By Dr. Stuart K Bisland.

Photodynamic therapy (PDT) is a non-invasive, oxygen-dependent treatment modality involving the dynamic use of specific-wavelength, visible light. The light is used to excite light-sensitive (photosensitizer) dyes within target cells and generate cytotoxic radical and non-radical derivatives of oxygen. The highly reactive nature of these derivatives limits the effects of PDT to the immediate vicinity of the photosensitizer. Consequently, the intracellular location of the photosensitizer plays an important role in determining the ultimate fate of the target cell(s) since certain subcellular structures within a cell are more prone to PDT-induced damage than others; mitochondria and nuclei are known to be particularly PDT-sensitive. It is therefore of interest to design photosensitizers which display precise intracellular targeting in order to maximize PDT-induced cell death. One such approach pioneered in our lab involves the use of short, synthetic peptide sequences which confer specific trafficking of photosensitizers into and inside targeted cells [1]. 
Using a similar branching strategy to that of Tam et al. [2,3], we have designed multi-branched peptide-based vehicles known as loligomers, harbouring eight identical branches each with a chlorin e6 molecule attached. Chlorin e6, a plant-derived photosensitizer, was seen as an ideal candidate for loligomer-mediated intracellular delivery based on two major criteria; its preponderance to accumulate predominately at the plasma membrane of cells [4] and its physicochemical properties: low-molecular weight, high purity, relatively high extinction coefficient at 401 nm (Soret maxima) in solution (e = 59,000 M-1cm-1) and Q-band absorbance in the red spectral region at ~654-664 nm. Furthermore, its structural formula is porphyrin-derived (see Fig. 1) with additional carboxyl-groups which can be activated using  dicyclohexylcarbodiimide and N-methyl-2-pyrrolidone [1] to facilitate attachment to N-terminal amino groups using standard solid phase peptide technology.

Despite the fact that the precise photochemical pathway responsible for PDT-induced phototoxicity remains controversial, it is accepted that the non-ionizing radiation associated with PDT is distinct from that of ionizing radiation (x-rays, g-rays) such that the use of PDT does not preclude the prior intervention of radiotherapy and other modalities including chemotherapy and surgery. Indeed, certain lesions unresponsive to ionizing radiation or chemotherapy, often respond to PDT [5].

Thus the combined potential for dual selectivity; with preferential retention of photosensitizers within neoplastic tissues and localized illumination of the tumor site, together with its cost-effectiveness and simplicity of use, resoundly justify the rational for developing clinical PDT along side already established protocols for the curative and palliative treatment cancers. Moreover, the application of PDT is not limited to the treatment of malignancies alone, and is also in clinical trials for the treatment of a number of non-cancerous conditions including, rheumatoid arthritis [6], fungal infections [7], vascular stenosis [8] and psoriasis [9].


In 1888, Marcacci unknowingly witnessed the photodynamic phenomena using quinine and cinchonamine which displayed enhanced toxicity towards plant and animal cells upon exposure to light [10]. The term ‘photodynamic action’, did not appear, however, for over a decade when in 1903, Professor von Tappeiner, inspired by the initial findings of his doctoral student, Oscar Raab, decided to pursue this phenomenon and introduced the term ‘photodynamic’ to describe the light-mediated enhanced cytotoxic of eosin on the skin [11]. Unfortunately, his interest in light-activated killing was not sustained, as collaboration between clinicians and basic scientists remained trivial and application for this photodynamic effect proved elusive. 

Despite the lack of government funding, interest was eventually rejuvenated in the 1960’s, when Schwartz, Winkelman and Lipson found that they could detect tumors in cancer patients using a crude mixture of light-activated porphyrin monomers and oligomers called “hematoporphyrin derivative” (HPD) to [12,13]. Skepticism persisted until the late 1970’s when Thomas J. Dougherty of Roswell Park Cancer Institute showed that HPD and red light provided a tumor-selective photodynamic therapy (PDT) for the treatment of animal tumor cells in culture and metastatic skin tumors in humans [14].

