by Acad. Georgi GEORGIEV and Alexandr SOBOLEV, Dr. Sc. (Biol.), Institute of Gene Biology, Russian Academy of Sciences

The main problem of drug treatment of cancer using chemo-, radio, photodynamic therapy or other methods is in grave side effects from which the patients suffer. All types of cancer cell exposure unfortunately involve normal, intensely dividing cells with the same characteristics. The task of scientists is to increase the selectivity of drug action. This can be attained by drug delivery to transformed cells by means of multimodular carriers.

THERAPEUTIC EFFECT OF LIGHT

Therapeutic and diagnostic methods based on recent progress in photochemistry, photobiology, nuclear physics and chemistry have been ever more often used in oncology in these two last decades. One promising approach is offered by photodynamic therapy (PDT), when a photosensitizer drug accumulated in the tumor destroys it by absorbing the irradiating light. The photosensitizer proper is nontoxic for the malignant tumor, but the so-called active oxygen forms (AOF), appearing during its illumination in the presence of O₂, are destructive: these are singlet oxygen (¹O₂), hydroxyl radical (^•OH), and others; many molecules in the cell (proteins, lipids, nucleic acids) and supramolecular structures formed by them serve as targets for AOF. An intricate succession of photochemical and photobiological processes leads to irreversible damage to cancer tumor cells and/or vessels.

As it often happens, the idea of using light-activated chemical compounds for the treatment of various diseases is not new. Vitiligo (skin disease) was treated in ancient Egypt using plants containing psoralenes* and sunlight. Recently (in 1903) a Danish scientist Niels R.

* Psoralenes are natural photosensitizers contained in extracts from seeds of umbellate plants (Ammi majus L.) and legumes (Psoralea corylifolia L.) and their synthetic analogs. - Auth.

стр. 40

Finsen won one of the first Nobel Prizes for treatment of lupus vulgaris with concentrated light rays. This was again followed by a "pause," fortunately lasting for a few decades only, and not thousands of years.

PDT attracted active interest in the 1970s, when a mixture of hematoporphyrine* derivatives, synthesized by Thomas Doherty from Cancer Institute (Rosswell Park, USA), was effectively used for the treatment of skin cancer. This interest has been increasing ever since. Soon the QLT Photo Therapeutics Firm (Vancouver, Canada) patented photophrine - a drug consisting of a purified mixture of hematoporphyrine derivatives. However, despite the rather high efficiency of this new drug in the treatment of some cancers (esophageal, skin, vesical) and some other than oncological diseases, it has a significant drawback: like all PDT drugs created later, photophrine nonspecifically accumulates in tumor cells and is very slowly eliminated from the body, which results in high photosensitivity of the skin and ocular retina, and often leads to severe complications.

What are the causes of these side effects? AOF generated by the photosensitizer during its illumination usually "run" the distance of no more than 40 nm in 10 - 20-u. cells. The comparison of these values shows that the drug efficiency depends not only on its relative distribution between the tumor and adjacent tissues (cancerous and normal cells), but also on its location inside them.

Importantly, the sensitivity of all structural components to AOF effects is different; the most sensitive is the nucleus. However, modern data on intracellular distribution of photosensitizers indicate that they are present in all organelles, but not in the cell nuclei. So, in order to achieve the desirable cytotoxic effect, the drug dose has to be increased, which causes complications and side effects, the more severe, the higher the dose.

It is noteworthy that none of the photosensitizers is characterized by any specificity towards certain cells (cancerous or vascular tumor ones), and, accumulating in these cells, they inflict similar damage. This "avoidance" of the drug from the most sensitive object prompts a search for modifications of photosensitizers in order to impart new properties to them, stimulating their accumulation in the target cell nuclei.

