Development of antibacterial agents and materials

目的1：探讨MAX1水凝胶的抗菌作用机制。迄今为止，我们的数据表明抗菌作用的机制涉及膜破坏，导致细胞与凝胶表面接触后细胞死亡。使用激光扫描共聚焦显微镜进行的活死分析表明，凝胶的表面具有杀菌作用，当细菌接触到表面时，它们会被迅速杀死。此外，我们在监测乳糖渗透酶缺陷大肠杆菌ML-35细胞质中β -半乳糖苷酶释放的实验中发现，凝胶表面会导致内外膜破坏。我们还发现，虽然可溶性的、非凝胶化的MAX1在凝胶状态下高浓度表现出抗菌活性，但可溶肽很少，是凝胶表面发挥活性的。目的2：确定氨基酸组成和序列如何影响凝胶抗菌活性。当自组装时，MAX1的赖氨酸侧链从其原纤维的溶剂暴露区域显示出来。我们的机制假设表明，这些侧链首先与细菌细胞表面接触，并可能解释凝胶的大部分活性。回顾AMP文献发现，虽然在许多AMP中发现赖氨酸，但也大量发现精氨酸（Arg）和色氨酸（Trp）。在一些amp中，当赖氨酸被精氨酸取代时，可能是由于精氨酸的胍基侧链能够与细菌膜中的脂质头基形成更强的相互作用，从而产生更强的肽。我们系统地用Arg替换了MAX1中所有的Lys残基，以研究Arg含量对抗菌活性的影响。我们发现含有精氨酸形式凝胶的肽在杀死革兰氏阳性和革兰氏阴性耐药菌株方面都非常有效。SAR研究表明，Arg含量的增加不仅与抗菌活性的增加有关，而且与溶血潜能的增加有关。然而，通过平衡精氨酸与赖氨酸含量的比例，我们发现可以制备有效的，但有选择性的凝胶。序列见附录。重要的是，这些研究表明凝胶的抗菌活性可以通过肽设计来调节。目标3：开发用于治疗递送的细胞穿透肽。我们正在开发一个CPPs家族，能够根据其改变的膜和细胞表面组成优先向癌细胞输送药物。这个目标是从我们最近的工作中发展而来的，我们设计了一种小的裂解肽（SVS-1），能够优先杀死癌细胞，如下所述。目的1：探讨多糖的内聚性、黏附性和蛋白质释放特性。由于葡聚糖的生物相容性和可利用性，我们最初正在探索葡聚糖作为多糖成分。在我们最初的研究中，我们制备凝胶时改变了三个参数，即葡聚糖分子量、醛含量和醛/胺（CHO/NH2）网络比，以测量它们对凝胶的内聚性和粘附性以及凝胶形成速度的影响。通过评估时间、频率和应变扫描实验中的存储模量，从流变学角度测量了材料的内聚性和形成速率。用猪皮进行拉伸动态力学分析，测定凝胶粘接强度。葡聚糖分子量在15-70 kD之间变化，25-40 kD的葡聚糖提供了最高的储存模量（G’）和粘附强度。我们还优化了合成方法，使双醛含量在25- 50%之间变化，发现虽然G′和粘接强度都随着双醛含量的增加而增加，但对粘接性能的影响最大。最后，以绿色荧光蛋白（GFP）作为模型蛋白交联剂，我们发现随着CHO/NH2比的降低，存储模量、粘接强度和凝胶速率都增加。这一数据表明，具有更多可用于交联的溶剂性胺的配方将产生更强、更快的凝固凝胶。这可以通过增加特定蛋白质的wt%或使用具有更多赖氨酸残基的蛋白质来实现。总之，我们可以改变这三个参数中的每一个来制备储存模量在102-105 Pa之间，粘附强度在1-6 kPa之间的生物胶粘剂凝胶，这与纤维蛋白胶（一种临床胶粘剂）的水平相当。我们已经确定，凝胶的模量为105，黏附强度为4千帕或更大，当引入体内时，可以很好地粘附在组织上，提供明确的形状。此外，从注射器输送后，材料形成的速度可以从几秒调整到几分钟。凝胶形成太快是有问题的，会堵塞注射器，但10-20秒的凝胶是最佳的。我们用这些机械特性作为评价新材料的基准。我们刚刚开始研究这些凝胶的释放特性，使用绿色荧光蛋白作为模型蛋白进行批量释放研究和体内实验。在这些早期实验中，只研究了一种凝胶组成（40 kD葡聚糖，25%氧化，CHO/NH2 = 8）。批量释放实验表明，蛋白质以爆发性释放，随后是缓慢的持续释放，蛋白质保持折叠和功能性荧光。当50微升gfp凝胶在裸鼠的侧翼内原位形成时，生物荧光测量表明蛋白质在14d内释放，其释放曲线与体外测量的相似。我们最初的实验表明，蛋白质可以直接用于形成生物黏附凝胶，以供自身输送。据我们所知，这些凝胶将是一流的生物黏附蛋白递送载体。目的2：确定可以递送的蛋白质范围。为此，我们研究了使用这种技术可以传递的蛋白质的范围。我们还研究了制备由两种蛋白质成分组成的复合凝胶的可能性，其中用于交联的大部分蛋白质是惰性的，廉价的填充蛋白，次要成分是生物活性蛋白。这允许在配方中使用较低浓度的高活性蛋白质。最后，我们提出了一种独特的材料类型，其中我们用多胺聚合物完全取代蛋白质成分来构建抗菌，可注射的伤口填充物。到目前为止，我们只研究了三种模型蛋白，白介素-2 （IL-2）、绿色荧光蛋白和肌红蛋白，以快速评估该技术的可行性。这些蛋白在分子量（15-30 kD）和pi（6-7）上略有不同，但在折叠（α和β -丰富）和可接近的赖氨酸残基数量（分别为11、16和20）上有所不同。我们表明，生物黏附凝胶可以使用这些蛋白质中的任何一种形成，并且溶剂可接近赖氨酸的数量会影响凝胶的黏附性和黏附性能。我们还通过CD和功能分析显示，释放的蛋白质保持折叠和功能。虽然数据很有希望，但所有这些蛋白质都是单体的，结构稳定。我们建议用更复杂的蛋白质来挑战我们的递送系统，同时探索可能影响材料机械和释放特性的蛋白质属性。

Aim 1: Investigate the mechanism of antibacterial action of the MAX1 hydrogel. To date, our data suggest a mechanism of antibacterial action that involves membrane disruption that leads to cell death upon cellular contact with the gel surface. Live-dead assays employing laser scanning confocal microscopy show that the gel's surface is bactericidal and that bacteria are quickly killed when they engage the surface. In addition, we showed that the gel surface causes inner and outer membrane disruption in experiments that monitor the release of beta-galactosidase from the cytoplasm of lactose permease-deficient E. coli ML-35. We have also shown that although soluble, non-gelled MAX1 shows antibacterial activity at high concentration in the gel state, there is little soluble peptide available and it is the gel's surface that exerts activity. Aim 2: Define how amino acid composition and sequence influences gel antibacterial activity. When self-assembled, the lysine side chains of MAX1 are displayed from the