has become a hepatocellular tumor, lung cancer, A new approach to brain cancer, pancreatic cancer, and non-genetically modified models of breast cancer.

Recently, the CRISPR/Cas9 system has also been applied to genetic modification of target gene suppression (CRISPRi) or activation (CRISPRa). Such a modification system can be used to develop a corresponding oncogene, and/or to inhibit the induction and reversible activation of a TSGs gene in a mouse model. For example, with the CRISPRa-based system, the purpose of studying its carcinogenic potential is achieved by activating the transcription of oncogenes.

Although the CRISPR/Cas9-based gene editing system has great potential, the system has certain defects in gene editing in vivo. For example, the current system strategy is not suitable for verifying the carcinogenic potential of potential oncogenes. In addition, the gene editing method of introducing Cas9 into somatic cells can cause a specific immune response of Cas9, resulting in the possibility that Cas9-expressing cells are cleared. In order to avoid these potential risks, it is possible to carry out corresponding experiments in immunodeficient mice, or to obtain a mouse model with immunological tolerance to Cas9 by genetic modification, and then carry out corresponding animal experiments. Finally, although it has been reported that the use of inducible Cas9n nickases that cause DNA single-strand breaks can reduce off-target effects, researchers must have a clear understanding of the application in order to completely avoid mediated by CRISPR/Cas9. Off-target mutations required for non-design are difficult.

3. Improved tumor model of tumor patient related sites

It is necessary and practical to construct an ideal mutation model for tumor patients, to study the role of target genes in tumorigenesis, and to effectively evaluate drug effects. Because in human tumor suppressor genes (TSGs), many germ cell-dependent germ cell mutations and somatic mutations are missense or nonsense mutations, leading to the formation of mutant products or possibly functional truncated proteins. Such mutations are difficult to achieve by conditional knockout mouse models, because the conditional genetic modification strategy is to completely knock out one or several exons in the target gene to achieve the function of inactivating the target gene. Some studies have shown that a mouse mutation model constructed with reference to a TSG mutation associated with a tumor patient can produce a different phenotype than a complete knockout of the target gene. For example, compared with Trp53 knockout mice, patient-associated Trp53 hot-spot mutant mice showed more pronounced carcinogenic activity.

Similarly, a conditional mouse model of the BRca1 gene mutation associated with BRCA1 breast cancer patients showed that breast tumors caused by mutations in specific RING regions of the Brac1 gene were more likely to be compared to the Brac1 complete knockout mouse model. Those drugs that destroy DNA are tolerant because their BRCA1 protein contains less RING activity. Studies have also shown that Brac1 protein contains less RING activity due to mutations, which appears to be more resistant to DNA-damaging drugs, and the results help to reveal the causal relationship between these mutations and the therapeutic response.

Application range of GEM model in oncology research

As a GEM model of primary tumorigenesis, it can be a systemic choice for in vivo analysis of interactions between cells themselves and cells during tumor formation, development, and tumor formation during metastasis. The human tumor GEM model has also been successfully applied to verify candidate drug targets, evaluate treatment effects, and evaluate drug tolerance mechanisms. Since the GEM model forms primary tumors in mice with intact immune systems, this model is more suitable for exploratory research in potential tumor immunotherapy. To establish a strategy and method for the close correlation between genetically modified mouse models and human diseases, and to provide a meaningful application platform for exploring and developing new methods and strategies for tumor therapy. It also provides information on clinical treatment effects and other information for the design and development of new anti-tumor treatments. The GEM model plays an important role in the research progress and contribution of tumor biology and transformational oncology in the following aspects.

