通常的癌症疫苗是通过促进肿瘤抗原（tumor antigen）的释放，并且吸引抗原提成细胞（antigen presenting cells, APCs）将释放的肿瘤抗原提呈给T细胞使其活化，发挥效应来杀灭癌细胞。但是，这个策略用在恶性黑色素瘤（melanoma）的治疗上，就面临一个问题——活化的T细胞在消灭肿瘤细胞的同时也误伤正常的黑色素细胞（melanocyte），因为许多与恶性黑色素瘤相关的抗原是正常黑色素细胞在产生黑色素过程中的蛋白。
在最新的《Nature Biotechnology》advance on-line publication上，Mayo Clinic的科学家却用了一种新的癌症疫苗策略来达到治疗目的：利用刻意杀死健康的皮肤细胞来启动免疫系统，最终消灭肿瘤。因为热休克蛋白70（heat shock protein 70, hsp70）是免疫耐受（immunological tolerance）的重要开关，可以阻断免疫耐受而使免疫系统对自身抗原（self-antigen）产生反应。通过皮内注射质粒DNA疫苗，表达作用于转录的细胞毒性基因和hsp70基因，研究者设想细胞毒性基因可以引起正常细胞的死亡，而hsp70就可以促发自体的免疫反应，而这个免疫反应进而消灭肿瘤细胞。
通过激活免疫系统使癌症疫苗的共同途径，但是不管用什么方式，都回面临相同的风险。如果被激活的免疫系统“矫枉过正”，开始大量进攻正常的组织和细胞，就会出现自体免疫病的副作用。这种新方法也不例外，研究者也有同样的担心。但是实验结果却出人意料，大量被激活的T细胞杀死了恶性黑色素瘤；不过，在它们引起自身免疫病之前，由于调节性T细胞（regulatory T cell）的出现，而被抑制了。
多么令人鼓舞啊：不但在多次治疗后，所有的实验小鼠的皮肤癌都被根除了；而且又不存在自身免疫病的问题。
接下去，研究会进一步向两个方面展开。一方面当然是研究人体对这种免疫治疗的反应，如果顺利，则开始进行临床实验。另一方面，研究者希望将这种策略扩展到其他的肿瘤类型，例如乳腺癌、肺癌和前列腺癌。
在通常情况下，对正常细胞的破坏是治疗中需要避免的。而这种新的治疗策略却反其道而行之，利用杀伤正常细胞触发免疫效应，使癌细胞也“同归于尽”。
A simple method to cure established tumors by inflammatory killing of normal cells
We describe a simple technology used to cure an established metastatic disease. Intradermal injection of plasmid DNA encoding a transcriptionally targeted cytotoxic gene, along with hsp70, not only promoted tissue-specific, inflammatory killing of normal melanocytes, but also induced a CD8+ T-cell–dependent, antigen-specific response in mice that eradicated systemically established B16 tumors. This CD8+ T cell response was subsequently suppressed in vivo within a few days. The data demonstrate that deliberate destruction of normal tissue can be exploited to generate immunity against a malignant disease originating from that tissue. This approach obviates the need to identify tumor antigens and does not require complex isolation of tumor cells or their derivatives. In addition, it provides a model system for studying the mechanisms underlying the etiology and control of autoimmune diseases. Finally, despite targeting normal tissue, therapy could be separated from development of overt autoimmune symptoms, suggesting that the strategy may be valuable against tumors derived from both non-essential and essential tissue types.
Figure 1. Intradermal injection of Tyr-HSVtk plasmid DNA leads to expression in melanocytes.
(a) Plasmids used in these studies. The Tyr-HSVtk plasmid uses a hybrid promoter of three tandem copies of a 200–base pair (bp) element of the murine tyrosinase enhancer47 upstream of a 270-bp fragment of the tyrosinase promoter28 to drive expression of the HSVtk gene27. In Tyr-LacZ, the HSVtk cDNA was replaced with the LacZ gene28. In CMV-hsp70, the murine hsp70 gene11 is driven by the CMV promoter in pCR3.1 (Invitrogen) (full details of plasmids available on request). (b) At 5 d after plasmid injections (days 1, 2, 3) and GCV or PBS treatment (days 1–5), skin samples at the site of intradermal plasmid injections (as indicated) from two separate mice per treatment were used to prepare cDNA, which was subsequently analyzed by PCR using primers specific for either murine tyrosinase or GAPDH as a loading control. (c) At 5 d after administration of nothing (none) or Tyr-HSVtk plasmid (10 g) (days 1, 2, 3) along with either PBS (Tyr-HSVtk + PBS) or GCV treatment (Tyr-HSVtk + GCV) on days 1–5, skin at the site of intradermal injection was analyzed for the L-DOPA oxidase activity of the tyrosinase enzyme in the samples. Results from two separate experiments (filled or hatched columns) are shown. Error bars represent standard deviations.
Figure 2. Immune consequences of Tyr-HSVtk + CMV-hsp70 intradermal injections.
(a–e) Sections of skin at sites of three daily intradermal injections of the following. (a) PBS; a displaced hair follicle at the needle injection site is shown, and white arrows indicate examples of pigmented hair bulbs, indicating the presence of intact melanocytes. (b) (10 g Tyr-HSVtk + 10 g empty plasmid) and GCV i.p. (c) (10 g CMV-hsp70 + 10 g empty plasmid) + GCV. (d,e) (10 g Tyr-HSVtk + 10 g CMV-hsp70) + GCV. Sections for a–e were recovered 72 h after the first injection, stained with H&E and examined for evidence of immune infiltrates. Regions encompassing dense infiltrates are delineated by white arrows. (f) Skin samples at the site of intradermal plasmid injections (72 h later, as indicated) from two separate mice per plasmid combination were snap frozen and used to prepare cDNA, which was subsequently analyzed by PCR using primers specific for murine cytokines including IFN- (f). PCR for GAPDH was used as a reference for levels of expression of each RNA.