The dimer-oligomer components of HPD which confer tumor-localizing were subsequently identified and enriched, forming a stable preparation of dihematoporphyrinester-dihematoporphyrinether called “Photofrinâ”(PF) with significantly higher chemical purity than HPD. As a first generation sensitizer, PF is currently the only light-activated (photosensitizer) chemical approved by the U.S. Food and Drug Administration (FDA) for the clinical treatment of cancers using PDT. The modality is now well established as an investigational anti-cancer therapy for advanced malignancies and a wide spectrum of dysplasias. In addition, as of April of this year, another photosensitizer called Verteporfin or Visudyne™, is approved by the FDA for the treatment of exudative age-related macular degeneration (ARMD), the leading cause of blindness in the western world in people over the age of 50 [15]. Visudyne therapy is also approved in Canada, Switzerland and Malta.The clinical use of PF-PDT was first approved in Canada for prophylactic treatment of recurrent papillary bladder cancer in 1993 and has since been approved for the reduction of obstruction and palliation of dysphagia and completely or partially obstructing esophageal cancer [16]. In the U.S., Germany, Italy and France, PF-PDT is approved for advanced obstructing esophageal cancer and early-stage lung cancer, in the Netherlands, for obstructive and early-stage lung cancers and esophageal cancer and in Japan, for early-stage lung, gastric, cervical cancers and superficial esophageal and gastric cancers[17,18]. Despite its significant advance into the clinic, with the intervention of cheap, easy-to-use, readily portable diode lasers for light delivery, the strategic application of PDT as an alternative to ionizing radiation or chemotherapy has yet to be fully embraced by practicing oncologists. This fact is undoubtedly fuelled by the limitations associated with PF-PDT which include the chemical complexity of PF, its relatively low absorbance maxima (Q-band l= 630 nm), together with its tendency to localize in skin which results in a prolonged skin photosensitivity for up to 4-8 weeks after treatment. The low Q-band of PF offers poor tissue penetration [19,20] since visible light penetration of tissues is wavelength-dependent with attenuation at ­£ 580 nm due to absorption by haemoglobin and a plateau between 700-800 nm [21]. Thus for optimal light penetration into target tissues the photosensitizer should have a Q-band approaching 800 nm. Alternatively, when minimal tissue penetration is required eg. for carcinoma in situ of the bladder wall [22] or at colorectal tumor resection margins [23] green light can be used.

The need to improve PF-PDT has propagated a whole new area of scientific research and industry in search of new photosensitizers with greater therapeutic efficacy. Many of these so called ‘second generation’ sensitizers, such as benzoporphyrins (eg. Verteporfin; QLT PhotoTherapeutics) [24], purpurins (eg. Purlytin; Miravant Medical Technologies) [25], chlorins (eg. meta-tetra (hydroxyphenyl)chlorin: Foscan; Scotia Pharmaceuticals) [26,27], phthalocyanine derivatives [28,29], the prodrug, 5-aminolevinic acid (Levulan; DUSA Pharmaceuticals) [30] and other porphyrin-related compounds (eg. lutetium texaphyrin: LuTex; Pharmacyclics Inc.) [31,32], are now in stages I-III of clinical appraisal [33]. Moreover, the introduction of modern day lasers coupled to various optic-fibre technologies, offering efficient endoscopic or bronchoscopic light delivery and dosimetry, has dramatically broadened the ailments for which PDT can now be considered [34,35,36,37]. Clinical trials are presently being conducted for cancers of the gastrointestinal and genitourinary tracts, lung, ovary, bladder, brain and head and neck [38,39,40], together with  a growing number of pre-clinical trials targeting prostate and nasopharyngeal cancers [41]. Skin is a considerable target organ for PDT and with more than 800,000 new cases of non-melanoma skin cancers reported in the U.S.A each year alone, there is a demand for alternate therapies to resection, excisional surgery or radiation therapy. Pre-clinical trials are currently aimed at cutaneous T-cell lymphoma, Kaposi’s sarcoma [29], primary basal and squamous cell carcinoma [42,43], metastatic or unresectable skin cancers, bowen’s disease [44] and melanoma [9,31]. Pigmented melanoma is insensitive to PF-PDT due to the high absorbance of melanin, only the high molar extinction coefficient (e) of naphthalocyanines at 780 nm (eabout 500,000 M-1cm-1) facilitates some measure of tissue transparency [45]. Also under investigation are preneoplastic lesions; actinic keratoses, benign prostatic hyperplasia and Barrett’s esophagus [38,46,47]. For a comprehensive review of new photosensitizers and their application in the clinic, readers are referred to a recent review by Schuitmaker et al., 1996 [17].


