Chiral Alkyl Groups at Position 3(1’) of Pyropheophorbide-a Specify Uptake and Retention by Tumor Cells and are Essential for Effective Photodynamic Therapy

Abstract

To investigate the importance of chirality and precise structure at position 3(1’) of pyropheophorbide-a for tumor cell specificity and photodynamic therapy (PDT), a series of photosensitizers (PSs) was synthesized: (a) with and without chirality at position 3(1’), (b) alkyl ether chain with variable number of chiral centers, (c) hexyl ether vs thioether side chain, and (d) methyl ester vs carboxylic acid group at position 17². Cellular uptake and specificity were defined in human lung and head/neck cancer cells. PSs without a chiral center, with an alkyl chain or thioether functionalities showed limited uptake and PDT efficacy. Replacing the methyl group at the chiral center with a propyl group or introducing an additional chiral center improved cellular retention and tumor cell-specificity. Replacing the carboxylic acid with methyl ester at position 17² lowered cellular uptake and PDT efficacy. A direct correlation between the PS uptake in vitro and in vivo was identified.

Graphical Abstract

Introduction

Chlorophyll-a-derived photosensitizers (PSs) have emerged as effective agents for photodynamic therapy (PDT)^1,2. While the first generation of clinically approved porphyrins, including Photofin and Foscan^3;4, gained broad application, the lasting skin toxicity has been found to be an undesirable side effect. The toxicity is caused by a strong retention of these porphyrins by non-tumor tissue such as skin. The development of the next generation PSs^1-5 has set the goal to identify phorphyrin derivatives that show enhanced specificity for tumor cells but reduced retention by stromal cells. Based on SAR analyses, chlorin derivatives with specific peripheral structural groups appeared to fulfill the sought-after properties^6,7.

Pyropheophorbide-a analogs with 3-(1’-alkoxy)ethyl substitution (the length of alkyl ether chain varied from 1- to 12 carbon) were found to exert a dominant function by determining the mode of cellular uptake⁸. By applying a number of murine tumor models, phyropheophorbides with either a hexyl or heptyl group at position 3, combined with a carboxyl group at position 17², emerged as best possible PSs for in vivo PDT^8,9. Subsequent cell biological analyses of the leading compound 3-[1’-(hexyloxy)ethyl]-3-devinylpyropheophorbide-a (HPPH) revealed that the alkyl group attached to position 3 affects four processes: (i) interaction with the plasma membrane, (ii) transmembrane diffusion into the cytoplasmic compartment, (iii) precise subcellular site of deposition, i.e. outer membrane of the mitochondria and ER, and (iv) kinetics of cellular release. The sum of these processes is perceived to be the basis for the cell type specificity of the PS retention¹⁰. The role of the alkyl group at position 3 onto these cell biological properties is most strikingly demonstrated in Figure 1 by three pheophorbides differing in the length of the alkyl group: without a side group (pyropheophorbide, PPH), with a hexyl group (HPPH) or with a dodecyl group (Dodecyl-PPH). Without a side group, PPH does not show any appreciable transient binding to the plasma membrane. It efficiently diffuses into cells and accumulates in mitochondria/ER. Intercellular retention is however short-lived because of the high egress rate of PPH. On the other end of the spectrum, Dodecyl-PPH, due to the presence of a large lipophilic side group, showed a strong binding to the plasma membrane without appreciable subsequent transmembrane diffusion. Cellular uptake occurs primarily through endocytosis followed by deposition and retention of Dodecyl-PPH in lysosomes. As noted for other lysosomally targeted PSs, there is no tumor cell-specific retention evident for Dodecyl-PPH. The addition of a 6-carbon ether chain to position 3 in HPPH appears to confer an activity that lies between PPH and Dodecyl-PPH and is marked by gaining a cell-specific retention.

Figure 1.

Effect of alkyl structure at position-3 of pheophorbides on binding to cell surface, uptake and retention by tumor epithelial cells. Co-cultures of HNT1 (p 61) and L324 T-Fb were treated with HPPH or PPH in serum-free DMEM for 30 min on ice (first column) or in DMEM containing 10% FBS for 24 h at 37°C (second column). The cultures were then chased for 24h in PS-free DMEM containing 10% FBS (third column). In a separate set of co-cultures, HNT1 (p 64) and CFSE-stained L364 T-Fb were similarly treated with Dodecyl-PPH. Phase and fluorescent images of the cultures were recorded.

What has not yet been defined is the role of the precise alkyl structure that directs the interaction with the phospholipid bilayer and its intrinsic components relevant for uptake and cell-specific retention of the porphyrin. Recent studies have implicated a marked contribution of isomeric forms of the tetrapyrrole^10,11 and side groups¹² on HPPH uptake and PDT in vitro and in vivo. Moreover, the application of HPPH to different patient-derived cancer cells isolated from lung and head and neck tumor lesions has indicated large individual differences in the specificity by which HPPH is retained¹⁰. In combination with the alkyl side group attached to position 3(1’), the carboxyl group at position 17² of HPPH enables long-term retention only in a subset of tumor cells. Replacing the carboxylic group with a methyl ester group altered uptake, subcellular distribution and prolonged retention in cells with reduced HPPH-binding activity¹¹.

Taken together, these findings suggest that the chiral center at position 3(1’) and precise alkyl structure may influence cell biological properties of the compound. A functional analysis of isolated epimers of HPPH, and the related aryl ether analog showed indistinguishable activity^11,12. Hence, we asked whether removal of the single chiral center at position 3(1’) or the introduction of a chiral center within the alkyl group will indicate a functional role of these for interaction with membrane components responsible of the tumor cell-specific retention. The importance of a methyl group forming the chiral structure at position 3(1’) has also been evaluated by replacing it with a propyl group. In addition, the contribution of the ether linkage was assessed by replacing the hexyl ether with a carbon chain or by a thio-ether functionality. This study presents the SAR analyses of 8 different alkyl group variants with and without chiral centers which are evaluated in combination with a carboxylic or methyl ester group at position 17². The data implicate the relevant contribution of specific structural arrangements for optimal activity.

Results and Discussion

Chemistry

For the preparation of pyropheophorbide-related PSs shown in Chart 1 following five synthetic strategies were applied:

Chart 1.

Partial structures of the PSs containing a variety of alkyl groups introduced at position 3 with ether, thioether or carbon-carbon linkages

PSs bearing an alkyl ether chain with two chiral centers:

These compounds were synthesized with the aim to investigate the impact of an additional chiral center in the existing hexyl ether chain (PS-1 and PS-2; Scheme 1), For the preparation of the pyropheophorbide-a analog 1, methyl pyropheophorbide-a 17 (PPH-ME) was first reacted with HBr/acetic acid, and the resulting unstable intermediate bromo-derivative was reacted with 2-hexanol. The methyl ester 1 isolated in 55% yield, which was hydrolyzed to the carboxylic acid 2 in quantitative yield.

Scheme 1:

Synthesis of HPPH analogs 1 and 2 containing an alkyl ether chain with two chiral centers

Alkyl ether analogs without chiral center at position 3(1’):

For the preparation of PSs without any chiral center at position 3(1’), 3-formyl pyropheophorbide-a methyl ester 18 was synthesized by following Tamiaki’s approach¹³ (Scheme 2). It was then converted to the corresponding hydroxymethyl analog 19, which on further reaction with HBr/AcOH /2-hexanol gave methyl ester analog 3 with a chiral center present only on hexyl ether side chain and no chirality at position 3(1’). Hydrolysis of the methyl ester with aqueous LiOH afforded the corresponding carboxylic analog 4. For the preparation of PS 5 bearing a hexyl ether group, but without chirality at position 3(1’), intermediate 19 was reacted with 1-hexanol, and the desired PS-6 was obtained by LiOH hydrolysis in 89% yield.

Scheme 2:

Synthesis of alkyl ether analogs of pyropheophorbide-a without chirality at position 3

Variable length of alkyl ether substituted pyropheophorbide-a analogs containing a propyl group instead of methyl substituent present at position 3(1’):

To investigate the importance of methyl functionality at position 3(1’) of HPPH for its biological properties compounds 7/8 and 9/10 were synthesized by following a similar approach (Scheme 3). In brief, 18 was reacted with allyl bromide in presence of activated Zinc¹⁴ (the activated Zinc is more reactive because the oxides at the surface are removed), yielded the allyl alkyl analog 20, which was first converted into the corresponding Zn(II) analog 21 before subjecting for hydrogenation to produce 22. Reaction of 22 with HBr gas/n-butanol under anhydrous reaction conditions gave PS-7 as major product along with a minor amount of compound 23. Hydrolysis of 7 with lithium hydroxide afforded the carboxylic acid analog 8. A similar approach was followed for the synthesis of PS-9 by reacting 22 with HBr gas/n-hexanol and the desired hexyl ether analog was isolated. As expected, the same dehydration product 23 was obtained as a minor component.

Scheme 3:

Synthesis of alkyl ether analogs with variable length of carbon units and the methyl functionality at position 3(1’) of HPPH is replaced with propyl group, and the dehydration product.

Position 3 alkyl substituted pyropheophorbides with variable number of carbon units:

To evaluate the importance of ether group in hexyl ether side chain, compounds 11/12 and 13/14 containing 4 and 8 carbon units, respectively, were synthesized (Scheme 4). PS-11 hydrolysis produced the respective carboxylic acid derivative 12. For the preparation of PS-13,3- formyl methyl pyropheophorbide-a 18 was reacted with heptyl triphenyl phosphonium bromide under Witting reaction conditions,¹⁵ and gave a mixture of 24 (major product) and 24a (minor). The isomeric mixture was separated by HPLC and the trans- and cis- configuration in both compounds was confirmed by NMR analysis. Hydrogenation of the mixture in presence of Pd-C afforded compound 13 in 95% yield, which on treating with LiOH in room temperature afforded PS-14 in 86% yield.