solvent exposed regions of its fibrils. Our mechanistic hypothesis suggests that these side chains are first to engage the bacterial cell surface and may account for much of the gel's activity. Review of the AMP literature indicates that although lysine is found in many AMPs, arginine (Arg) and tryptophan (Trp) are also found to a large extent. In some AMPs, when lysine is substituted with arginine, a more potent peptide is generated presumably due to the guanido side chain of Arg being able to form stronger interactions with lipid head groups contained in the bacteria's membrane. We have systematically replaced all of the Lys residues in MAX1 with Arg to study how Arg-content influences antibacterial activity. We found that peptides containing Arg form gels that are extremely effective at killing both gram-positive and gram-negative drug-resistant strains of bacteria. SAR studies show that an increase in Arg content correlates not only with an increase in antibacterial activity, but also an increase in hemolytic potential. However, by balancing the ratio of Arg relative to Lys content, we showed that potent, but selective, gels could be prepared. See addendum for sequences. Importantly, these studies show that the gel's antibacterial activity can be modulated by peptide design. Aim 3: Develop cell penetrating peptides for therapeutic delivery. We are developing a family of CPPs capable of preferentially delivering drugs to cancer cells based on their altered membrane and cell surface composition. This Aim evolved from our recent work where we designed a small lytic peptide (SVS-1) capable of preferentially killing cancer cells as described below. Design of Novel Bioadhesive Gels for the Local Delivery of Proteins Aim 1: Explore polysaccharide composition on cohesive, adhesive, and protein-release properties. We are initially exploring dextran as the polysaccharide component due to its biocompatibility and availability. In our initial studies, gels were prepared varying three parameters, namely, dextran molecular weight, aldehyde content, and the aldehyde/amine (CHO/NH2) network ratio to measure their effects on the cohesive and adhesive properties of resulting gels as well as the rate of gel formation. Cohesiveness and the rate of material formation were measured rheologically by assessing the storage moduli in time-, frequency-, and strain-sweep experiments. Gel adhesive strength was measured using tensile dynamic mechanical analysis employing porcine skin. Dextran molecular weight was varied from 15-70 kD, with 25-40 kD dextran providing gels with the highest storage moduli (G') and adhesive strength. We also optimized synthetic methods to vary the dialdehyde content from 25-50 % and found that although both G' and the adhesive strength increases with dialdehyde content, it is the adhesive properties that are most influenced. Lastly, using green fluorescent protein (GFP) as a model protein crosslinker, we showed that the storage modulus, adhesive strength, and rate of gelation all increase as the CHO/NH2 ratio decreases. This data indicates that formulations having more solvent accessible amines available for crosslinking will result in stronger, faster setting gels. This can be accomplished