Clearly, PDT is in its infancy as regards clinical application. In order for one to fully exploit the photodynamic effect and be able to harness this effect to specific target sites within the body, a better understanding of the photophysical and photobiological processes which underlie PDT at the cellular level is imperative. A comprehensive knowledge of structural and electronic features of the drug which predict pharmacokinetic and biodistribution behaviour is also essential. This topic was thoroughly reviewed recently by Boyle and Dolphin [48].

1.2.1Photochemistry of PDT

The process of PDT can essentially be divided into two stages. Photosensitizers are initially administered, usually systemically (i.v. or i.p.), and given time to localize into target tissue(s). Then using a strong monochromatic light source or laser, a specific wavelength of non-thermal, visible light (in the red or infrared region of the electromagnetic spectrum) is subsequently delivered to excite the sensitizer. The sensitizer in turn undergoes a sequence of photooxidation reactions that culminate in the generation of highly cytotoxic, oxygen-derived species most notably, singlet oxygen [49,50,51,52]. Other metabolites of oxygen implicated in this process include hydrogen peroxide, superoxide anions and the very potent hydroxyl radicals. An in-depth review of the many photochemical processes relating to PDT is provided by Ochsner [53]. 

In brief, the photosensitizer is activated to an excited singlet state upon the absorption of photons at the appropriate wavelength, it may subsequently decay from its singlet state back to ground state and emit fluorescence or alternatively undergo intersystem crossing to populate stable, triplet state orbitals [16,54]. Once excited to a triplet state energy, the photosensitizer can either undergo intermolecular transfer of electrons with nearby cellular membranes or amino acids and/or endogenous alcohols in a type I reaction. More commonly, the photosensitizer transfers energy to ground-state molecular oxygen in a type II reaction. Either reaction can result in significant intracellular damage of membranes and organelles although, the relative contribution of each will depend largely on the type of sensitizer being used as well as the intracellular environment; ie. the availability of molecular oxygen. For instance, Foscan can induce tumor necrosis via type I or type II pathways depending on the availability of oxygen in the tumor tissue [55], with type I reactions predominating under hypoxic conditions and type II predominating at high oxygen concentrations. A maximal type II phototoxic effect has been proposed to occur when intracellular oxygen pressure reaches 10 mmHg [56]. Similarly, type II reactions predominate in hydrophobic environments due to the enhanced solubility and lifetime of singlet oxygen (1O2) in lipophilic solvents [57,58], while type I-associated radicals exist more readily in polar, aqueous media [59]. 