1. Verify potential oncogenes

On the basis of a large number of tumor sample sequencing studies to obtain increasing potential tumor genes, it is necessary and practical to establish a strategy for rapid verification of these potential tumor-associated genes in vivo. Given the speed and relative simplification factors, GEM-ESC and CRISPR/Cas9 technologies are the preferred method for quickly validating potential tumor genes. In particular, the use of somatic-based CRIPPR/Cas9-mediated gene editing technology to establish a non-genetically modified mouse model to achieve high-throughput in vivo verification of potential tumor genes. For example, a combination of DNA injection and in vivo electroporation is used to introduce 15 gRNAs/Cas9 expression plasmid mixtures of 13 different major tumor suppressor-related genes in pancreatic ductal adenocarcinoma (PDAC) into mature mice. In the pancreas, a mouse model in which these 13 genes were simultaneously modified was constructed. The results showed that more than 60% of these target genes in this PDAC mouse showed gene knockout, suggesting that CRISPR/Cas9-mediated mutations induced tumor formation. Similarly, the GEM model that induces Cas9 expression using Dox was also used to validate a variety of known intestinal tumor genes (such as Apc and Trp53). In addition to modifying TSGs, CRISPR/Cas9 technology is also used to verify the carcinogenicity of chromosomal rearrangements, such as the fusion of the Eml4-Alk gene observed in lung cancer patients.

In addition, the use of the GEM model to validate candidate potential oncogenes from clinical patients and screened has also become a common strategy for studying tumor-related gene function. For example, Professor Hu Zhuowei's group recently studied the role of pseudokinase Trib3 in promoting the formation of acute promyelocytic leukemia (APL) by constructing conditional overexpression and knockout GEM models. The results showed that it was also in mouse bone marrow cells. Specific expression or knockdown of Trib3 and oncogenic protein PML-RARa (PR) fusion gene, Trib3 gene can significantly increase the role of PR-induced APL formation. Professor Lu Yi's group was the first to confirm the specific overexpression of Sry gene male/female mice in the liver tissue by constructing the GEM model of the Y chromosome sex determining region (Sry) gene-specific chemical carcinogen (DEN) Inducing the formation of hepatocellular carcinoma (HCC) in mice is more sensitive, suggesting that the Sry gene plays an important role in the formation of HCC.

2. Study the dependence of oncogenes

Oncogene dependence refers to the fact that certain tumor formation is completely dependent on a single consensus oncogene. Since the modification of genes by the conditional GEM model is irreversible, it is not suitable for studying oncogene dependence. Therefore, it is necessary to select different regulatory induction strategies for corresponding research, such as the fusion of oncogenes and ERT to control their expression. It has been reported that the Trp53-ERT variant replaces the homozygous Trp53-established homozygous knock-in mouse, and the Trp53-ERT mouse induces Trp53 expression only in the presence of Tamoxifen, and has formed a tumor mouse model. On the basis of the study, the effect of re-purification of p53 function on existing tumors was studied. The results showed that on the basis of lymphoma caused by Eu-Myc, the recovery of Trp53 activity can produce rapid apoptosis and significantly increase the survival rate of mice. In addition, the reversible induction system of Doxycycline (Dox) regulating gene expression was also applied to the establishment of the GEM model, and the expression of human MYC proto-oncogene was induced by this system to cause tumor formation. After shutting down the expression of the MYC gene, the corresponding reaction of the formed tumor after the proto-oncogene was inactivated was observed. This study is a Tet-off induction system using Dox, which continuously expresses the human MYC transgene in mouse hematopoietic stem cells, induces the formation of malignant T-cell lymphoma and acute myeloid leukemia in mice, on the basis of which, if induced by the addition of Dox After stopping the expression of MYC, the tumor phenotype was also found to be weakened, and it was confirmed that this process was associated with tumor cell cycle death. The study also found that different tumor types are different for the long-term effects of stopping the activation of MYC expression in this reversible induction system. For example, transient inhibition of MYC expression in osteosarcoma, as sarcoma cells differentiate into mature bone cells, will continue to shrink sarcoma. On the contrary, although the inhibition of MYC expression causes diffuse atrophy of liver cancer, the remaining tumor cells are still in a latent state, and after re-opening MYC expression, their tumor characteristics can be quickly restored.