Scheme 4:

Pyropheophorbide analogs bearing alkyl groups with 4 and 8 carbon units at position 3. was obtained by hydrogenation of pyropheophorbide-a 23 as methyl ester, which on base

Importance of hexyl ether side chain vs thio-ether analog:

Methyl pyropheophorbide-a 17 was converted into the corresponding thio-ether analog by reacting with HBr/AcOH first and then with 1-hexyl thiol (Scheme 5). The methyl ester 15 was obtained in 61% yield, which on reacting with aqueous LiOH afforded the carboxylic acid derivative 16 in 88% yield.

Scheme 5:

Conversion of methyl pyropheophorbide-a to thioether analog

Functional characterization of PS-1 to PS-16 in primary human cancer cells

Several types of cell culture assays aimed to determine for each compound the five cell biological reactions indicated in the introduction and pictured in Figure 1: (1) extent of binding to cell surface or diffusion into cells when the cells are incubated in medium at 0°C (suppressed endocytosis), (2) kinetics of internalization and uptake, (3) subcellular site of deposition, (4) kinetics of cellular release, and (5) tumor cell specificity. In order to gain the information relevant for cancer cell types for which HPPH is intended to be used as PS for PDT, primary cultures of tumor epithelial and stromal cells were generated from lung and head and neck cancer tissue resected from patients treated in our institute. Since there is a substantial variation in tumor cell phenotypes in regard to the specificity of PS uptake and retention¹⁰ and the fact that this phenotype is subject to change as a function of progressive passages in culture (including senescence), reference sets of cultures of tumor and stromal cells with pre-established HPPH uptake characteristics were also applied to confirm to what extent the biological features of the new PSs are manifested. Considering the anticipated heterogeneity of primary cell cultures (no clonal selections have been applied) the location and quantity of PS in each culture set were visualized by fluorescent microscopic imaging of larger areas of the cultures. The high-resolution digital images (entire camera frames) were integrated in most of the composite figures shown in the following result section and in the supplementary section. The cell type specificity of the PSs was evaluated by comparative analyses of tumor epithelial and stromal cells. These analyses relied either on the assessment of monotypic cell cultures with pre-established HPPH uptake characteristics or, more frequently, on treatment of reconstituted co-cultures involving tumor and stromal cells on collagen-1 matrix¹⁶. To enhance distinction of cell types, especially in dense cell cultures, stromal cells were fluorescently tagged prior to incorporation into co-cultures (e.g. shown in Figure 1). The functional role of specific structures in pyropheophorbides was evaluated by using pair-wise applications of compounds differing only in that change introduced. HPPH or HPPH-ME served as common reference in most analyses.

In the following section, we present examples of the various experimental tools used to characterize PS-1 to PS-16. Results of these assays have assisted in the categorization of the PSs and have directed the analysis of other cellular functions of the same compounds. The data are then incorporated into the subsequent section, which summarize and discuss the specific functional properties of each of the structural elements introduced into the PSs (Chart 1).

Determination of cellular uptake of pheophorbides as a function of alkyl group at position 3 in combination with the presence of a carboxyl or methyl ester group at position 17^2:

An initial functional assessment of alkyl group at position 3(1’) determined its influence on the relative level of internalized PS and its retention by various tumor cell preparations. The example in Figure 2 involved HN-143 T-EC (tongue) cultures with the ability for long-term retention of carboxylated pheophorbide derivatives, in particular HPPH. Equivalent analyses of the PSs were carried with other cancer cell preparations with similar HPPH-binding activity, which includes HNT1 (tongue; early passages), HN-90 T-EC (tongue), FaDu-49 (hypopharynx, clonal line) and TEC-21 (lung adenocarcinoma; early passages). The assay consisted of incubating the cells with medium containing 10% serum and 1600nM of PS for 24 h followed by a chase period with PS-free medium for up to 72 h. The cellular level of PS was determined daily by fluorescent microscopy and quantification of the fluorescent signal (Figure 3). The assay allowed a rapid assessment of those pheophorbide derivatives with functional properties distinct from HPPH.

Figure 2.

Uptake and retention of HPPH derivatives by head and neck cancer cells HN-143 T-EC (p 27) were treated for 24 h with RPMI containing 10% FBS and the PSs indicated at the right. Retention of the PSs were tracked for 72 h by daily changes of culture medium. The fluorescent images of the cultures (exposure at 500 msec.

Figure 3.

Time course of PS retention by HN-143 T-EC cultures shown in Fig. 2. Net fluoresence intensity values for each original image frames were quantified.

The level of initial 24-h uptake of the carboxylate derivatives was similar for HPPH, PS-4 and PS-2, but lower for PS-8, PS-16 and PS-10, and particularly low for PS-14. The level of PSs taken-up gradually decline over 72 h-chase period with characteristic differences in egress rates reflecting the influence of the side group structures. In HN-143 T-EC cultures, PS-8 displayed the slowest release rate among the compounds with high initial uptake resulting in this derivative being the highest retained PS after 48 h. A distinctly different behavior has been determined for the methyl ester derivatives. As demonstrated by the examples in Figure 2, uptake of the methyl ester derivatives by HN-143 T-EC is consistently lower than that of the corresponding carboxylate forms. Of note is that PS-1, among all methyl ester derivatives, showed highest uptake and a slow release rate over the subsequent 72 h-chase period. At this time point, the retention level of PS-1, along with HPPH-ME, was comparable to, or even exceeding that of the best carboxylate derivatives (Figure 3).

Imaging of the cells at higher magnification over the course of uptake and chase period coupled with inclusion of organelle-specific tracker molecules (data not shown, see detailed description of the tools used by us in ref 11) indicated that carboxylate derivatives were primarily retained in the mitochondria and ER compartment (Figure 4). In contrast, all methyl ester derivatives showed dual site of deposition that included to the various degree mitochondria/ER and dense granules. The latter organelles are distinct from lysosomes, Golgi, and peroxisomes¹¹.

Figure 4.

Effect of carboxyl and methyl ester group at position 17²- on subcellular localization of PSs. Fluorescent images of HN-143 T-EC cultures were recorded at 400X magnification after 72 h chase period.

The association of the methylated PSs with non-mitochondrial subcellular structures correlates with the marked long-term cellular retention as we have already observed previously with an analogous of 3(1)’iodobenyloxy) ethyl derivative¹¹.

The evaluation of the pheophorbide derivatives has been extended to tumor epithelial cells, which have evolved an altered specificity for retaining porphyrins. This cell type exhibits a substantially reduced capability to retain carboxylate pheophorbides (e.g. HPPH) but still strongly retains the corresponding methylated forms. These cells permit a more refined assessment of the functional role the alkyl structures for uptake and retention. An example of a particularly low HPPH-retaining cell culture HN-77 T-EC (larynx) is shown in Figure 5A. Similar results were also obtained with other epithelial cancer cell preparations with low HPPH-binding activity, including HN-142 T-EC (larynx), FaDu-124 (hypopharynx, clonal line), HNT1 (tongue, high passage), L237 (lung adenocarcinoma) and TEC-1 (lung squamous carcinoma). While the 24-h cellular uptake of PS-2 exceeds that of HPPH, within 24 h chase most of PS-2 and HPPH are removed again (Figure 5B). In contrast, both uptake and 48 h retention of PS-1 (methylated form of PS-2) are above that for HPPH-ME (Figure 5A & B), suggesting PS-1 as being a particularly effective PS in this tumor cell culture. The relative high retention of PS-1 has also been evident by the comparison with other methyl ester compounds such as PS-3 and PS-5 (Figure 5B). The assessment thus far has provided a rough categorization of the potential function of the alkyl structures at position 3(1’) in the context of the carboxylate and methyl ester version of the pheophorbide. The results indicated that the definition of biological activity of the PSs under investigation strongly depends on the cell type used for analyses and the time course of the PS treatment.

Figure 5.

A & B Altered specificity of PS retention in a subgroup of epithelial cancer cells. Confluent cultures of HN-77 T-EC (p 13) were treated for 24 h with RPMI containing 10% FBS and 1600 nM PS followed by a 48h chase with RPMI containing 10% FBS. Fluorescent images of the cultures were recorded (A) and net fluorescence intensity values quantified (B).

To evaluate the role of the alkyl structures on the initial binding of the PS to the target cells, sets of selected tumor cells were incubated for 30 min at 0°C with serum-free medium containing 1600 nM compound. The treatment at reduced temperature was chosen to suppress endocytosis but also to reduce transmembrane diffusion. Previous studies have already demonstrated that under such condition both HPPH and HPPH-ME show predominant binding to the cell surface with minimal evidence of diffusion into cells (see Figure 1). In contrast, pyropheophorbide (PPH; without alkyl structure at position 3(1’) displays low level surface binding at 0°C but substantial diffusion into the cells (Figure 1). Surface-associated PSs, such as HPPH or HPPH-ME, do efficiently diffuse into the cells as soon as the temperature is raised to 37°C. The uptake by diffusion is distinct from the kinetically slower endocytosis^17,18. A striking example of how the peripheral structure on pheophorbide is able to affect membrane binding is shown in the example of PS-15 and PS-16 interacting with hypopharyngeal carcinoma cells FaDu-124 in comparison to the activity of HPPH and HPPH-ME (Figure 6).

