(本文) Biological systems typically confront a formidable problem of DNA condensation: the necessity, for example, of fitting approximately 1.5 m of human diploid DNA into a 20-μm-diameter cell-and into a still smaller nucleus throughout most of the cell cycle. In eukaryotes, nucleosomes partially solve this problem; each nucleosome compacts the DNA associated with it 6- or 7-fold. Human DNA, however, must be compacted approximately 75,000-fold; this obviously involves more than nucleosome formation. DNA also exists in other compact structures such as viruses and prokaryotic nucleoids that lack nucleosomes. In the T2 or T4 bacteriophage, 60 μm of DNA must be compacted 600-fold to fit into a virion whose largest dimension is 0.10 μm; and in E. coli, 1.2 mm of DNA fits into a cell no more than 2 μm long.

The various biological phenomena in cell such as proliferation, transcription are thought to be performed in the specific structure of DNA condensed in nucleus. Human body is composed of enormous numbers of cells which are all originated from only one cell through cell division and differentiation. Although the process and mechanism of differentiation is not clearly understood, it is thought to be realized by the combination of specific kind of genes to be selected to function as a specific function of operation system in computer is realized by using only several partial programs not all in it, as supported by observation of chromosomal DNA in different kinds of cells. By what method the selection of specific kind of genes is performed is not obvious. The elucidation of the mechanism may be realized by the structural investigation of DNA condensed in cell. When a system in human body is damaged, cells in the system are proliferated and the function of the system is restored, which is thought to be realized by response of those cells to any signal from the damaged system, in more details, several specific proteins are related to the process and interact with any specific genes in the chromosome to make them function although the mechanisms of restoration are not yet fully clarified. Therefore the elucidation not only the structure of DNA condensed in nucleus of living cell but also its relationship to their biological functions is important for elucidation of biological mechanism of human body. And it is also important for developing a drug for a treatment human of various diseases including a genetic disorder. In these respects, thermal analysis leading to structural analysis of DNA may be important for the clarification of biological process and therapy of human diseases together with electronic scattering, atomic force scattering or x-ray scattering.

In the respect of gene therapy, also, thermal analysis can be useful. A reason that gene vector is required for gene therapy is to protect the gene for therapy from various nucleases in human body. Presumably DNA is protected by gene vector to block the access of those nucleases to DNA. Although DNA can be protected by gene vector quite effectively, it yet may be not sufficient for practical use. To elucidate the complex structure, that is, the relative binding position of gene vector to DNA, can be lead to clarify the mechanism of protection of DNA from nucleases. If it is clarified, it leads to develop a more effectively protecting gene vector from nuclease.

To clarify the structure of the condensed DNA many theoretical and experimental studies have been done. Thus as a force inducing the condensation to elongated DNA, several kind of forces have been suggested, including fluctuation of counterions, charge ordering (or charge reorganization), release of low molecular counterions, release of water molecules, delocalization of counterion. But they are not still controversial. If the thermodynamic parameters can be obtained using calorimetry, their theoretical condensation mechanisms may be evaluated. However, formation of precipitates followed by DNA condensation makes difficult the evaluation of DNA condensation. to make the evaluation possible a new condensing agent suppressing the condensed DNA not to form precipitates is required.

Our group has synthesized condensing agents, for example, poly(ethylene glycol)-block-polylysine (PEG-PLL), able to suppress formation of precipitation to condensed DNA which was developed by connect covalently an end of a linear hydrophilic poly(ethylene glycol) (PEG) to an end of a cationic polylysine (PLL). Considering the fact that DNA condensed by PEG-PLL, it is indicated that PEG chain surrounds the condensed DNA by cationic PLL segment, that is, a micelle structure. Using these block copolymers, DNA condensation without being interfering by forming precipitates can be investigated.

Actually these condensing agents were developed for efficient gene vector for gene therapy. For a gene vector to be useful for gene therapy, first of all, the problem of formation of precipitates has to be overcome. Although, to overcome the problem, many researches have been performed by many researchers also, our result is one of the most successful developments for gene therapy. These block copolymers are supported by several experimental results as a promising gene vector.