3. Crack the mechanism of spontaneous transfer formation

Despite the ever-improving strategy of choice for cancer treatment, metastatic disease remains the leading cause of cancer deaths. The metastatic process is a complex multi-step process formed by the mutual interaction of tumor cells and the tumor microenvironment. The vast majority of preclinical metastasis studies in the past were performed using cell line inoculation models, and such models do not truly reflect the metastatic process in tumor patients. The GEM model can cause primary tumor development and metastasis formation, and is therefore an indispensable tool for studying the process of spontaneous metastasis formation of tumors that were unknown in the past. Due to the excessive growth of the primary tumor, the mice generally have to be sacrificed before the formation of a large range of metastases, which is a potential deficiency of the GEM model. This limitation can be solved by orthotopically transplanting GEM-derived tumor tissue, such as by surgical transplantation, to achieve the effect of intratumoral heterogeneity of the donor tumor, so that the metastasis process is close to the clinical common metastasis. disease.

Some important findings have been obtained by using the GEM model to study the tumor metastasis process. Past studies have suggested that tumor metastasis occurs in the late stages of tumor formation. However, studies by the BALB-NeuT and MMTV-PyMT mouse breast tumor models have shown that early damage transfected cells have the ability to spread to bone marrow and lung tissue to form micrometastases. In addition, epithelial-to-mesenchymal cell metastasis (EMT) is thought to play a very important role in tumor cell transmission and metastasis. However, studies using pancreatic cancer and breast cancer GEM models have shown that tumor cells not only retain their epithelial cell characteristics, but also appear in metastatic lesions, suggesting that EMT is not necessary for tumor metastasis formation in these models. Furthermore, in exploring the complex relationship between tumor cells and the immune system during the process of tumor metastasis formation, the GEM model has clearly played a prominent role. For example, bone marrow immune cells, such as macrophages and neutrophils, play a crucial role in promoting the metastasis of different types of cancer. Recent studies have shown that breast tumors cause systemic inflammation, that is, the expansion of IL-17-derived T cells and secondary immunosuppressive neutrophils, which can trigger the spontaneous metastasis of the GEM model of lobular breast cancer, leading to GEM transplantation. Model of spontaneous metastasis of disease.

The GEM model also plays an important role in revealing the involvement of related genes in inhibiting tumor metastasis. Recently, Liu Baohua's group applied Tet-ON to induce the expression of Sirt7 in GEM, revealing the mechanism of Sirt7 inhibiting the metastasis of primary pancreatic cancer. The results confirmed that Sirt7 induced by Dox significantly inhibited MMTV-PyMT mouse breast tumors. The role of lung metastasis, and its mechanism of action is achieved by regulating the TGF-β signaling pathway.

Therefore, the GEM model plays an indispensable role in revealing the complexity of tumor metastasis and challenging the generally accepted theory that tumor metastasis is a metastatic process including advanced cancer cells including EMT. These important findings may provide an important reference for the treatment of patients with metastatic cancer.

4. Study the role of tumor microenvironment

The GEM model has played an irreplaceable role in revealing the role of tumor cell external factors such as cancer-associated fibroblasts (CAFs) and immune cells such as cancer-associated fibroblasts (CAFs) and immune cells and tumor formation processes. CAFs regulate ECM and basement membrane formation by synthesizing extracellular matrix (ECM) components such as collagen, fibronectin, and laminin. Moreover, CAFs are a source of various soluble mediators, including matrix metalloproteinases (MMPs), which play an important role in promoting ECM transformation and enhancing their homeostasis in ECM. GEM model studies have shown that CAFs have a dual role in tumor formation. Using a K4-HPV6 squamous skin cancer mouse model, it was found that CAFs can stimulate tumor development by enhancing inflammation, angiogenesis, and ECM reconstitution during epithelial cell malignant transformation.