The involvement of a type II reaction-associated generation of 1O2 in the majority of PDT-induced cytotoxic responses, both in vitro and in vivo , is supported by deuterium enhancement and azide inhibition of oxidative reaction rates [60,61] and direct phosphorescence (at 1270 nm) detection [62]. Chemical scavengers offer indirect detection of 1O2 although, the specificity, sensitivity and time resolution of such techniques may be limited to non-cellular systems only [63].The highly reactive nature of 1O2 limit its lifetime within cellular systems from 250 ns in the cytoplasm [64] down to as little as 100 ns within lipid membrane regions of the cell [65]. Consequently, in cells the potential sphere of influence of 1O2 has been estimated to be £ 50 nm of its site of primary generation [66], only 1:2000 of a human cell, assuming a diameter of 100 µm. By comparison, the absence of cellular materials with which 1O2 can interact in solutions, significantly prolong its lifetime to 3-4 µs in water [67,68] and 200 µs in organic solvents [69]. It is clear however, these values offer little information for predicting the behaviour of 1O2 in vitro, or in vivo, beyond the efficiency of the photosensitizer to generate 1O2

Conceivably, the ideal photosensitizer would possess a high triplet state energy (> 7900 cm-1), prolonged triplet state lifetime to allow efficient energy transfer to oxygen (tT³ 500 ns, a high triplet state quantum yield (³ 0.7) to maximize singlet oxygen production and high absorbance and molar extinction coefficient in the red/near infra-red region of the spectrum (e= 800-900 nm) to maximize tissue penetration and photon absorption respectively [57,70]. Even with optimal characteristics like these, the limited lifetime of 1O2 predicts that the parameter most influential in achieving the maximum PDT effect for any potential photosensitizer, is the ability to accurately deliver that agent to specific PDT-susceptible sites within the target cell(s) or tissue.

1.2.2Subcellular targets of PDT

Targets of 1O2-induced oxidative damage include membranes which may undergo lipid peroxidation, protein cross-linking and/or loss of ionic homeostasis [71], cytoskeletal elements such as nonpolymerized tubulin or actin [72,73], mitochondria; resulting in respiratory failure, endoplasmic reticulum and golgi apparatus; disrupting protein synthesis/routing [74] and lysosomes; which rupture releasing harmful acidic hydrolases into the cytosol [75]. Intranuclear DNA is generally not a primary target for 1O2 as lipophilic photosensitizers tend to accumulate at the nuclear membrane rather than inside the nucleus. However, HPD has been shown to inactivate enzymes involved in DNA repair in CHO cells [76] and cause chromosome aberrations, sister chromatid exchanges and single-strand DNA breaks in CHO and NHIK cells [77,78,79]. Similarly, DNA-protein cross-linking is also evident in mouse lymphoma L5178Y cells following chloroaluminium phthalocyanine (AlPc)-PDT [80]. 

PDT-induced mutagenicity has also been reported in Chinese hamster V79 and CHO cells following treatment with HPD [79] and following PF, AlPc or Pc4-mediated PDT in various cell lines (LY-S1, LY-R16 and WTK1) in which the tumor suppressor gene, p53, is inactivated [81,82]. The potential for expression and survival of PDT-induced mutants is reduced on homologous chromosomes, since an active copy of the gene is often available to function in place of the targeted one. From these studies, it is apparent that the propensity for mutation correlates more with the type of cell line being treated and the specific genes being attacked by 1O2 rather than the subcellular localization of the photosensitizer.

1.2.3Mechanisms of PDT-induced cell death.

Ultimately, the physico-chemical properties of a given photosensitizer will determine its pharmacokinetic behaviour and hence its phototherapeutic potential to kill cancer cells, as well as the mechanism of cell killing [48]. Death of PDT-targeted cell(s) can occur via one or more of three very distinct and well defined mechanisms: direct cell kill; indirect cell kill or immunological-induced cell kill. Direct Cell Kill