Figure 6:

Linkage of hexyl moiety at position 3 affects cell surface binding. FaDu-124 cellls were treated for 30 min on ice with serum-free DMEM containing 1600 nM PS as indicated at the right. The cell-associated fluorescence was recorded. The cells were incubated with fresh serum-free medium for additional 30 min at 37C for detection of internalization by diffusion. The cultures were then incubated for additional 24 h in DMEM containing 10% FBS and 1600 nM PS followed by a 24 h chase with DMEM containing 10% FBS. Fluorescent images of the cells at 400X magnification were recorded.

The substantial difference in cell surface binding between HPPH and HPPH-ME is largely eliminated by substituting the hexyl ether by a hexyl thioether structure. Both PS-15 and PS-16 display similar levels of surface binding and subsequent diffusion into the cells that is intermediate of that for HPPH and HPPH-ME. Prolonged incubation of the cells with PS yielded comparable levels of intracellularly accumulated PSs with the characteristically distinct site of deposition for the carboxylate and methyl ester derivatives (Figures 4 and 6). Within 24h chase period, the carboxylate PSs are removed while the methylated compounds remained. The observed difference in uptake and retention process suggest that the alkyl structure at position 3 in combination with the structure at position 17 affect interaction with plasma membrane, but the position 17 structure primarily dictates the intracellular organelle association and egress process.

The cell type-specific retention of PSs has been interpreted to result from the distinct egress rates detected in tumor epithelial and stromal cells^10,19. The influence of modified alkyl structure on this property has been determined by either comparative treatment of monotypic cultures of tumor-derived epithelial and stromal cells or in reconstituted co-cultures of these cell types^10,19 (Figure 1).

Manifestation of structure-dependent changes on tumor cell-specific retention is shown by the examples of PSs with carboxylic acid groups in Figure 7 using a co-culture of tongue tumor-derived HNT1 and stromal cells. HPPH indicates the prominent 24-h retention by the tumor epithelial cells with substantially lower retention by the stromal cells. The hexyl thioether PS-16 attenuates the tumor-cell specificity by an enhanced retention by stromal cells. A drastically switch is found for PS-14 in which a 7-carbon chain promotes a stronger retention in stromal cells with loss of tumor cell specificity. This altered cellular retention correlates with an association of PS-14 with dense granules rather than mitochondria/ER. PS-2 with two chiral centers exhibits the opposite trend with a further reduced retention by stromal cells and, thus, yielding a more pronounced tumor cell specificity.

Figure 7.

Tumor cell-specific retention of PSs. Co-cultures of HNT1 cells (p 103) with CFSE-labeled HN-101 T-Fb were incubated for 24 h in DMEM containing 10% FBS and 1600 nM of the indicated PS followed by 24 h chase with PS-free medium.

The relative contribution of the various alkyl structures attached to position 3(1’) of the 16 PSs (Chart 1) to cellular uptake and retention was evaluated by applying the experimental approaches as illustrated in Figures 2-7. Since most of the SAR analyses were carried out on primary tumor cell cultures, which often are only available in low quantity and have limited proliferative activity in vitro, thus, preventing us to perform replicate analyses, we validated the basic findings by using cell cultures from separate cancer cases. Attention was given to the distinct PS preferences of each culture (i.e., ability to retain HPPH or HPPH-ME). The functional information is summarized below in reference to the four groups of alkyl structures tested: (1) number of chiral centers (PS-1 to PS-6); (2) chiral propyl group at position 3 (PS-7 to PS-10); (3) carbon chain at position 3 (PS-11 to PS-14); and (4) hexyl thioether at position 3 (PS-15 and PS-16).

Chiral methyl group promotes cellular retention

PSs containing a carboxylic acid functionality e. g., PS-2, PS-4 and PS-6 display cell surface binding and uptake similar to (Figure 2) or even exceeding (Figure 5) that of HPPH. However, egress rates for these three PSs are higher than for HPPH resulting in a long-term retention level that equals or is below that of HPPH. In few cases of epithelial tumor cells with low retaining activity for HPPH, PS-2 and PS-4 show retention levels after 24h chase periods that are up to two-fold higher than HPPH (Figure S34). In all cell systems tested, retention of PS-6 was below than that of HPPH (Figure S35).

The functional role of the different chiral alkyl groups becomes more apparent when tested as methyl ester pheophorbides PS-1, −3 and −5 (Figure 8, see also Figures 2 and 5). The presence of an isohexyl ether group (PS-1 and PS-3) contributes to a several-fold higher binding activity to and uptake by all cell types (EC and Fb). Of note is that PS-3 and PS-5, both missing the chiral methyl group at position 3(1’), exhibit a substantially reduced binding to the plasma membrane of tumor and stromal cells but diffuse effectively into the cells (Figures 8 and 9). The same property has been detected in lung cancer cells (Figure S36). Moreover, a notable difference is found in the egress rates. While PS-1 shows a slow release from the cells, PS-3 has an accelerated egress that is particularly evident in fibroblasts. This difference accounts for the pronounced differential PS-3 retention by EC in co-culture with stromal fibroblasts (Figures 2 and 9). The analysis of PS-5 [without a chiral center at position 3(1’)] indicated accumulation and retention properties similar to HPPH-ME (Figure 8). Taken together, these results suggest that the methyl chiral structure introduced by isohexyl ether group enhances adhesion to cell surface and the methyl chiral structure at position 3(1’) enhances retention of the methyl ester derivatives to dense granules. The chemical structure of PS-5 and PS-6 proved, however, to be unfavorable because these derivatives precipitate in the standard Tween-80-containing formulation when stored at 4°C and, thus, limits their further use.

Figure 8.

Cellular binding, uptake and retention of methyl ester derivatives of PSs. Subconfluent cultures of HNT1 cells (p 91) were treated for 30 min on ice with serum-free DMEM containin the indicated PSs. After fluorescence imaging (exposure 4 sec at 100X), the cultures were incubated for additional 24 h in DMEM containing 10% FBS and 1600 nM PS (fluorescnet images taken at 500 msec). The cultures were then incubated for additional 24 h in DMEM containing 10% FBS and imaged (exposure 2 sec).

Figure 9.

Cell-specific retention of chiralic PS structures. Co-culture of HNT1 cells (p 72) and CFSE-stained HN-87 T-Fb were treated for 24h with DMEM containing 10% FBS and 1600 nM of the indicated PSs. After a 24-h chase with DMEM containing 10% FBS, the fluorescence of the cultures were imaged under idential conditions. PS-3 is not retained by stromal cells.

Chiral propyl group in combination with butyl ether yields a highly retained PS

The contrasting functionality detected when comparing HPPH and PS 1 / 2 with PS 5 / 6 (Figures 8 and 9) suggests that the methyl chiral structure at position 3(1’) contributes to higher uptake and retention. To identify whether this function is specific to a methyl group, it was replaced by a propyl group (PS-9 and PS-10). Since the modification increased the peripheral structures at the position 3 from a 7-carbon to a 9-carbon moiety and, thus, raising lipophilicity from a LogP value of 6.01 (HPPH) to 6.56 (PS-10), a compensatory modification was generated by replacing the hexyl ether moiety by a butyl ether moiety (PS-7 and PS-8). This modification lowered the LogP value for PS-8 to 6.01 (Table S1). Functional analyses using T-EC preparations with appreciable HPPH-binding activity (e.g., Figure 2) indicated that the carboxylated forms (PS-8 and PS-10) differed in uptake and retention. PS-8 is generally retained better than HPPH (Figure 3), whereas PS-10 is taken up at a much lower level. PS-8, like HPPH, is primarily localized to mitochondria/ER and is removed with kinetics similar to that for HPPH (Figures 2 and 3). In contrast, PS-10, despite its lower uptake, is retained for extended time period (Figures 2 and 3) what is in part due to its association with dense granules (Figure 4).The difference in cellular retention of PS-8, and PS 10 is even more pronounced in cells with low HPPH-binding activity (Figure S37). In these cells, PS 8, like HPPH, PS-2 and PS-4, are rapidly removed within 24h, while PS-10, although being a carboxylic acid analog, is retained by tumor cells similar to HPPH methyl ester (HPPH-ME). Microscopic analyses confirmed the distinct subcellular site of retention of the PS-8 and PS-10 (Figure S38).

A test of PS-7 to PS-10 for tumor cell-specific retention in co-cultures revealed that PS-7, PS-9 and PS-10 had, like HPPH-ME, low level uptake but strong retention by fibroblast. This property obscured the manifestation of tumor cell-specific retention as evident in reconstituted co-cultures. Optimal properties were detected for PS-8 that demonstrated high retention by epithelial cells but rapid egress from fibroblasts as HPPH. For a number of T-EC preparations that indicated modest to high HPPH-binding activity, PS-8 yielded most prominent tagging of tumor cell clusters (Figure 10). This favorable biological property of PS-8 has been further studied in the context of tumor tissue and in vivo (see below).

Figure 10.

Effect of alkyl structures on cellular retention of pyropheophorbide derivatives. Co-culture of HNT1 cells (p 25) and CFSE-stained HN-100 T-Fb were treated for 24h with DMEM containing 10% FBS and 1600 nM of the indicated PSs. Followed by a 24 h chase with DMEM containing 10% FBS.

Carbon chain at position 3 reduces solubility and cellular uptake

The functional role of the hexyl ether moiety at position 3(1’) has been evaluated by substituting it by a propyl (PS-11 and PS-12) or heptyl (PS-13 and PS-14) group. The preparations of PS-11, PS-12 and PS-13 indicated low solubility in the standard Tween-80-based formulation. The modification of pheophorbide with 3-carbon chain (PS-11 and PS-12) yielded compounds with biological properties that are comparable to those noted for PPH-ME and PPH (Figure 1). These include low level binding to cell surface but immediate transmembrane diffusion into the cells (Figure S39). While PS-11, due to the methyl ester function at position 17 becomes firmly associated with dense granules, PS-12, like PPH, is rapidly released again. The extended carbon chain rendered PS-13 poorly soluble and also to be taken up at very low level by all cell types. This property eliminated PS-13 from further characterization. The carboxylate derivative PS-14 remained soluble in the standard formulation. PS-14 is taken up by all cell types at much lower level than HPPH. PS-14, unlike HPPH, is primarily associated with dense granules (Figure S38). The distinct subcellular deposition of PS-14 also eliminates the tumor cell specific retention property of HPPH (Figure 7). These observations suggest that hexyl ether group in HPPH serves as a key element for determining mode of uptake and site of deposition.