In contrast, studies by two independent pancreatic cancer GEM models have shown that inhibition of CAFs in vivo has an effect of accelerating the process of tumor formation, suggesting a preventive effect of CAFs on tumors. Such contradictory phenomena are understandable for immune cells. Initially, immune cells are thought to be cells that can transform tumors by attack, inhibiting the process of tumor formation. However, recent studies have shown that these immune cells also have a function of promoting tumors. Studies of several different tumor types using mouse models have revealed a correlation between inflammation and tumors. For example, a mouse model of colitis-associated cancer, which specifically knocks out the NF-jB signaling system in myeloid immune cells, slows tumor growth in mice, indicating that it has a tumor-promoting effect.

In addition, the K4-HPV6 mouse model study also showed that mast cells and bone marrow-derived cells play a role in promoting squamous skin cancer formation by activating MMP9 and re-adjusting the matrix structure. Applying the same skin cancer model, it is found that chronic inflammation has the effect of promoting neonatal tumor formation. So far, research has begun on the promotion of inflammation-induced tumor-associated macrophages and neutrophils. For example, based on the MMTV-PyMT breast cancer mouse model, knocking out an important macrophage-associated gene CSF1 (Colony-stimulating factor 1) revealed that the malignant process of breast tumors in this mouse was delayed. Similarly, inhibition of CXCR2, a chemokine that mediates neutrophil migration, has the effect of inhibiting intestinal tumor formation in APC mice. In summary, these studies emphasize the role of immune cells in the process of tumorigenesis and development.

5. Determine the source of tumor cells

Revealing the source of cells during tumorigenesis will provide a very important theoretical basis for the development and improvement of therapeutic strategies. The application of the GEM model has successfully elucidated the cellular origin of certain different tumor types. In a small cell lung cancer (SCLC) study, Trp53 and Rb1 genes were expressed in Clara cells, neuroendocrine (NE) cells, and type II alveoli (SPC) by intratracheal injection of cell-specific Adeno-Cre viral vectors. Cells were specifically knocked out and analyzed for different times of tumorigenesis and tumor phenotype. The results indicate that NE cells are the main source of cells responsible for SCLC formation relative to SPC cells. In addition, cell-derived studies can provide unexpected results that are different from previous studies. For example, in the past BRCA1-based breast cancer research, the source cells of this type of cancer are basal epithelial stem cells. In the application of GEM model to BRCA1-induced basal-like breast cancer research, it is found that luminal progenitor cells are the true source of basal-like tumors.

Recent studies from two different experiments have shown that genetic variation (such as the Pik3ca mutation) can significantly affect stem cell composition. Pik3ca point mutations (such as H047R) cause loss of the ability of mammary epithelial cells with lineage characteristics to differentiate into pluripotent stem-like states. Moreover, the cell source of Pik3ca breast tumors dominates the degree of malignancy, indicating that it is of great practical significance to accurately find the source of tumor cells in terms of improving the specificity of anticancer drugs and therapeutic effects.

6. Verify new drug targets

Considering that not all oncogenes are necessary to maintain tumor formation, verify inactivated TSG or reduce oncogenes before the human clinical trials, ie, in preclinical animals, against the corresponding target drug. A test that can cause atrophy of an ongoing tumor becomes very important. The application of inducible mouse model can be used to verify the correlation of oncogene maintenance of tumors. For example, in the mouse model of breast cancer, the expression of oncogene Pik3ca is induced to induce atrophy of some tumors, suggesting that these tumors are "dependent". Sexually active P13K signals are closely related. However, most tumors eventually relapse due to an increase in Met or Myc, suggesting that these genetic lesions may induce tolerance to P13K inhibitors. This example demonstrates that the inducible GEM model can be applied to preclinical studies to achieve not only the goal of validating drug targets, but also to reveal the mechanism of drug tolerance formation.