A direct cell death may occur in which the photosensitizer generates sufficient 1O2 during PDT to result in irreparable intracellular damage to the target cell(s). Treated cells subsequently undergo apoptotic cell death or necrosis [16]. Photosensitizers implicated in this type of death include benzoporphyrin derivative monoacid ring A (verteporfin) and disulfonated metallophthalocyanines (zinc or aluminium), both of which are amphiphilic in nature [83]. The propensity for direct cell kill correlates with the enhanced ability of cationic, lipophilic sensitizers to bind the plasma membrane and gain access to the cell interior [84], while the exact subcellular localization of drug is determined primarily by its net ionic charge. Once inside the cell, cationic compounds localize mainly in the mitochondria, lysosomes and/or intracellular membranes [85,86]. It is conceivable therefore, that death occurs due to mitochondrial dysfunction (disruption of the electron transport chain and mitochondrial electrochemical gradient), oxidative phosphorylation, or rupturing of lysosomes. However, a precise correlation between cell death and subcellular localization is further complicated by the fact that charged photosensitizers redistribute to other sites within the cell during the course of PDT treatment. Indeed, the primary target of PDT-induced redistribution appears to be the nucleus [87]. Indirect Cell Kill -Vascular Damage

The effects of PDT can be far more dynamic in the presence of a functioning vascular system, as in the case of a growing solid tumor. PDT-induced vascular damage usually results in an indirect cell kill as cells are deprived of life-sustaining nutrients and oxygen. A number of different cell types, are sensitive to PDT and contribute to PDT-induced vascular shutdown including endothelium, macrophages, neutrophils and platelets. A physiological cascade of events involving endothelial cell damage, vascular leakage and/or vasoconstriction, mirocagglutination of blood cells (leukocytes, neutrophils) and platelet aggregation, all combine to result in blood flow stasis [88]. Eicosanoids; leukotriene B4, thromboxane and cycloxygenase are thought to be the main biochemical mediators responsible for this response as inhibitors of eicosanoid release significantly abate the tumor response to PDT [89,90]. The release of eicosanoids is, in turn, thought to be initiated via the phospholipase C/phospholipase A2 and prostaglandin endoperoxide pathways [91], while PDT-induced inhibition of the endothelium-derived relaxing factor (EDRF), nitric oxide, potentiates the eicosanoid-induced vascular shutdown [92]. 

The dense stromal network and high interstitial pressures within a solid tumor limit the penetration of activating light into the tumor tissue (~1-2 mm) and the inherently deficient neovascular supply provides poor oxygen delivery to the cells [93]. As a result many cells within the centre of a growing solid tumor survive under highly hypoxic, low pH conditions and are resistant to standard radiation therapy. The high extravascular pressures of crowding cells and tortuous course of blood vessels within a tumor also provide a substantial barrier to drugs extravasating and internalizing into target tumor cells, thus the potential for PDT to ablate large solid tumors via direct cell kill is significantly reduced [94,95]. It would appear therefore, that solid tumors provide for a very unfavourable environment in which to deliver therapies reliant on intracellular uptake (ie. chemotherapeutic agents) or adequate intracellular oxidation such as radiation therapy or PDT. This makes it all the more fitting that the impenetrable barrier provided by tumor microvasculature should also turn out to be the ‘achilles heel’ of solid tumors subject to PDT. 

Solid tumors have to establish an adequate vascular network providing the cells with essential oxygen and nutrients in order to grow beyond a certain size (~1-2 mm diameter), disruption of this delicate network can result in nutrient deprivation, tissue hypoxia and eventual starvation and death of tumor cells (96). Tumor microvasculature, either derived from the host and incorporated into the growing tumor mass or angiogenically derived neovasculature is known to be anatomically, physiologically and biochemically distinct from non-tumor associated vasculature (my thesis). The potential for augmenting the therapy of any given PDT agent by influencing the vasoactivity of vessels which feed directly into a solid tumor is apparent. With his information, a growing number of PDT photosensitizers have been shown to mediate a tumor response via damage to the tumor microvasculature, including HPD, NPe6, purpurins, PF and protoporphyrin IX [88]. Unfortunately, the vascular effect, although certainly more pronounced in tumor microvasculature compared with the surrounding normal vasculature, is not exclusive to it [97]. 