Here was investigated the effect of the secondary condensation on the ITC curve by comparing PLL and PEG-PLL binding to DNA using ITC. And also the transmittance was measured under the same condition with the ITC experiment. Two distinctive endothermic binding stage of PLL and PEG-PLL to DNA were observed, one attributed to the binding of them without DNA condensation, the other that with DNA condensation. Their ITC curves were not so much different with each other at low salt concentration (10 mM), but they showed great difference between them with increase in salt concentration. That is, the heat accompanied by the binding of PLL to DNA decreases to zero at the lower molar ratio of lysine to nucleotide, while in case of PEG-PLL the such molar ratio did not decrease so much as PLL. Because this result seems to mean that the amount of PLL bound to DNA decrease remarkably with increase in salt concentration, the amount of free PLL in the solution was investigated by titrating by the poly(ethylene glycol)- poly(aspartic acid) (PEG-P(Asp)) the free PLL in the supernatant obtained after centrifugation. Based on that result, it is cleared that the amount of PLL bound to DNA did not decrease even if high salt concentration. These results indicate that the difference between PLL and PEG-PLL binding to DNA is attributed to the interference of the heat accompanied by the secondary condensation to the heat accompanied by binding of PLL to DNA. The dependence of the molar ratio at which the endothermic heat disappears on salt concentration is also consistent with the result of the transmittance.

The binding of a variety of poly(ethylene glycol)-block-polylysine (PEG-PLL) to DNA in various salt concentration was investigated thermodynamically using isothermal titration calorimetry (ITC). Thermodynamic parameters of two distinctive endothermic binding processes were obtained by curve fitting of each titration data; the first parameters attributed to the binding process of PEG-PLL to DNA without DNA condensation, and the second to the process with DNA condensation. The titration result exhibited that PEG-PLL having high degree of polymerization (DP) of PLL is more effective to form stable complex. Based on the thermodynamic parameters, it became clear that the binding of PEG-PLL to DNA is accompanied not only by the large increase in entropy, which is thought to be attributed to the release of low molecular counterions or water molecules bound to in the vicinity of DNA and PEG-PLL, but also by the large decrease of free energy which is thought to be attributed to the electrostatic interaction of PEG-PLL with DNA.

And also we discussed about the structure of condensed DNA induced by PEG-PLL based on the thermodynamic parameter obtained here. These thermodynamic observations may lead to the elucidation of mechanism of PEG-PLL protecting DNA from the enzymes such as endonuclease in serum for efficient gene delivery or the mechanism of reproduction and transcriptional regulation.

Actually for thermodynamic parameters to be obtained from the ITC curve, it should be fitted to any arbitrary bind model. However the conventional fitting models may not be suitable for the polycation binding to DNA followed by conformational change in DNA. Thus originally new fitting model was suggested. Data shown here was obtained using the model.

PLL binding to DNA was accompanied not only by increase in entropy but also decrease in free energy, based on the thermodynamic parameters. Increase in entropy may be attributed to release of low molecular counterion, release of water or delocalization of counterions. Decrease in free energy may be attributed to the electrostatic interaction between DNA phosphates and PLL. Similarly with the case of the PLL binding to DNA, DNA condensation was indicated that it was accompanied not only by release of counterions or release of water or delocalization of counterions but also by electrostatic interaction between them, considering the thermodynamic parameters which shows not only large increase in entropy but also large decrease in free energy.

In future these data and the method of thermal analysis may be effective not only for evaluating the structure of genomic DNA condensed in living cell and its relation on the function but also for developing more stable gene vector against nuclease by clarifying the mechanism to protect DNA from nuclease.

　本論文は、正電荷を持つ物質の結合によって誘起されるDNA凝縮を熱分析の対象とし、カチオン性物質のDNAへの結合力やDNA凝縮の程度などに関する熱力学変数に基づく評価が、遺伝子治療における非ウイルス性遺伝子ベクターの設計・開発指針を与えることを示している。以下、章ごとにその内容を解説し、本論文の審査結果を述べる。

　第1章では、序論としてDNA凝縮に関する先行研究を研究手法や凝縮剤の種類、DNA凝縮機構などの主題別にまとめ、DNA凝縮と凝縮構造の研究の重要性が述べられている。ここでは、DNA凝縮が一種の相転移現象であるため、熱測定が重要な研究手法であるにも関わらず、凝縮DNA同士の二次凝集や沈殿形成の効果を単一DNA分子レベルで生じるDNA凝縮と区別して検証する実験的方法論が未確立であったため、凝縮転移の詳細は未解明であったこと、熱分析による特性解析も十分な検討がなされていなかったことが指摘されている。また、遺伝子キャリアとして開発されたブロック共重合体を凝縮剤として用いることにより、二次凝集の生じない実験系が得られ、DNA凝縮機構の解明や遺伝子キャリアとDNAとの結合様式の研究が可能になることとともに、本論文の主題である熱分析によるDNA凝縮機構解明の意義が述べられている。