Hexyl thioether group alters interaction with cell surface and uptake but not egress from mitochondria/ER compartment

The role of the hexyl ether group has also been evaluated by incorporating a hexyl thioether in place of a hexyl ether at position 3 (PS-15 and PS-16). This modification alters the binding activity of both compounds to cell surfaces (Figure 6). Yet, the quantitative levels of uptake of both compounds are generally below that of HPPH or HPPH-ME (Figures 2 and 3). The subcellular sites of deposition and kinetics of subsequent egress for PS-15 and PS-16 are comparable to those determined for HPPH-ME and HPPH, respectively. However, a trend of PS-16 to become associated with dense granules has been detected in various epithelial and stromal cell preparations what contributes to the attenuated differential retention defining cell type specificity of PS-16 in co-cultures (Figures 7 and 10). These results suggest that the precise structure of the hexyl linkage to position 3 affects interaction with membranes at the cell surface and intracellular structures. Although the thioether linkage causes a marked change in PS binding activity of pheophorbide, the overall effects of this has not improved its biological properties over that of HPPH.

Photoreaction mediated by the new series of PSs

As evident from the time course studies (e.g. Figures 2, 3, 5 and 6) the level of intracellular porphyrin is strongly dependent on the overall structures of the pheophorbide, the phenotype of the cellular target and time point and conditions of treatment. Hence, the photosensitizing action and therapeutic potential of each compound at the time of light excitation is not predictable a priori without determining the actual cellular level of the compounds. Considering that the photophysical properties of each of the 16 PSs have been noted to be comparable since alkyl ether substitutions did not affect the photophysical properties (singlet oxygen, fluorescence quantum yield) including the photobleaching of chromophore⁶, we have used the level of PS-specific fluorescence as predictor for the quantitative manifestation of the expected photoreaction. Recent studies using different primary cultures of tumor cells have confirmed a close correlation of fluorescence with light-mediated photoreaction and lethal PDT response^10,19. As biomarker for the light-dependent PS action served the oxidative covalent cross-linking of latent STAT3^20,21. The co-current activation of the p38 stress MAPK pathway and degradation of EGFR assisted in the evaluation of the cellular PDT response¹⁰.

The PS dose-dependent photoreaction in HN-85 T-EC cultures after treatment with PS-1 and PS-2 is shown as representative example (Figure 11). Inclusion of HPPH and HPPH-ME allows a quantitative grading of the PS efficacy. The results confirm the correlation of PS fluorescence with STAT3 crosslinking. Of note is that this STAT3 biomarker is independent of the specific pheophorbide structure and of the cell type used for the assay. The result in the case of HN-85 T-EC cultures after a 24-h chase period indicates PS-1 as being the most effective PS. The same outcome has been determined for other tumor cell cultures with low HPPH retaining activity, e.g., HN-77 T-EC (Figure 5) or TEC-1-2 cells.

Figure 11.

PS dose-dependent photoreaction identified by the molecular modification of signaling proteins. HN-85 T-EC (p 2) in 24-well plates were treated for 24h with RPMI containing 10% FBS and the indicated concentrations of PSs. After a 24-h chase with RPMI containing 10% FBS the cells were imaged for quantification of PS fluorescence and then exposed for 9 min to 665 nm light (3 J/cm²). Cells were immediately extracted and proteins analyzed by Western blotting for the indicated proteins

In agreement with previous findings¹⁰ that demonstrated the close relationship of PS concentrations used for incubation of the target cells, level of PS taken up and retained by the cells, and conditions of light treatment correlate with the degree of cellular damage leading to impaired proliferation and cell death²⁴, we could confirm the same for the PSs used in this study. A representative example is shown in Figure 12, that indicates the PS dose-dependent PDT response in mediating death of lung tumor-derived TEC-1-2 cells. The calculated IC50 values (ranging from 50 to 500 nM under the conditions used in Figure 12A) correlated closely with the cellular level of the PSs determined prior to light treatment. Cellular survival or lethal outcome outcome was found to be determined by just few-fold difference in PS concentrations (Figure 12A). This difference is sufficient to achieve a tumor cell-specific elimination as demonstrated by the comparison of HPPH and PS-2 (Figure 12B). Due to the higher egress rate, the level of both PSs in the tumor stromal cells fell below lethal concentration accounting for the post-PDT recovery of stromal cell proliferation.

Figure 12.

PS-specific photoreaction resulting in tumor cell death. A, TEC-1-2 cells (p 34) were treated for 24 h with DMEM containing 10% FBS and increasing concentrations of the indicated PS. After washing with medium without PSs, the cultures were treated with 665 nm light for 9 min (3 J/cm²) and then incubated for additional 24 h. Surviving cells were counted and expressed as percentage of the control cultures. B, Co-cultures of TEC-1-2 cells (p57) and L318 T-Fb (p10, stained with CFSE) were treated for 24 h with DMEM containing 10% FBS and 1600 nM of HPPH or PS-2. The cultures were incubated for an additional 24 h in PS-free medium. The cell-associated fluorescence was imaged and the cultures were exposed to PDT as in A. The cells were incubated for additional 4 d followed by imaging the surviving cultures. The level of PS-2 was sufficiently high in TEC-1-2 cells to mediate lethal outcome, whereas HPPH was close to the IC50 level with surviving tumor cell clusters emerging (indicated by arrow). In each culture, co-cultured tumor stromal cells survived PDT.

Enhanced binding activity of PS-8 identifiable in primary tumor cell preparations and xenografts

The assay of the 16 pyropheophorbide compounds in different primary human cancer cell cultures indicated PS-8 as being highly effective in a number of cancer cases noted to retain carboxylate pyropheophorbides. Often, retention of PS-8 exceeded that of HPPH (e.g., Figures 2, 3 & 10). Since the in vitro assay cannot appropriately reproduce the in vivo conditions defining uptake and retention by cancer cells in the context of tumor tissue, we asked whether the specific activity of PS-8 could be experimentally confirmed in vivo. Among our collection of tumor cell cultures derived from patient’s samples, we Identified several cases that could also be propagated as xenografts in SCID mice. Cell cultures of three cases indicated retention of PS-8 that exceeded that of HPPH. The derivative of a lung adenocarcinoma, TEC-21, was used for further analysis. Tissue cultured TEC-21 cells demonstrate uptake of PS-8 that exceeded HPPH and PS-4 by ~2-fold (Figure 13A). A subsequent 48 h chase period indicated a particularly effective retention of PS-8 resulting in several-fold higher level than for HPPH or PS-4. Light treatment triggered a photoreaction and STAT3 crosslinking that was proportional to the fluorescent signal of the accumulated PSs (Figure 13B). The same cells were grown as tumors in SCID mice. Injection of either HPPH or PS-8 into TEC-21 xenograft-bearing mice resulted within 24 h in an accumulation of the PSs that mirrored the PS accumulation specificity of the same cells in tissue culture (Figure 13C).

Figure 13.

Enhanced binding of PS-8 by TEC-21 lung tumor cells in vitro and in vivo. A, TEC-21 cells (p 8) were treated for 24 h with RPMI containing 10% FBS and the indicated PSs. After imaging the cell-associated PS fluorescence, the cultures were chased with RPMI containing 10% FBS for 48h and imaged again for determining the level of retained PSs. B, The cells were treated with 665 nm light for 9 min followed by immunoblot analysis of the indicated proteins. C & D, Aliquots of the same TEC-21 cell cultures were grown as xenografts in SCID mice. HPPH and PS-8 were injected intravenously into two tumor-bearing mice. After 24 h, the tumor tissue distribution of the PSs was determined by fluorescent microscopy of 10-mm cryosections (C). Two consecutive cryosections of one tumor removed from a non-PS-treated mouse were used to define HPPH and PS-8 binding by in vitro treatment of the section with the PSs (D). Fluorescent images were matched to the H&E staining pattern of the same cryosections.

As a part of independent study of tumor tissue, we have discovered that cell cultures, following formalin fixation, but avoiding organic solvent extraction, bind porphyrins with the qualitative profiles of live cells. The subcellular site of retention was essentially identical to that observed in live cells. The same PS binding activity is manifested by tumor cryosection following treatment ex vivo with PSs (Tracy et al, in preparation). This information was used to confirm the binding specificity of PS-8 in comparison with HPPH. Formalin-fixed cryosections of TEC-21 xenografts were treated for 24 h with PS-containing media followed by a 24-h-chase with PS-free medium as done with live TEC-21 cell cultures in Figure 13A. The tissue distribution and relative binding of HPPH and PS-8 (Figure 13D) proved to be the same as detected following in vivo PS treatment (Figure 13C). This result indicates that the alkyl structures at position 3 of PS-8 confer unto the pheophorbide an improved binding to intracellular membranes of TEC-21 tumor cells. Future studies involving membrane fractions of subcellular organelles isolated from tumor cells with distinct PS preference will attempt to define the molecular components responsible for pheophorbide-binding specificity. Lastly, the present work has not only yielded new reagents to probe porphyrin-binding process but also indicated new diagnostic tool for testing primary tumor tissue for its PS preference. The latter may assist in designing optimized PDT procedures for suitable cancer cases.