The propensity for photosensitzers to promote vascular damage was addressed by Fingar et al., [98] in which the PDT effect of various zinc phthalocyanines of differing hydrophobicity was studied using a chondrosarcoma tumor model in Sprague-Dawley rats. A clear correlation between hydrophobicity and arteriole or venule constriction, venule leakage of albumin and increased tumor interstitial pressure was found, all of which could be abolished upon prior treatment with the specific cyclooxygenase inhibitor, indomethacin. The use of targeting modalities which are specific to tumor vasculature, whether it be peptide-based delivery vehicles or receptor-mediated uptake, have not been reported and vascular targeting has thus far relied primarily on the unremarkable differential in uptake of photosensitizer into tumor-associated vasculature as opposed to normal vasculature, together with the use of very localized activating light using optic fibres. Immunological response of PDT

A third means by which of PDT can result in cell death involves the release of cytokines, chemokines and related mediators of the inflammatory response. Damaged cells and host cells involved in immunity such as macrophages and lymphocytes are known to release cytokines which can contribute to both direct cell kill and vascular damage [99,100). The identity of these cytokines and their precise role in promoting cell survival or cell death via apoptosis or necrosis, remains to be defined. 

A number of inflammatory cytokines are up-regulated with PDT including endothelial-derived clotting factors (eg. factor VIII:C) [101], platelet-derived growth factor and granulocyte colony-stimulating factor (G-CSF) produced by endothelial cells, monocytes/macrophages, platelets and lymphocytes [102]. Interleukin-1b (IL-1b), IL-6 and TNF-a are thought to mediate the enhanced production of GCSF by PDT via distinct mitogen-activated protein kinase signaling pathways (MAPK), which in turn promote the enhanced recruitment and endothelial-adherence of neutrophils within the tumor [102,103,104]. Gollnick et al., [105] has previously shown that PF-PDT transiently elevates IL-6 mRNA expression in EMT6 mammary cells in vitro, with prolonged IL-6 expression (up to 24 h) and protein synthesis (up to 48 h) in vivo

The expression of proto-oncogenes, c-jun and c-fos, is also increased following PF-PDT of epithelial HeLa cells in vitro [106], and may be responsible for the transcriptional regulation of various genes which encode proteins that are induced by PDT including those mentioned above and others associated with protective, stress-response pathways; NF-kB, glucose-regulated proteins (eg. GRP78), haem oxygenase, heat-shock proteins (eg. HSP70) [107,108] and/or apoptotic cell death, such as the interleukin-1b-converting enzyme-related proteases (caspases). 

Indeed, it is becoming increasingly evident that PDT elicits a non-specific, pro-inflammatory response similar to that described for endotoxin; for further review of inflammatory pathways associated with endotoxin, TNF-a and interleukin 1, the readers are referred to that of Henry Movat, 1987 [109].  Perhaps more surprisingly, a specific, T-cell mediated immune response has also been implicated following PDT [110], suggesting that photosensitizers are able to be processed into the MHC-class I /II pathways.

1.2.4 PDT-induced apoptosis

Caspases 3,8 and 9 [111] and ceramide [112] are integral to PDT-induced apoptosis. Singlet-oxygen is thought to facilitate opening of mitochondrial transition pores which allow for the triggered release of cytochrome C from the mitochondria into the cytosol. The electron transport chain is subsequently disrupted upon the loss of mitochondrial cytochrome C while the newly appropriated cytosolic cytochrome C activates certain cytoplasmic proteins eg. apoptosis-activating factor 1 (APAF-1) [111]. APAF-1 in turn activates a cascade of caspases which result in apoptotic cell death. 