by increasing the wt% of a particular protein or using a protein having more Lys residues. In sum, we can vary each of these three parameters to prepare bioadhesive gels that range in storage moduli from 102-105 Pa and adhesive strengths from 1-6 kPa, which is on the order of fibrin glue, a clinical adhesive. We have determined that gels having moduli on the order of 105 and adhesive strengths of 4 kPa or greater adhere nicely to tissue affording well-defined shapes when introduced in vivo. In addition, the rate of material formation can be tuned from seconds to minutes after delivery from syringe. Gels that form too quickly are problematic, clogging the syringe, but gels that set in the regime of 10-20s are optimal. We use these mechanical characteristics as benchmarks when evaluating new materials. We have just begun to investigate the release properties of these gels using GFP as a model protein in bulk release studies and in vivo experiments. In these early experiments, only one gel composition was investigated (40 kD dextran, 25% oxidation, CHO/NH2 = 8). Bulk release experiments indicate that protein is released with a burst followed by a slow sustained release profile, with the protein remaining folded and functionally fluorescent. When 50 microliter of GFP-gel is formed in situ within the flanks of nude mice, biofluorescence measurements indicate that protein is released over a period 14d, with a similar release profile as that measured ex vivo. Our initial experiments suggest that proteins can be used directly to form bioadhesive gels for their own delivery. To our knowledge, these gels will be first-in-class as bioadhesive protein delivery vehicles. Aim 2: Determine the scope of proteins that can be delivered. In this aim, we investigate the scope of proteins that can be delivered using this technology. We also investigate the possibility of preparing composite gels comprised of two protein components where the majority of protein used for crosslinking is an inert, inexpensive filler protein and the minor component is a bioactive protein. This allows a lower concentration of highly active protein to be used in the formulation. Lastly, we propose a distinct material type in which we replace the protein component altogether, with polyamine polymers to construct antimicrobial, injectable wound fillers. To date, we have studied only three model proteins to quickly assess the feasibility of the technology, interleukin-2 (IL-2), GFP, and myoglobin. These proteins vary slightly in molecular weight (15-30 kD) and have similar pIs (6-7), but differ in their folds (alpha and beta-rich) and number of accessible Lys residues (11, 16, and 20 respectively). We showed that bioadhesive gels could be formed using any of these proteins and that that the number of solvent accessible lysines influences gel cohesive and adhesive properties. We also showed via CD and functional assays that released proteins remain folded and functional. Although the data is promising, all of these proteins are monomeric and structurally stable. We propose to challenge our delivery system with more complex proteins while exploring protein attributes that might influence the material's mechanical and release characteristics.