Conclusion

The results presented in this study suggest that in pyropheophorbide-a series, including the lead photosensitizer HPPH (trans-reduced ring D), the chiral center (1’-hexyloxy)ethyl group plays an important role in tumor cell specificity and in vitro/in vivo PDT efficacy. The removal of the chirality at position 3(1’) significantly diminishes cellular uptake/retention and, consequently, the level of photoreaction and PDT efficacy. These findings are complementary to those previously made with the structurally related m-(iodobenzyloxy)ethyl pheophorbide-a derivative¹¹. By replacing the methyl group at 3(1’) position with another alkyl functionality (e. g. propyl) the resulting product retains a biological activity comparable to HPPH, suggesting the methyl group at the chiral center is not functionally essential. Addition of another chiral center in the alkyl ether side chain of HPPH enhances the cellular uptake/retention and PDT of HPPH. Replacing the hexyl ether sidechain with an alkyl group (with variable number of carbon units) diminishes the tumor cell specificity of HPPH. All compounds investigated possess similar absorption and fluorescence characteristics and overall lipophilicity, yet, indicated structure-dependent differences in binding, uptake and retention by cells derived from human head & neck and lung cancer tissues. These properties vary among tumor cells derived from different cancer cases providing new tools to define cancer cell phenotypes. We also provide evidence that the alkyl structure-dependent binding specificity of tumor cells can be identified in patient-derived tumor tissue sections. Efforts are currently underway to extend this knowledge to other NIR photosensitizers (e.g., phthalocyanines, bacteriochlorins and BODIPY analogs), and to investigate the interaction of these photosensitizers with plasma proteins and cellular membrane components to improve the selection of photosensitizers for PDT application in vivo.

Experimental Procedures

The reactions were carried out under dry argon atmosphere. The completion of the reactions was monitored by thin layer chromatography (TLC) on pre-coated silica-gel sheets (layer thickness: 0.2 mm). Column chromatography was performed using silica gel 60, and/or preparative TLC. All compounds, including intermediates were >95% pure. UV/Vis spectra were recorded on an FT UV/Vis spectrophotometer using methanol as a solvent. Mass spectrometry analyses were performed at the Mass Spectrometry Facility, University of Buffalo, NY (USA). ¹H and ¹³C NMR spectra were recorded on a Bruker Avance III HD spectrometer at 28 °C (400 MHz for ¹H and 100 MHz for ¹³C). Chemical shifts are reported in parts per million (ppm) relative to tetramethylsilane (TMS) and were calibrated to the residual solvent peaks of CDCl₃ (δ = 7.26, 77.0 ppm). Coupling constants (J) are reported in hertz (Hz) and refer to apparent and visible multiplicities. ¹H NMR multiplicities are reported as follows: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, br = broad.

Synthesis of compound 1:

Compound 17²² (50.0 mg, 0.09 mmol) was dissolved in 2 mL HBr/AcOH, the reaction mixture was stirred for 1 hour at room temperature. The solvent was evaporated under high vacuum, the resulting crude was dried, and the entire crude was dissolved in dry CH₂Cl₂ (10 mL). A 50 mg portion of dry K₂CO₃ was added to this mixture, followed by addition of 1-hexanethiol (0.336 mL, 2.7 mmol). The resulting reaction mixture was stirred for 1 h at room temperature and then worked up with water and DCM. The crude mixture was purified by flash column chromatography (silica gel, 3% MeOH in CH₂Cl₂). Evaporation of the appropriate elutes gave the desired compound 1, 50% (29.7 mg) yield. MS (ESI) m/z: 651.39 (M + H⁺). HRMS (ESI): calcd for C₄₀H₅₁N₄O₄ (M + H⁺) 651.3929; found, 651.3904. UV-vis (MeOH, λ_max, nm (ε)): 660 (4.59 × 10⁴), 603 (9.46× 10³), 536 (9.46× 10³), 505 (9.46× 10³), 407 (9.18× 10⁴).

Synthesis of compound 2:

Compound 1 (15 mg, 0.024 mmol) was dissolved in degassed THF (4 mL) and 2.5 mL degassed methanol and a degassed solution of LiOH ·H₂O (21.16 mg, 30 equiv) (2.5 mL, 1:1 v/v) was added. The entire reaction mixture was then stirred under nitrogen atmosphere at room temperature for 2 h, and the resulting mixture was diluted with dichloromethane (20 mL). The reaction mixture was washed three times with water. The organic layer was separated and dried over Na₂SO₄, and the solvent was removed under reduced pressure. The residue obtained was purified by flash column chromatography (silica gel, 5% methanol in CH₂Cl₂). Evaporation of the appropriate elutes gave the desired compound 2, 86% (12.62 mg) yield. MS (ESI) m/z: 637.4 (M + H⁺). HRMS (ESI): calcd for C₃₉H₄₉N₄O₄ (M + H⁺) 637.4028; found, 637.4031. UV-vis (MeOH, λ_max, nm (ε)): 660 (4.59 × 10⁴), 603 (8.81 × 10³), 536 (9.43× 10³), 505 (9.02× 10³), 406 (9.04× 10⁴), 318 (2.42 × 10⁴).

Synthesis of compound 3:

Compound 19 (40.0 mg, 0.073 mmol) was dissolved in 1.5 mL HBr/AcOH, the reaction mixture was stirred for 1 hour at room temperature. The solvent was evaporated under high vacuum, the resulting crude was dried, and the entire crude was dissolved in dry CH₂Cl₂ (10 mL). A 40 mg portion of dry K₂CO₃ was added to this mixture, followed by addition of 2-hexanethiol (0.37 mL, 2.92 mmol). The resulting reaction mixture was stirred for 1 h at room temperature and then worked up. The crude mixture was purified by flash column chromatography (silica gel, 3% MeOH in CH₂Cl₂). Evaporation of the appropriate elutes gave the desired compound 3, 45% (21 mg) yield. ¹H NMR (400 MHz, CDCl₃, δ ppm): 9.50 (s, 1H), 9.48 (s, 1H), 8.53 (s, 1H), 5.75 (d, J = Hz, 1H), 5.64 (d, J = Hz, 1H), 5.27 (d, J = 20.0 Hz, 1H), 5.10 (d, J = Hz, 1H), 4.42-4.52 (qd, 1H), 4.22-4.52 (m, 1H), 3.85-3.95 (m, 1H), 3.66-3.73 (m, 2H), 3.67 (s, 3H), 3.61 (s, 3H), 3.38 (s, 3H), 3.25 (s, 3H), 2.64-2.76 (m, 1H), 2.50-2.63 (m, 1H), 2.24-2.37 (m, 2H), 1.81 (d, J = Hz, 4H), 1.70 (t, J = 7.6 Hz, 3H), 1.52-1.58 (m, 2H), 1.46 (d, J = 6.4 Hz, 3H), 1.24-1.36 (m, 3H), 0.86 (t, J = Hz, 3H), 0.47 (brs, 1H), −1.71 (s, 1H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 196.22, 173.50, 171.39, 160.09, 155.32, 150.87, 148.96, 144.99, 141.44, 137.85, 136.99, 136.09, 135.05, 134.13, 130.48, 128.22, 106.08, 104.02, 97.38, 92.98, 75.16/75.14, 61.36, 51.71, 51.67, 49.99, 48.06, 36.65, 30.96, 29.89, 27.97/27.95, 23.12, 22.85/22.84, 19.89, 19.49, 17.43, 14.04, 12.05, 11.23, 11.22. MS (ESI) m/z: 651.4 (M +H⁺). HRMS (ESI): calcd for C₄₀H₅₁N₄O₄ (M + H⁺) 651.3910; found, 651.3906. UV-vis (MeOH, λ_max, nm (ε)): 669 (4.45 × 10⁴), 612 (6.77 × 10³), 540 (6.77 × 10³), 513 (1.11 × 10⁴), 411 (7.82× 10⁴).

Synthesis of compound 4:

Compound 3 (30 mg, 0.0471 mmol) was dissolved in degassed THF (6 mL) and 3.5 mL degassed methanol and a degassed solution of LiOH ·H₂O (21.16 mg, 30 equiv) (3.5 mL, 1:1 v/v) was added. The entire reaction mixture was then stirred under nitrogen atmosphere at room temperature for 2 h, and the resulting mixture was diluted with dichloromethane (20 mL). The reaction mixture was washed three times with water. The organic layer was separated and dried over Na₂SO₄, and the solvent was removed under reduced pressure. The residue obtained was purified by flash column chromatography (silica gel, 5% methanol in CH₂Cl₂). Yield: 25.81 mg, 88%.¹H NMR (400 MHz, CDCl₃, δ ppm): 9.47 (s, 1H), 9.46/9.45 (s, 1H), 8.51 (s, 1H), (qd, J = 12.0 Hz, 2H), 5.25 (d, J = 20.0 Hz, 1H), 5.09 (d, J = 20.0 Hz, 1H), 4.41-4.51 (q, 1H), 4.23-4.32 (d, J = 8.0 Hz, 1H), 3.82-3.94 (m, 1H), 3.64-3.72 (m, 2H), 3.64 (s, 3H), 3.37 (s, 3H), 3.23 (s, 3H), 2.56-2.76 (m, 2H), 2.19-2.40 (m, 2H), 1.80 (d, J = 7.6 Hz, 3H), 1.68 (t, J = Hz, 3H), 1.54-1.66 (m, 2H), 1.37-1.54 (m, 2H), 1.45 (d, J = 6.0 Hz, 3H), 1.24-1.36 (m, 2H), 0.85 (td, 3H), −1.69 (brs, 1H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 196.41, 176.54, 171.37, 160.00, 155.40, 150.90, 148.99, 145.02, 141.48, 137.84, 137.04, 136.11, 135.04, 134.19, 130.40, 128.22, 106.06, 104.05, 97.38, 92.99, 75.17/75.15, 61.34, 51.56, 49.98, 48.01, 36.63, 30.53, 29.65, 27.96/27.95, 23.11, 22.84, 19.88, 19.49, 17.42, 14.04, 12.03, 11.22, 11.21. MS (ESI) m/z: 623.36 (M +H⁺). HRMS (ESI): calcd for C₃₈H₄₇N₄O₄ (M + H⁺) 623.2631; found, 623.2639. UV-vis (MeOH, λ_max, nm (ε)): 669 (4.45 × 10⁴), 612 (6.77 × 10³), 540 (6.77 × 10³), 513 (1.11 × 10⁴), 411 (7.82× 10⁴).