The integral role of mitochondria in PDT-associated apoptotic cell death and the fact that cancer therapy prognosis often correlates with the propensity for malignant cells to undergo apoptosis, implicates mitochondria as an important subcellular target for PDT. Drug resistance or suboptimal tumor response may relate to inappropriate targeting of the sensitizer to PDT-insensitive sites within the cell with subsequent insusceptability of target cells to apoptose [113]. Alternatively, the resistance of certain cell lines to PDT may also relate to their inability to undergo apoptosis, as exemplified by CHO cells expressing the bcl-2 gene encoding for a mitochondrial outer membrane protein which prevents the release of mitochondrial factors involved in apoptosis [114,115], or that demonstrate altered mtiochondria function [116]. It is conceivable that commital to an apoptotic cell death may require factors derived from the tumor, which are not present in vitro, as in the case of radiation-induced fibrosarcoma (RIF-1) cells exposed to Pc4-PDT [117]. Thus, besides genetic differences defining a cell’s susceptability to a given therapy such as PDT, it is also imperative to consider optimizing subcellular targeting of PDT agents in order to achieve maximum tumor response.


2.1 Physico-chemical properties of photosensitizers determine biodistribution

The application of novel strategies which target photosensitizing drugs to malignant cells is now an intense area of research. Since the pharmacokinetic behaviour of a PDT drug and its ability to accumulate into cells is largely predicted by its molecular structure, mass, charge and lipophilicity, many of the second generation photosensitizers are themselves being modified to provide favourable intracellular and intratissular localization patterns to further improve the efficacy of PDT. The subcellular site of photosensitizer is also known to determine both the type and extent of cellular apoptosis induced by PDT. The conjugation of specific peptide sequences to a photosensitizer offers a very versatile, reproducible means of optimizing delivery by manipulating each of the afore-mentioned physico-chemical parameters, thereby directing the drug to regions within a cell which are highly sensitive to PDT. 

Polymeric structures beyond a certain molecular size (~1000 Da) tend to enter cells by actively (energy-dependent) binding to phospholipid components of the cellular plasma membrane, being packaged into endosomal, coated pits  formed within the membrane, which eventually envelope to form endocytic vesicles, and entering into the cytosol [118]. This process is temperature dependent and considerably reduced at 4 oC compared with 37 oC. Based on molecular size, smaller molecules such as Ce6 (MW = 596.4) are more likely to enter cells via a temperature-independent passive diffusion. Unfortunately, like many photosensitizers which display a favourable interaction with negatively charged plasma membrane, Chlorin e6 localizes largely at cellular membranes [119] which may compromise its photodynamic effect. 

Insertion of polar substituents (eg. carboxylate, sulphonate or hydroxyl groups) confer hydrophilicity to allow systemic injection in vivo, while the simultaneous inclusion of unsubstituted rings (ie. hydrophobic matrix), result in an amphiphilic compound with the ability to cross lipid membranes and internalize tumor cells [120]. An example of which is Mono-L-aspartyl-chlorin e6,which can be regarded as the amphiphilic version of chlorin e6 [121]. The enhanced photodynamic activity of mono-L-aspartyl Ce6 compared with Ce6, may relate to its ability to endocytose, resulting in greater intracellular accumulation [122]. 

The importance of charge was also illustrated by Soukas et al [123] who found that by using differently charged poly-L-lysine conjugates of Ce6, the time course, mechanism of uptake and intracellular distribution of conjugates into human epidermoid squamous cell carcinoma cells (A431) or human endothelial hybrid cells (Eahy926), could be varied significantly. Cationic conjugates showed the highest uptake into cells (maximal by 6 h) and in contrast to anionic (maximal uptake at 6 h) and neutral conjugates (maximal uptake at 24 h), the process of internalization was not inhibited at 4oC. The importance of intracellular localization is further emphasized by the fact that the conjugate of neutral charge, with the least amount of uptake into either cell type tested, but displayed a discrete perinuclear localization and was the most potent photsensitzer. The nucleus and perinuclear region of a cell are known targets for phototoxicity [124,125] resulting from 1O2-mediated DNA damage. The authors elude to an indirect mechanism whereby aggregates of cationic conjugate to anionic regions of the plasma membrane are too large for absorptive endocytosis and suggest that this may also account for their relatively poor phototoxicity. This study therefore implies that the overall molecular charge of a photosensitizer may predict whether cellular uptake is via an ATP-dependent or ATP-independent mechanism. 