TSG is also likely to be an effective drug target. Loss of p53 function in tumors is caused by a dominant negative or inhibitory mutation of the p53 gene, as well as an increase/overexpression of its specific inhibitors MDM2 and MDM4. Genetic studies using a GEM model that reversibly inhibits p53 activity have shown that restoring p53 gene function can rapidly resolve established tumors, suggesting the development of inhibitory MDM2 molecules, thereby restoring p53 function, or restoring mutant p53 to wild The clinical significance of anti-tumor drugs with type-function p53. Similarly, the use of a GEM model that induces knockdown of APC to study colorectal cancer also suggests that induction of APC function can rapidly cause rapid and extensive tumor cell differentiation and sustained recurrence-free atrophy, which is an in vivo evaluation of APC/Wnt. The pathway serves as a therapeutic target for colorectal cancer caused by APC mutations.

7. Clarify therapeutic effects and tolerance

In order to minimize the risk of failure of new anti-tumor therapies in clinical trials, it is more important to establish a pre-clinical and predictive in vivo model to objectively assess the corresponding drug effects and tolerability. A study of the GEM model of lung cancer and pancreatic cancer caused by the Kras mutation found that the response effect of the GEM model on targeted therapy and conventional chemotherapy is very similar to that of the corresponding patient.需要加以关注的是小鼠与人在药物代谢方面表现明显不同，比如，参与肝脏药物代谢的细胞色素P450酶底物特异性方面，不同的物种存在较大的差异。该类问题可借助人源化小鼠模型加以解决。因此，建立人源化的GEM模型作为临床前药物效果的研究，将有助于优化针对靶特异抗肿瘤药物的研发，以及寻找与确定治疗效应的关键因素，并使其成为肿瘤病人特征的预测性生物标记。另外，GEM模型也可应用于探索治疗敏感性肿瘤获得性耐药的形成机制。

在探讨肿瘤治疗效应与耐受机制方面，K14cre; Brca1-f/f; Trp53-f/f (KB1P) 小鼠是作为BRCA1突变乳腺癌临床前GEM模型的一个非常有说服力的例子。KB1P小鼠可形成完全模拟类人BRCA1突变乳腺癌组织病理学特征的乳腺肿瘤，而且，对含铂类药物和PARP抑制剂也具有高敏感性。临床试验证实，PARP抑制剂Olaparib可以治疗卵巢肿癌，乳腺癌和结直肠癌。 虽然该药物并不对所有这类癌症病人的治疗都有效果，但其对BRCA1突变携带者表现有明显治疗效果，可能与PARP抑制剂合用引起的协同致死作用与BRCA1缺乏有关。BRCA1突变细胞对PARP抑制剂表现为更加容易被损伤，因为PARP抑制剂诱发的单一链DNA断裂，可导致DNA复制时的双链断裂，造成BRCA1缺失细胞无法实施同源重组机制来修复损伤的DNA。

根据Olaparib在临床试验中获得的理想效果，FDA于2014年12月批准了该药用于BRCA1/2突变卵巢癌病人的治疗。尽管病人对该药物有很好的反应效果，然而，在病人和GEM模型研究中都发现了获得性药物耐受性。通过临床前KB1P小鼠模型研究证实，这类耐药机制与药物运输物及同源重组恢复等数量的增加有关。这些研究结果有助于了解临床上耐药性的产生, 以及设计针对Olaparib耐药病人的改进治疗策略。

关于肿瘤治疗效应与药物耐受性的关系，现在也是越来越清楚，即其影响不仅受肿瘤细胞自身因素，而且与成纤维细胞和免疫细胞等基质因素有关。通过PDAC的GEM模型进行肿瘤干预研究的结果表明，治疗是抑制旁泌性相关信号通路，减少促结缔组织增生肿瘤基质，增加肿瘤脉管系统，导致促进抗肿瘤药物导入肿瘤部位的过程。然而，关于攻击PDAC中的肿瘤基质的概念，最近有两个研究对该理论进行了挑战。这两研究结果表明，基质因素可能通过阻止肿瘤血管形成，引起不是促进而是抑制PDAC生长。所以，这些研究都揭示了肿瘤微环境在治疗耐受性方面，发挥了比过去想象的更加重要的，且复杂的作用。