Synthesis of compound 5:

To a solution of compound (18) (150 mg) in CH₂Cl₂: MeOH (15:1 mL), sodium borohydride (20 mg) was added at 0 °C and the reaction mixture was stirred under Ar at room temperature for 20 min. The reaction was quenched with 2% AcOH in H₂O (50 mL), diluted with CH₂Cl₂ (150 mL) and the separated organic layer was washed with H₂O (3 x 100 mL). The organic layer was dried over Na₂SO₄, concentrated, and purified over a silica column using 4% acetone in CH₂Cl₂ as eluant to afford 122 mg (81%) of the product 19

Compound 19 (40.0 mg, 0.073 mmol) was dissolved in 1.5 mL HBr/AcOH, the reaction mixture was stirred for 1 hour at room temperature. The solvent was evaporated under high vacuum, the resulting crude was dried, and the entire crude was dissolved in dry CH₂Cl₂ (10 mL). A 40 mg portion of dry K₂CO₃ was added to this mixture, followed by addition of 1-hexanethiol (0.365 mL, 2.92 mmol). The resulting reaction mixture was stirred for 1 h at room temperature and then worked up. The crude mixture was purified by flash column chromatography (silica gel, 4% MeOH in CH₂Cl₂). Evaporation of the appropriate elutes gave the desired compound 5, 49% (22.86 mg) yield. ¹H NMR (400 MHz, CDCl₃, δ ppm): 9.48 (s, 1H), 9.46 (s, 1H), 8.53 (s, 1H), 5.69 (s, 2H), 5.27 (d, J = 19.6 Hz, 1H), 5.10 (d, J = 19.6 Hz, 1H), 4.44-4.57 (td, 1H), 4.26-4.33 (m, 1H), 3.81 (t, J = 6.4 Hz, 2H), 3.64-3.71 (m, 2H), 3.65 (s, 3H), 3.61 (s, 3H), 3.38 (s, 3H), 3.24 (s, 3H), 2.64-2.76 (m, 1H), 2.51-2.69 (m, 1H), 2.24-2.37 (m, 2H), 1.82 (d, J = 7.2 Hz, 3H),1.73-1.78 (m, 2H), 1.70 (t, J = 7.6 Hz, 3H), 1.40-1.50 (m, 2H), 1.23-1.34 (m, 4H), 0.83 (t, J = 7.2 Hz, 3H), 0.42 (brs, 1H), −1.72 (s, 1H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 196.19, 173.49, 171.36, 160.12, 155.28, 150.84, 148.95, 144.99, 141.33, 137.84, 136.83, 136.12, 134.70, 134.23, 130.49, 128.25, 106.08, 104.01, 97.29, 92.96, 70.83, 64.10, 51.70, 51.66, 49.98, 48.05, 31.71, 30.95, 29.98, 29.88, 26.06, 23.11, 22.60, 19.47, 17.41, 13.99, 12.03, 11.25, 11.22. MS (ESI) m/z: 651.4 (M +H⁺). HRMS (ESI): calcd for C₄₀H₅₁N₄O₄ (M + H⁺) 651.3910; found, 651.3906. UV-vis (MeOH, λ_max, nm (ε)): 669 (4.45 × 10⁴), 612 (6.77 × 10³), 540 (6.77 × 10³), 513 (1.11 × 10⁴), 411 (7.82× 10⁴).

Synthesis of compound 6:

Compound 5 (30 mg, 0.0449 mmol) was dissolved in degassed THF (6 mL) and 3.5 mL degassed methanol and a degassed solution of LiOH ·H₂O (21.16 mg, 30 equiv) (3.5 mL, 1:1 v/v) was added. The entire reaction mixture was then stirred under nitrogen atmosphere at room temperature for 2 h, and the resulting mixture was diluted with dichloromethane (20 mL). The reaction mixture was washed three times with water. The organic layer was separated and dried over Na₂SO₄, and the solvent was removed under reduced pressure. The residue obtained was purified by flash column chromatography (silica gel, 5% methanol in CH₂Cl₂). Yield: 26.10 mg, 89%. ¹H NMR (400 MHz, CDCl₃, δ ppm): 9.45 (s, 1H), 9.44 (s, 1H), 8.52 (s, 1H), 5.67 (s, 2H), 5.25 (d, J = 20.0 Hz, 1H), 5.09 (d, J = 20.0 Hz, 1H), 4.42-4.51 (qd, 1H), 4.56-4.32 (m, 1H), 3.79 (t, J = 6.8 Hz, 2H), 3.63-3.71 (m, 2H), 3.63 (s, 3H), 3.36 (s, 3H), 3.28 (s, 3H), 2.56-2.75 (m, 2H), 2.20-2.40 (m, 2H), 1.80 (d, J = 7.20 Hz, 3H), 1.74 (t, J = 7.6 Hz, 2H), 1.68 (t, J = 7.6 Hz, 3H), 1.38-1.49 (m, 2H), 1.22-1.38 (m, 4H), 0.81 (t, J = 6.8 Hz, 3H), −1.70 (brs, 1H). ³C NMR (100 MHz, CDCl₃, δ ppm): 196.43, 176.95, 171.36, 160.06, 155.38, 150.89, 148.99, 145.02, 141.39, 137.83, 136.90, 136.14, 134.69, 134.31, 130.39, 128.26, 106.07, 104.05, 97.29, 92.98, 70.82, 64.08, 51.56, 49,97, 48.00, 31.71, 30.61, 29.96, 29.63, 26.05, 23.11, 22.60, 19.47, 17.41, 13.98, 12.02, 11.24, 11.21. MS (ESI) m/z: 623.36 (M + H⁺). HRMS (ESI): calcd. for C₃₈H₄₇N₄O₄ (M + H⁺) 623.3610; found, 623.3591. UV-vis (MeOH, λmax, nm (ε)): 669 (4.45 × 104), 612 (6.77 × 103), 540 (6.77 × 103), 513 (1.11 × 104), 411 (7.82× 104).

Synthesis of compound 7:

A saturated methanol (8 mL) solution with Zn(OAC)₂.2H₂O (910.3 mg) was added to a dichloromethane (16 mL) solution of compound 20 (120 mg) and the reaction mixture was stirred for 45 min under reflux conditions. After completion of the reaction the reaction mixture was diluted with 30 mL water. The organic layer was separated and dried over anhydrous Na₂SO₄, and concentrated in vacuo to give corresponding Zink complex 21.

The Zink complex of chlorin (21) was dissolved in THF (10 mL). Trimethylamine (3 drops) along with palladium in activated carbon (15 mg) were added. The reaction mixture was stirred at room temperature under H₂ atmosphere for 24 h. It was then filtered and the filtrate was evaporated to afford a crude residue 22, 94% (147.8 mg) yield.

The crude residue 22 (60.0 mg, 0.0831 mmol) was dissolved in dry CH₂Cl₂ (12 mL), and HBr gas was bubbled into the solution for 5 min. The reaction mixture was stirred for 8 min at room temperature. The solvent was evaporated under high vacuum, the resulting crude was dried, and the entire crude was dissolved in dry CH₂Cl₂ (12 mL). A 60 mg portion of dry K₂CO₃ was added to this mixture, followed by addition of n-butanol (0.1 mL). The resulting reaction mixture was stirred for 1 h at room temperature and then worked up. The crude mixture was treated with 10 mL of TFA/DCM (1:1) for 1 h at room temperature after completion of the reaction the solvent was evaporated and adjust the pH 6.0 and worked up with DCM and water. The crude mixture was purified by flash column chromatography (silica gel, 4% acetone in DCM). Evaporation of the appropriate elutes gave the desired compound 7, 50% (27.08 mg) and compound 23, 15% (7.19 mg) yield. ¹H NMR (400 MHz, CDCl₃, δ ppm): 9.78 (s, 1H), 9.52 (s, 1H), 8.52 (s, 1H), 5.73 (t, J = 6.8 Hz, 1H), 5.25 (d, J = 20.0 Hz, 1H), 512 (d, J = 20.0 Hz, 1H), 4.50-4.54 (q, 1H), 4.28-4.35 (m, 1H), 3.67-3.76 (m, 2H), 3.68 (s, 3H), 3.59-3.66 (m, 2H), 3.61/3.61 (s, 3H), 3.88/3.88 (s, 3H), 3.27 (s, 3H), 2.64-2.76 (m, 1H), 2.50-2.64 (m, 2H), 2.20-2.43 (m, 3H), 1.80-1.85 (m, 3H), 1.72 (t, J = 7.6 Hz, 6H), 1.35-1.55 (m, 3H), 1.02 (t, J = 7.2 Hz, 3H), 0.83-0.90 (td, 3H), 0.47 (brs, 1H), −1.68/−1.69 (s, 1H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 196.18, 173.48, 171.38/171.37, 160.17/160.13, 155.24, 150.83, 149.03, 144.96, 141.41/141.37, 138.99/138.98, 137.75, 136.26, 136.10/136.02, 132.94/132.87, 130.42, 128.21, 105.92/105.91, 104.09, 98.19, 92.53, 77.22/77.02, 76.93/76.87, 69.39/69.38, 51.66/51.64, 50.07, 48.04, 41.12/41.07, 32.31/32.30, 30.90/30.88, 29.89, 29.70, 23.14/23.13, 19.71/19.69, 19.57/19.51, 17.44, 14.15/14.14, 13.95, 12.06, 11.28, 11.22/11.21, 1.03. MS (ESI) m/z: 651.4 (M + H⁺). HRMS (ESI): calcd for C₄₂H₅₅N₄O₄ (M + H⁺) 651.3905; found, 651.3883. UV-vis (MeOH, λ_max, nm (ε)): 660 (4.62 × 10⁴), 604 (8.76 × 10³), 536 (9.16 × 10³), 505 (8.86 × 10³), 407 (9.09× 10⁴), 318 (2.58× 10⁴).