　第2章では、本研究の主たる実験的方法論として位置づけた熱分析手法についての解説がなされている。ここでは様々な熱分析測定法の中で、特に本研究で用いた等温滴定法の測定原理や優位性が述べられている。この方法の利点として、研究対象とする系の化学量論比やエンタルピー、エントロピー、自由エネルギーなどの熱力学変数が得られるだけでなく、凝縮剤のDNAへの結合に伴って生じるDNA凝縮転移を追跡出来る点が挙げられている。また、熱力学変数を得るために用いられる解析手法、すなわち等温滴定曲線を解析する3つの従来法を紹介すると共に、それらがDNA凝縮の系に適用できないことを指摘し、新しいフィッティング法を開発する必要があることが述べられている。

　第3章では、DNA凝縮過程を考慮した新しいフィッティング法の開発について述べられている。具体的には、DNAに対する凝縮剤の二種類の結合過程、すなわちDNA凝縮を伴わない結合過程とDNA凝縮を伴う結合過程を仮定し、必ず後者が前者の後に起こることをフィッティング法に反映させた。二つの結合過程はどちらも第2章で述べられたフィッティング法の一つであるSingle Set of Identical Sites Model(1種類の結合サイトモデル)を用いており、この基本モデルを組み合わせることによって新しいフィッティング法が成り立つことが明らかにされている。

　第4章では、凝縮DNA同士の二次凝集の効果を熱的に評価することを試みた結果が述べられている。二次凝集が抑えられない凝縮剤であるポリリシン(PLL)ホモポリマーと二次凝集を抑える凝縮剤であるポリエチレングリコール(PEG)とポリリシン(PLL)から成るブロック共重合体(PEG-PLL)をそれぞれDNAに添加しながら熱測定を行った。その結果、両者の差は明らかであり、この差を二次凝集の生成に起因するものと仮定すれば二次凝集は発熱過程であること、また二次凝集を生成する系が単一DNA分子の凝縮に対して誤った解析結果をもたらすことを指摘している。

　第5章では、ブロック共重合体(PEG-PLL)のプラスミドDNAへの結合に伴う熱測定を行い、PLLの重合度や溶液の塩濃度が結合に与える影響が調べられている。得られた等温滴定曲線に対して第3章で開発した新しいフィッティング法を適用し熱力学変数が求められている。全体的な傾向として、重合度の減少と塩濃度の増加に伴いPEG-PLLのDNAへの結合力が減少することが示されている。また、結合に伴うエンタルピー変化は小さく、自由エネルギーとエントロピーの変化は同程度で大きいことから、自由エネルギー減少に伴って本来系の外に出るはずの熱が系の内部でのエントロピー増加に用いられることが明らかにされている。さらに、純粋なDNAの構造変化に対応する熱力学変数を、DNA凝縮を伴う結合過程における熱力学変数から凝縮を伴わない初期の結合過程における変数の差によって表されることを示し、この結果からもエントロピー効果に関して同様の結論を得ている。

　第6章では、総括としてPEG-PLLを用いることにより二次凝集による熱の影響を取り除き、DNA凝縮に関する熱力学変数を得ることが出来、その結果に基づきPEG-PLLのDNAへの結合力、DNA凝縮程度に対するPLL重合度や塩濃度の影響を議論している。さらには、この熱力学変数をエネルギー計算に還元することにより、結合構造や結合状態などの予測や解析に援用できることを指摘している。

　上記の通り、本論文では構造変化を伴うDNAと凝縮剤との結合における熱測定のための試料溶液の調製法および熱力学変数を得るための熱分析手法が確立されている。またDNA凝縮過程と凝縮の程度に関する熱分析による解析結果は、既存の遺伝子ベクターの最適化やより優れた新規遺伝子ベクターの材料設計・開発に対して新規な指針を与えるものとなっている。

　よって本論文は博士(工学)の学位請求論文として合格と認められる。