8. 肿瘤免疫治疗

在过去十年，通过对免疫反应的深刻揭示与认识，建立了利用病人免疫系统攻击肿瘤的治疗策略。近年来的黑色素瘤和肺癌临床病人试验已证实，包括抗CTLA-4和抗PD-1的免疫检查点抑制物，在增强病人有效抗肿瘤免疫及改善生存率等方面具有巨大潜力。这些临床试验基础是来自过去几十年在实验小鼠模型上开展的基础研究，从而揭示了CTLA-4和PD-1在阻止免疫反应中的重要性，特别是通过对CTLA-4和PD-1敲除小鼠出现严重与温和程度的自发性自身免疫表型的明确证实，应用CTLA-4抑制物引起接种肿瘤小鼠的抗肿瘤T细胞反应增加，产生肿瘤排斥作用，提示释放T细胞上的刹车可能是抵抗肿瘤的一个潜力的应对策略。尽管如此，同时也应该清楚的了解，仍有一定比例的病人对此类免疫治疗没有反应，且目前的挑战是还不知道其真正原因。

目前，虽然绝大部分的免疫学研究都是在肿瘤移植小鼠模型的基础上进行的，但就现在研究情况预测表明，今后GEM模型应用于该领域的研究将会越来越多。应用GEM模型研究的部分结果表明，在新形成的肿瘤过程中，因肿瘤引起的耐受机制，T细胞丧失其对肿瘤细胞的反应性，特别明显的是，如果将来自GEM模型小鼠形成的肿瘤细胞，接种至免疫缺陷小鼠体内，肿瘤会快速生长，而野生小鼠则能排斥这些肿瘤细胞。提示这些肿瘤细胞没有失去其免疫原性，T细胞仍然能识别这些细胞并发挥其攻击作用。但在原发肿瘤小鼠的体内这些T细胞却无能为力。

肿瘤常常被认为是慢性炎症的结果，这种炎症可引起局部和系统免疫抑制，从而不利于T细胞发挥其有效的功能。而且，肿瘤常表现为树状突细胞（DC)功能缺失，导致T细胞启动缺损。例如，在MMTV-PyMT乳腺肿瘤小鼠模型中发现，原本具有潜在激活抗肿瘤T细胞的DC细胞，会被大量存在的巨噬细胞竞争抑制，这些巨噬细胞起到了阻止特定T细胞激活的作用。最近的研究也证实，促进DC细胞功能或阻止骨髓细胞引起的免疫抑制，可以达到改善免疫检查点抑制物的抗肿瘤效果。因此，对肿瘤病人实施耐受T细胞免疫激活疗法的时候，结合应用针对改善免疫抑制或则增强T细胞启动的靶点药物，在临床上可能会起到更好的治疗效果。

相对于肿瘤接种模型，应用GEM模型进行免疫治疗研究需要不同的方法。考虑到GEM模型中肿瘤是在每个独立小鼠体内发生的，如同病人肿瘤的发生过程，具有特殊的肿瘤抗原。因而，具有其异质性的特点，从而确保区别鉴定反应和非反应肿瘤之间的分子不同，也有助于临床预测的生物标记物的建立。然而，对于大多数GEM模型来源的肿瘤来说，辨认可被T细胞识别的表达肿瘤抗原却是未知的。为了克服这点，可借助基因修饰的方式，将临床上相关肿瘤抗原引入小鼠体内，使其具有诱发肿瘤特异性T细胞反应的效果。比如，应用低免疫原性的肿瘤（如肉瘤和肺癌），将肿瘤特异性抗原引入至GEM模型后，这些肿瘤的免疫原性增加，引起潜在而短时间的抗肿瘤T细胞反应。起初的抗肿瘤T细胞反应很快发生，然后是可调控T细胞介导的免疫抑制。因此，这些模型将有助于目前与未来相关研究，以达到揭示免疫逃逸的复杂机制，最终研发具有改善的肿瘤免疫治疗的新策略的目的。