Synthesis of compound 8:

Compound 7 (20 mg, 0.0449 mmol) was dissolved in degassed THF (4 mL) and 2.5 mL degassed methanol and a degassed solution of LiOH ·H₂O (26.1 mg, 30 equiv) (2.5 mL, 1:1 v/v) was added. The entire reaction mixture was then stirred under nitrogen atmosphere at room temperature for 2 h, and the resulting mixture was diluted with dichloromethane (20 mL). The reaction mixture was washed three times with water and adjusted pH 6.5. The organic layer was separated and dried over Na₂SO₄, and the solvent was removed under reduced pressure. The residue obtained was purified by flash column chromatography (silica gel, 4% methanol in CH₂Cl₂) to afforded pure compound 8. Yield: 17.8 mg, 91%. MS (ESI) m/z: 665.4 (M +H⁺). HRMS (ESI): calcd for C₄₀H₅₁N₄O₄ (M + H⁺) 665.4061; found, 665.4037. UV-vis (MeOH, λ_max, nm (ε)): 661 (4.6 × 10⁴), 604 (8.87 × 10³), 538 (8.87 × 10³), 505 (8.87 × 10³), 408 (8.92 × 10⁴), 320 (2.36× 10⁴).

Synthesis of compound 9:

The crude residue 22 (60.0 mg, 0.0832 mmol) was dissolved in dry CH₂Cl₂ (10 mL), and HBr gas was bubbled into the solution for 5 min. The reaction mixture was stirred for 8 min at room temperature. The solvent was evaporated under high vacuum, the resulting crude was dried, and the entire crude was dissolved in dry CH₂Cl₂ (12 mL). A 60 mg portion of dry K₂CO₃ was added to this mixture, followed by addition of n-hexanol (0.1 mL). The resulting reaction mixture was stirred for 1 h at room temperature and then worked up. The crude mixture was treated with 10 mL of TFA/DCM (1:1) for 1 h at room temperature after completion of the reaction the solvent was evaporated and adjust the pH 6.0 and worked up with DCM and water. The crude mixture was purified by flash column chromatography (silica gel, 4% acetone in DCM). Evaporation of the appropriate elutes gave the desired compound 9, 55% (31.07 mg) and compound 23, 12% (5.75 mg) yield. ¹H NMR (400 MHz, CDCl₃, δ ppm): 9.78 (s, 1H), 9.52 (s, 1H), 8.52 (s, 1H), 5.73 (t, J = 7.2 Hz, 1H), 5.27 (d, J = 20.0 Hz, 1H), 5.12 (d, J = 20.0 Hz, 1H), 4.45-4.53 (q, 1H), 4.27-4.39 (dd, 1H), 3.66-3.76 (m, 2H), 3.68 (s, 3H), 3.58-3.65 (m, 2H), 3.61/3.61 (s, 3H), 3.38 (s, 3H), 3.27 (s, 3H), 2.65-2.76 (m, 1H), 2.50-2.65 (m, 2H), 2.21-2.44 (m, 2H), 1.80-1.85 (m, 3H), 1.67-1.77 (t, J = 7.6 Hz, 6H), 1.30-1.55 (m, 3H), 1.18-1.26 (m, 4H), 1.02 (t, J = 7.2 Hz, 3H), 00.75-0.82 (t, J = 6.8 Hz, 3H), .48 (brs, 1H), −1.68 (s, 1H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 196.19, 173.48, 171.38, 160.16/160.13, 155.25, 150.84/150.83, 149.03, 144.97, 141.41/141.37, 138.99/138.97, 137.75, 136.26, 136.09/136.02, 132.95/132.88, 130.42, 128.21, 105.92/105.91, 104.08, 98.20, 92.53, 77.02, 76.90/76.85, 69.70, 51.66/51.64, 50.04, 48.04, 41.11/41.07, 31.71/31.71, 30.90/30.88, 30.19/30.17, 29.89, 29.70, 26.07/26.06, 23.14/23.13, 22.57, 19.71/19.70, 19.51, 17.43, 14.15/14.14, 13.95, 12.06, 11.31, 11.23/11.22. MS (ESI) m/z: 679.4 (M +H⁺). HRMS (ESI): calcd for C₄₀H₅₁N₄O₄ (M + H⁺) 679.4130; found, 679.4128. UV-vis (MeOH, λ_max, nm (ε)): 660 (4.55 × 10⁴), 603 (8.7 × 10³), 536 (8.85 × 10³), 504 603 (8.7 × 10³), 406 (8.85 × 10⁴), 320 (2.33 × 10⁴).

Synthesis of compound 10:

Compound 9 (20 mg, 0.03 mmol) was dissolved in degassed THF (4 mL) and 2.5 mL degassed methanol and a degassed solution of LiOH ·H₂O (21.16 mg, 30 equiv) (2.5 mL, 1:1 v/v) was added. The entire reaction mixture was then stirred under nitrogen atmosphere at room temperature for 2 h, and the resulting mixture was diluted with dichloromethane (20 mL). The reaction mixture was washed three times with water and adjusted pH 6.5. The organic layer was separated and dried over Na₂SO₄, and the solvent was removed under reduced pressure. The residue obtained was purified by flash column chromatography (silica gel, 4% methanol in CH₂Cl₂) to afforded pure compound 10. Yield: 17.8 mg, 91%. MS (ESI) m/z: 665.4 (M +H⁺). HRMS (ESI): calcd for C₄₀H₅₁N₄O₄ (M + H⁺) 665.4061; found, 665.4037. UV-vis (MeOH, λ_max, nm (ε)): 661 (4.55 × 10⁴), 605 (8.85 × 10³), 537 (8.85 × 10³), 505 (8.85 × 10³), 408 (8.8× 10⁴), 320 (2.36× 10⁴).

Synthesis of compound 11:

The compound 23 (20 mg) was dissolved in THF (5 mL). Trimethylamine (2 drops) along with palladium in activated carbon (6 mg) were added. The reaction mixture was stirred at room temperature under H₂ atmosphere for 24 h. It was then filtered and the filtrate was evaporated to afford a crude residue. The crude mixture was purified by flash column chromatography (silica gel, 3% acetone in DCM) to afforded pure compound (11), 95% (19.00 mg) yield.¹H NMR (400 MHz, CDCl₃, δ ppm): 9.44 (s, 1H), 9.17 (s, 1H), 8.44 (s, 1H), 6.15 (t, 1H), 5.25 (d, J = 20.0 Hz, 1H), 5.08 (d, J = 20.0 Hz, 1H), 4.42-4.49 (m, 1H), 4.24-4.30 (m, 1H), 3.78 (t, J = 7.6 Hz, 2H), 3.60-3.71 (m, 2H), 3.64 (s, 3H), 3.62 (s, 3H), 3.28 (s, 3H), 3.23 (s, 3H), 2.63-2.74 (m, 1H), 2.50-2.61 (m, 1H), 2.24-2.36 (m, 2H), 2.08-2.18 (m, 2H), 1.81 (d, J = 7.2 Hz, 3H), 1.69 (d, J = 8.0 Hz, 4H), 1.10 (t, J = 7.6 Hz, 3H), 0.62 (brs, 1H), −1.60 (s, 1H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 196.20, 173.51, 171.72, 159.83, 155.25, 150.44, 149.10, 145.03, 142.32, 140.29, 137.46, 137.44, 135.63, 131.66, 130.19, 127.83, 105.78, 104.15, 96.12, 92.3, 77.02, 51.65, 51.54, 50.06, 47.99, 34.51, 30.92, 29.87, 25.70, 23.05, 22.92, 19.46, 17.42, 14.12, 12.00, 11.21, 11.14. MS (ESI) m/z: 579.33 (M + H⁺). HRMS (ESI): calcd for C₃₆H₄₃N₄O₃ (M + H⁺) 579.3330; found, 579.3327. UV-vis (MeOH, λ_max, nm (ε)): 656 (4.40 × 10⁴), 600 (9.46 × 10³), 535 (9.46 × 10³), 504 (9.46 × 10³), 406 (9.5× 10⁴), 320 (2.55× 10⁴).