The inclusion of charged functional groups within the pyrrole rings of porphyrin actually acts to limit the tendency for aggregation due to electrostatic repulsion and steric hinderance  [126]. Aggregation is also influenced by different metal ions with differing peripheral and/or axial ligand coordinates, as in the case of the various metallophthalocyanines, while the addition of benzene moieties to each pyrrole, as in the case of benzoporphyrin derivative, can enhance red light absorption. 

A recent study by Noodt et al., [127] using V79 Chinese hamster fibroblasts exposed to a number of porphyrin-based sensitizers, implicates the degree of lipophilicity as an important determinant of photosensitizer subcellular localization.

In general, hydrophilic PDT drugs (eg. Highly sulphonated phthalocyanines and meso-tetra(4-sulphonatophenyl)porphine) tend to associate with albumin serum proteins in the blood, with prolonged retention (up to 120 h in the case of sulphonated chloro-aluminium (III)-phthalocynanine) in reticuloendothelial rich organs (ie. liver, kidney and spleen) and limited retention (up to 48 h) within the interstitial space and vascular stroma of tumor, due to a poor capacity for diffusion across lipid membranes into cells [128,129]. Clearance via the hepatic pathways is however more rapid from the skin (24-48 h). 

The expression of low-density-lipoprotein (LDL) membrane receptors is elevated in neoplastic cells and proliferating endothelia compared with non-proliferating tissues due to the heightened demand for exogenous cholesterol with increased rate of membrane synthesis [130,131]. Consequently, lipophilic sensitizers (eg. PF, HPD, unsulphonated pthalocyanines and purpurins) provide enhanced tumor localization by virtue of binding both high density (HDL) and low density lipoproteins (LDL) [132], with greater retention than hydrophobic drugs; up to 168 h in the case of zinc (II)-phthalocyanine [133]. Furthermore, clearance from reticuloendothelial tissues via the bile gut pathway is relatively fast (within 72 h). Strategies aimed at targeting lysosomes which are also potential targets; by releasing cytotoxic hydrolases into the cytosol (75), include the use of LDL conjugates [134,133]. Lysosomes are the site of LDL degradation, where upon an attached sensitizer can be subsequently released to have its phototoxic effect in that area of the cell. Receptor-mediated endocytosis of LDL-sensitizer conjugates offers the added advantage of cell-specific targeting. 

It is clear that a combination of lipophilicity and polarity will endow a photosensitizer with adequate water solubility for systemic injection (i.p. or i.v.), low skin retention, as well as the ability to selectively localize at sites of neoplastic growth and therein gain access to the interior of tumor cells. 


mPDT studies are conducted using an interstitial brain tumour model in rats. Tumours are grown to a size of 3-4 mm diameter within the rat brain at which time, treatment is given via permanently implanted optical fibres directly into the tumour mass. Custom-made, pocket sized light sources have been designed to deliver either short duration (10-30 mins) light treatments or continuous, low dose light over extended periods of time (up to 5 days). Immunocompetent and immnuodeficient animals are used to allow the growth of both murine- and human-derived cell lines and to assess the role of the immune system on PDT response. Throughout the light treatments animals are able to move freely throughout their cage by virtue of our custom-designed contact swivel and mPDT backpack. After treatment the extent of cells, both tumour and surrounding brain, undergoing apoptotic cell death is quantified from histological slices through the brain and tumour.

Stereotactic apparatus for intracranial implantation of tumour cells and optic fibre.

Shows Fischer rat receiving intracranial PDT via permanently implanted optical fibre. The high scattering properties of brain tissue effectively disperses the light.

Lasers of mPDT Link 5