9. 与临床试验并行的 GEM 模型

最近推出的“与临床共试验” 范例，目的是将临床前GEM实验与人体临床试验同时开展，从而达到预测治疗效果的作用。该策略已经应用于前列腺癌治疗中，并成功揭示了由雄性激素诱发的耐受性与某些遗传关键因子有关，以及克服去势难治性的新综合疗法。同样，应用NSCLC的GEM模型的一起临床试验表明，Kras/Lkb1突变肺癌较Kras or Kras/p53突变肿瘤对临床上的Docetaxel和MEK抑制剂Selumetinib的联合疗法，表现为更加具有耐受性。揭示了LKB1是临床试验中对此类药物联合疗法耐受的潜在决定因素。这类研究表明，应用GEM模型作为人肿瘤临床前药物效果研究，能发现新的生物标记物和联合疗法。

GEM在肿瘤学研究中发展趋势与未来展望

许多抗肿瘤药物在临床试验中未能达到临床前实验的期望目的，已经成为目前肿瘤学与转化癌症医学所面临的巨大挑战，如何改善肿瘤学领域的临床前研究结果的预测性，也是人们十分关注的热点。因此，如何选择更能真实反映人肿瘤疾病发生发展过程的临床前肿瘤动物模型，将显得更加重要了。为了实现这一目的，首先需要考虑的是如何建立能真实反映肿瘤本身及外在特征的临床前小鼠模型。比如，临床前模型应该含有病人特异性突变，该种突变具有诱发恶性肿瘤的趋势，且在一定病人群中显示遗传变化特征。另外，原发的肿瘤过程是在自然微环境中进行的，如同在人肿瘤进行过程中所见的肿瘤细胞与肿瘤微环境之间相互作用（包括免疫细胞，成纤维细胞，以及淋巴细胞和血管的潜入）。再有，由于绝大多数进入临床试验的病人已经是处于广泛肿瘤转移疾病过程中，因此，实验设计时就应该考虑选择能模拟病人疾病不同进程状态的小鼠模型，开展相应的临床前的药物效果评估研究。在现实的临床试验中，往往进入临床试验的病人之前多是经过不同程度的治疗，因而极有可能会干扰治疗效果的验证。而临床前的动物研究则是建立在从来没有接受任何治疗的基础上进行的，从而导致高估治疗效果的结果。从另外一方面讲，临床试验中对那些接受过治疗的严重病人没有效果，却仍有可能对没有接受过治疗的严重病人有益处。

最近基因修饰模式动物技术的进展，促进了快速研制更加精准的能引起原发肿瘤的小鼠模型，该类模型结合了特异性肿瘤病人形成发生发展过程中肿瘤细胞本身和细胞外的特征。预测这些新一代的GEM模型和基于GEM模型的移植模型，将是真实模拟病人肿瘤发生发展过程，研究自发性转移疾病的最佳模式动物。这些模型可作为研究肿瘤形成的复杂过程（包括肿瘤的起始，器官特异性转移的形成，肿瘤微环境的参与等方面）的重要而有用工具。但是，对于肿瘤病人而言，更加重要的是，这些模型能更为深入地揭示免疫治疗中的反应性与耐受性，以及疾病的复发等相关机制。期待未来，应用新一代GEM模型对抗肿瘤新药物临床前的评估研究，将会增加预测其在临床试验中的成功率，从而加速抗肿瘤新药策略设计与临床实施，达到改善防治肿瘤病人的病情的最终目的。

About the Author

俞晓峰博士，国际知名模式动物和细胞生物学专家，先后就任于耶鲁大学医学院、iTL基因打靶公司和纽约大学医学院以及美国ASC生物技术公司等机构，在遗传修饰模式动物领域有超过20年的研发和管理经验。目前任职于赛业生物科技，任高级副总裁和高级科学家，主要负责基因修饰模式动物平台的技术工作，其研究成果多次发表在Nat Immunol 、 Mol Cell Biol等高水平杂志上。

主要参考文献：

1. Li K, Wang F，Cao WB, and ZW et al (2017) TRIB3 Promotes APL Progression

through Stabilization of the Oncoprotein PML-RARa and Inhibition of p53-Mediated Senescence. Cancer Cell 31, 697–710