HITTING THE TARGET

Conversions of a photosensitizer in the body are very important. Only its aggregated forms remain free, and they constitute no more than 1/10 of the dose and are rarely important for therapy, while nonaggregated forms bind various proteins and supramolecular complexes of blood. The future of these noncovalent compounds (their distribution in tissues, cells, and subcellular components) depends not on the photosensitizer, but on the proteins bound to it. In addition, in a complex with proteins the photosensitizer can generate AOF otherwise than if free.

Again: is it justified to rely on blood proteins alone for delivery of the photosensitizer to specific cells? Can the further fate of the drug be predetermined by fixing it (before administration) to the transporter with preset properties?

One more aspect. Some photosensitizers (for example, cationic) after their absorption by the cell and subsequent illumination lead to damage to mitochondria and cell death. Although these organelles are not the most sensitive ones, maybe we should stop here and not make the drug more complex by adding a special transporter to it? Unfortunately, two factors prevent it. First, the mitochondria are not the only and even not predominant site for localization of these photosensitizers, which again brings us to the problem of higher doses. In addition, the problem of drug delivery to the target cells, where blood proteins cannot transport them, is still there.

We believe that delivery of photosensitizers to the most sensitive compartments of target cells by special transporters will solve these problems. The expediency of creating such a mechanism for other antitumor drugs

Artificial multimodular nanoproteins (AMMNP) structure and stages of their delivery into the target cell nucleus.

* Hematoporphyrine is a purple pigment forming as a result of exposure of hematin, hemin, and hemoglobin to strong acids. - Auth.

стр. 41

Mice with 14-day-old malignant tumors (melanoma). Top: PDT with a common photosensitizer; bottom: with the same photosensitizer, but delivered by AMMNP. No tumor.

(radionuclides emitting alpha-particles, or alpha-emitters) is even more obvious: we have known during more than 60 years that the cell nucleus is the most sensitive target for this radiation. Alpha-emitters themselves, just like photosensitizers, are not specific for any cell type or organelles.

Now a few words should be said about these radionuclides and the interest in them which has emerged during recent years. A comparison with the beta-particle emitters, still the main "tool" for endoradiotherapy of cancer, shows that, first, alpha-radiation causes a greater number of irreparable DNA ruptures, second, more chromosome lesions, and, third, a longer delay of cell division than beta-radiation does. Exposure to alpha-particles leads to more serious irreparable injuries to cancer cells.

Microdosimetric estimates show that by cytotoxicity, one decay of ²¹¹ At alpha-emitter on the cell surface is equivalent to 1,000 decays of ⁹⁰Y beta-emitter. This, and new potentialities, which appeared due to progress in accelerator physics, radiochemistry, biotechnology, and other spheres have increased the interest of oncologists in alpha-emitters. The most promising technique is offered by endoradiotherapy of brain tumors (glioblastoma* treatment at the Medical Center of Duke University, Durham, North Carolina, USA) and acute myeloblasts leukemia** (Sloan Kettering Memorial Oncological Center, New York, USA). However, by now the best result has been to deliver alpha-emitters to target cancer cells (usually by means of antibodies) whose nuclei they have to destroy by emitting alpha-particles. Only few prove to be effective. Hence, it is desirable to create means for transportation of these radionuclides directly into target cell nuclei, which can guarantee the needed result.

GENE ENGINEERING HAS ITS SAY

This approach was realized at the Institute of Gene Biology of the Russian Academy of Sciences: artificial multimodular nanoproteins (AMMNP) were developed there for delivery of destructive agents (including photosensitizers and alpha-emitters) precisely to the target. AMMNP have to "recognize" the target cell among the many other, mainly normal cells, of which the body consists, penetrate this cell, and, finally, hit the cell nucleus.

Intermolecular "recognition" (at a subcellular level) means that the molecules selecting each other form a complex characterized by high affinity constants (values inverse to the equilibrium dissociation constants) of at least 10⁸ l/mol. Invasion into the cell can be achieved by different methods, but the most important thing is that it should come after the "recognition," otherwise the selective action will fail and AMMNP will get into the wrong cell.