2. Tang X，Shi L，and Liu B et al (2017) SIRT7 antagonizes TGF-β signaling and inhibits breast cancer metastasis. Nature Communications 8: 318

3. Liu C, Ren YF, Yi Lv and Xu-Feng Zhan et al (2017) Activation of SRY Accounts for Male-Specific Hepatocarcinogenesis: Implication in Gender Disparity of Hepatocellular Carcinoma. Cancer Letters 410: 20-31

4. Chunga WJ, Daemena A, Melissa R, and Junttila MR et al (2017) Kras mutant genetically engineered mouse models of human cancers are genomically heterogeneous. PNAS Dec 4, E10947–E10955

5. Annunziato S, Kas SM, Nethe M, and Drenth AP et al (2016) Modeling invasive lobular breast carcinoma by CRISPR/Cas9-mediated somatic genome editing of the mammary gland. Genes Dev 30: 1470 – 1480

6. Drost R, Dhillon KK, and Schut E et al (2016) BRCA1185delAG tumors may acquire therapy resistance through expression of RING-less BRCA1. J Clin Invest 126: 2903 – 2918

7. Maresch R, Mueller S, and Barenboim M et al (2016) Multiplexed pancreatic genome engineering and cancer induction by transfection-based CRISPR/ Cas9 delivery in mice. Nat Commun 7: 10770

8. Chiou SH, Winters IP, Wang J, and Chuang CH et al (2015) Pancreatic cancer model-ing using retrograde viral vector delivery and in vivo CRISPR/Cas9-mediated somatic genome editing. Genes Dev 29: 1576 – 1585

9. Weber J, Öllinger R, and Engleitner T et al (2015) CRISPR/Cas9 somatic multiplex-mutagenesis for high-throughput functional cancer genomics in mice. Proc Natl Acad Sci USA 112: 13982 – 13987

10. Dow LE, O'Rourke KP, and Lowe SW et al (2015b) Apc restoration promotes cellu-lar differentiation and reestablishes Crypt homeostasis in colorectal cancer. Cell 161:1539 – 1552

11. Henneman L, van Miltenburg MH, and Schlicker A et al (2015) Selective resistance to the PARP inhibitor olaparib in a mouse model for BRCA1-deficient metaplastic breast cancer. Proc Natl Acad Sci USA 112: 8409 – 8414

12. Clohessy JG, Pandolfi PP (2015) Mouse hospital and co-clinical trial project- from bench to bedside. Nat Rev Clin Oncol 12: 491 – 498

13. Cong L, Ran FA, Cox D, and Marraffini LA et al (2013) Multiplex genome engineering using CRISPR/Cas systems. Science 339: 819 – 823

14. Jamieson T, Clarke M, and Nibbs RJB et al (2012) Inhibition of CXCR2 profoundly suppresses inflammation-driven and spontaneous tumorigenesis. J Clin Invest 122: 3127 – 3144

15. Drake AC, Chen Q, Chen J (2012) Engineering humanized mice for improved hematopoietic reconstitution. Cell Mol Immunol 9: 215 – 224

16. Liu P, Cheng H, and Fox EA et al (2011) Oncogenic PIK3CA-driven mammary tu-mors frequently recur via PI3K pathway-dependent and PI3K pathway- independent mechanisms. Nat Med 17: 1116 – 1120

17. Martins CP, Brown-Swigart L, and Evan GI (2006) Modeling the therapeutic efficacy of p53 restoration in tumors. Cell 127: 1323 – 1334

18. Shibata H, Toyama K, Shioya H, and Toyoshima K et al (1997) Rapid colorectal adenoma formation initiated by conditional targeting of the Apc gene. Science 278: 120 – 123

Mattress

We are Hospital Electric From China Manufacturer Medical Bed, also known as Hospital Bed Home Care, Hospital Equipment, Nursing Bed, etc., is a hospital bed used by patients when they are hospitalized in the hospital.