This formulation of the problem prompted the idea of using a normal process taking place in both sound and cancerous cells-the receptor-mediated endocytosis. During this process not only low- and high-molecular-weight compounds, but also supramolecular complexes are brought into the cell, and this process is realized in a high selective mode because it is sustained by the high-specificity binding of the transporter substance (its common name is ligand) to an appropriate receptor.

Hence, by comparing the "receptor portraits" of normal cells and target cancerous cells we had to choose the ligand so that receptors to it be numerous on cancer cells but scanty or none on normal cells. This substance was found and became one of the AMMNP modules: the ligand one.

Starting its motion, however, AMMNP will have to follow the chosen track, being inside the endocytic vesicles and moving together with them. Thus it may get into lysosomes, where it will degrade under the

* Glioblastoma is a malignant tumor of the central nervous system. - Auth.

** Myeloblastic leukemia is a malignant disease of hemopoietic organs. - Auth.

стр. 42

effect of acid hydrolases before reaching the cell nucleus. This circumstance dictates the necessity of supplying a module for it, which will make it possible to "jump" from the endocytosis track and not get to lysosomes. Another, no less important reason for it, is that cell proteins, realizing the first steps of AMMNP delivery to the target (the cell nucleus), are located outside these vesicles.

We know of many macromolecules, which can make holes in biological membranes, but few of them are fit for the module in question. For example, pore-forming molecules, effective in neutral and slightly alkaline media (characteristic of the blood and tissue liquids) are no good, as they perform their function directly at the first contact with any cell, and the desired selectivity will be lost. That is why we selected the pore-forming molecules active at pH 5.5 - 6.0 (characteristic of those very endocytic bubbles-endosomes, preceding lysosomes) and inert in the neutral medium. They became an endosomolytic module.

Normal cytoplasmatic-nuclear transport can be used for drug delivery into the cell nucleus, as this process is intrinsic to both cancerous and normal cells. The AMMNP in this case should have a module with an amino acid sequence (nuclear localization signal), "recognized" by special importine proteins located in the structure-free part of the cytoplasm. Another carrier module is needed for attachment of the "transported" drug to AMMNP.

Thus, using gene engineering methods, we have designed polypeptides containing only those fragments of known proteins, which are capable of performing one of the preset four functions directly: attach the active agent (photosensitizer or alpha-emitter), bind the specific receptor on the surface of the target cancer cell, get into the cytoplasm and into the cell nucleus.

NEW QUALITY OF TREATMENT

Remarkably, none of the natural proteins or their complexes has a sum total of the functional characteristics realized by 4 modules in the small man-made AMMNP molecule: the ligand, endosomolytic, one containing a nuclear localization signal, and a destructive agent (antitumor drug) carrier.

As found out, the shift to the nanolevel was associated with modification of the particles' characteristics. The immunogenic characteristics that prevent therapeutic use were attenuated and the treatment efficiency increased appreciably due to the availability of all active groups in AMMNP and decrease in size. As a result, the antitumor agent acquired selective toxicity only towards preset type of cells, and due to penetration into their nuclei, the treatment efficiency increased 500 - 3,000 times.

The new technology made it possible to rapidly create "transportation" for drugs reaching cancerous cells of different tumors: melanoma, epidermal carcinoma, glioblastoma, squamous cell carcinoma of the head and neck, etc. It is enough to change the ligand module for hitting a different "target." The module can be translocated by rearranging the AMMNP for various purposes.

Experiments on animals show that these substances make antitumor drugs efficient in cases when they are of little effect or cannot be used at all, for example, in photodynamic therapy of pigmented tumors. Delivered by the photosensitizers directly into their cell nuclei, AMMNP can do with even small quanta of penetrating light.

We hope that the proposed module and combinatorial approach to the specific target delivery of photosensitizers, alpha-emitters, and other locally acting drugs is the first step towards a new generation of drugs.