Compound 12:

Compound 11 (12 mg, 0.0449 mmol) was dissolved in degassed THF (3 mL) and 2 mLdegassed methanol and a solution of LiOH ·H₂O (26.1 mg, 30 equiv) in water (2 mL, 1:1 v/v) were added. The entire reaction mixture was then stirred under nitrogen atmosphere at room temperature for 2 h, and the resulting mixture was diluted with dichloromethane (20 mL). The reaction mixture was washed three times with water and adjusted pH 6.5. The organic layer was separated and dried over Na₂SO₄, and the solvent was removed under reduced pressure. The residue obtained was purified by flash column chromatography (silica gel, 4% methanol in CH₂Cl₂) to afforded pure compound 12. Yield: 10.44 mg, 89%. MS (ESI) m/z: 565.4 (M +H⁺). HRMS (ESI): calcd for C₄₀H₅₁N₄O₄ (M + H⁺) 565.3187; found, 565.3188. UV-vis (MeOH, λ_max, nm (ε)): 657 (4.45 × 10⁴), 602 (9.56 × 10³), 537 (9.56 × 10³), 504 (9.36 × 10³), 407 (9.6× 10⁴), 319 (2.58× 10⁴).

Synthesis of compound 13:

The compound 24 (20 mg) was dissolved in THF (4 mL). Trimethylamine (1 drops) along with palladium in activated carbon (6 mg) were added. The reaction mixture was stirred at room temperature under H₂ atmosphere for 24 h. It was then filtered and the filtrate was evaporated to afford a crude residue. The crude mixture was purified by flash column chromatography (silica gel, 3% acetone in DCM) to afforded pure compound (13), 95% (19.0 mg).¹H NMR (400 MHz, CDCl₃, δ ppm): 9.46 (s, 1H), 9.18 (s, 1H), 8.44 (s, 1H), 5.24 (d, J = 20.0 Hz, 1H), 5.08 (d, J = 20.0 Hz, 1H), 4.41-4.50 (qd, 1H), 4.23-4.50 (m, 1H), 3.78 (t, J = 7.2 Hz, 2H), 3.64-3.73 (m, 2H), 3.65 (s, 3H), 3.61 (s, 3H), 3.27 (s, 3H), 3.24 (s, 3H), 2.62-2.74 (m, 1H), 2.49-2.61 (m, 1H), 2.22-2.36 (m, 2H), 2.08-2.19 (m, 2H), 1.81 (d, J = 6.8 Hz, 3H), 1.70 (t, J = 7.6 Hz, 3H), 1.60-1.7 (m, 2H), 1.43-1.53 (m, 2H), 1.23-1.40 (m, 6H), 0.87 (t, J = 6.8 Hz, 3H), 0.64 (brs, 1H), −1.59 (s, 1H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 196.21, 173.51, 171.73, 159.84, 155.27, 150.45, 149.11, 145.03, 142.33, 140.36, 137.47, 137.46, 135.64, 131.65, 130.20, 127.85, 105.78, 104.17, 96.13, 92.38, 51.65, 51.55, 50.07, 48.00, 32.35, 31.92, 30.93, 29.87, 29.60, 29.31, 25.99, 23.05, 22.67, 19.47, 17.43, 14.09, 12.01, 11.22, 11.16. MS (ESI) m/z: 635.4 (M +H⁺). HRMS (ESI): calcd for C₄₀H₅₁N₄O₃ (M + H⁺) 635.3956; found, 635.3972. UV-vis (MeOH, λ_max, nm (ε)): 656 (4.38 × 10⁴), 600 (9.41 × 10³), 535 (9.41 × 10³), 504 (9.21 × 10³), 406 (9.45× 10⁴), 320 (2.53× 10⁴).

Synthesis of compound 14:

Compound 13 (15 mg, 0.024 mmol) was dissolved in degassed THF (3 mL) and 2.0 mL degassed methanol and a degassed solution of LiOH ·H₂O (26.1 mg, 30 equiv) (2.0 mL, 1:1 v/v) was added. The entire reaction mixture was then stirred under nitrogen atmosphere at room temperature for 2 h, and the resulting mixture was diluted with dichloromethane (20 mL). The reaction mixture was washed three times with water and adjusted pH 6.5. The organic layer was separated and dried over Na₂SO₄, and the solvent was removed under reduced pressure. The residue obtained was purified by flash column chromatography (silica gel, 4% methanol in CH₂Cl₂) to afforded pure compound 8. Yield: 12.6 mg, 86%. MS (ESI) m/z: 621.4 (M + H⁺). HRMS (ESI): calcd for C₃₉H₄₉N₄O₃ (M + H⁺) 621.4032; found, 621.4037. UV-vis (MeOH, λ_max, nm (ε)): 657 (4.38 × 10⁴), 602 (9.41 × 10³), 537 (9.41 × 10³), 504 (9.21 × 10³), 407 (9.45× 10⁴), 319 (2.53× 10⁴).

Synthesis of compound 15:

Compound 17 (80.0 mg, 0.04 mmol) was dissolved in 2 mL HBr/AcOH, the reaction mixture was stirred for 1 hour at room temperature. The solvent was evaporated under high vacuum, the resulting crude was dried, and the entire crude was dissolved in dry CH₂Cl₂ (10 mL). A 80 mg portion of dry K₂CO₃ was added to this mixture, followed by addition of 1-hexanethiol (0.35 mL). The resulting reaction mixture was stirred for 1 h at room temperature and then worked up. The crude mixture was purified by flash column chromatography (silica gel, 40% ethyl acetate in hexane). Evaporation of the appropriate elutes gave the desired compound 15, 61% (59.32 mg). ¹H NMR (400 MHz, CDCl₃, δ ppm): 9.77 (s, 1H), 9.49 (s, 1H), 8.56 (s, 1H), 5.54-5.62 (q, 1H), 5.28 (d, J = 20.0 Hz, 1H), 5.13 (d, J = 20.0 Hz, 1H), 4.46-4.55 (m, 1H), 4.27-4.35 (m, 1H), 3.66-3.76 (m, 2H), 3.66 (s, 3H), 3.64 (s, 3H), 3.47 (s, 3H), 3.31 (s, 3H), 2.66-2.76 (m, 1H), 2.53-2.66 (m, 2H), 2.40-2.50 (m, 1H), 2.24-2.37 (m, 5H), 1.84 (d, J = 7.6 Hz, 3H), 1.72 (t, J = 8.0 Hz, 3H), 1.55-1.68 (m, 2H), 1.18-1.35 (m, 2H), 1.05-1.18 (m, 4H), 0.72 (t, J = 7.2 Hz, 3H), 0.46 (brs, 1H), −1.70 (s, 1H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 196.11, 173.46, 171.38, 160.28/160.27, 154.88/154.87, 150.87, 149.01, 144.93, 141.28, 139.73/139.71, 137.78, 136.18, 135.80/135.79, 132.23, 130.44, 128.29, 105.97, 104.08, 97.38, 92.66, 51.64/51.63, 50.04, 48.01, 36.29/36.27, 32.45/32.44, 31.31/31.30, 30.89, 29.86, 29.60/29.57, 28.50/28.50, 23.13/23.12, 22.37/22.37, 19.47, 17.42, 13.85, 12.01, 11.71, 11.28. MS (ESI) m/z: 667.36 (M + H⁺). HRMS (ESI): calcd for C₃₈H₅₂N₄O₃ (M + H⁺) 667.3652; found, 667.3655. UV-vis (MeOH, λmax, nm (ε)): 662 (4.6 × 10⁴), 607 (8.5 × 10³), 538 (8.5 × 10³), 507 (8.65 × 10³), 408 (8.64× 10⁴), 319 (2.25 × 10⁴).

Synthesis of compound 16:

Compound 15 (30 mg, 0.0449 mmol) was dissolved in degassed THF (10 mL). Degassed methanol and a solution of LiOH ·H₂O (56.52 mg, 30 equiv.) in water (10 mL, 1:1 v/v) were added. The entire reaction mixture was then stirred under nitrogen atmosphere at room temperature for 2 h, and the resulting mixture was diluted with dichloromethane (20 mL). The reaction mixture was washed three times with water. The organic layer was separated and dried over Na₂SO₄, and the solvent was removed under reduced pressure. The residue obtained was purified by flash column chromatography (silica gel, 5% methanol in CH₂Cl₂). Yield: 25.84 mg, 88%. ¹H NMR (400 MHz, CDCl₃, δ ppm): 9.62/9.58 (s, 1H), 8.95/8.94 (s, 1H), 8.54 (s, 1H), 5.45-5.57 (m, 1H), 5.05 (d, J = 19.6 Hz, 1H), 4.87 (d, J = 19.6 Hz, 1H), 4.34-4.48 (m, 1H), 4.03-4.18 (m, 1H), 3.38-3.47 (m, 2H), 3.39/3.38 (s, 3H), 3.22 (s, 3H), 3.12 (s, 3H), 2.32-2.62 (m, 5H), 2.22 (d, J = 7.2 Hz, 5H), 1.90-2.04 (m, 1H), 1.75 (d, J = 6.8 Hz, 3H), 1.51 (t, J = 7.2 Hz, 3H), 1.06-1.30 (m, 5H), 0.97 (m, 3H), 0.50-0.61 (m, 3H). ³C NMR (100 MHz, CDCl₃, δ ppm): 196.44, 177.20, 171.90, 160.53, 154.38, 149.84, 147.97/147.93, 144.06, 140.57, 139.06, 136.18, 135.13, 135.02, 131.85, 128.39, 126.64, 104.74, 102.67, 96.24/96.05, 92.09, 50.98, 49.22, 35.34/35.29, 31.79, 31.35, 30.30, 29.79, 28.66, 27.41, 21.73/21.63, 21.54, 21.33, 18.05, 15.83, 12.29, 9.97, 9.83, 9.48. . MS (ESI) m/z: 653.3 (M + H⁺). HRMS (ESI): calcd for C₃₇H₅₀N₄O₃ (M + H+) 653.3496; found, 653.3501. UV-vis (MeOH, λmax, nm (ε)): 662 (4.5 × 10⁴), 607 (8.3 × 10³), 538 (8.3 × 10³), 507 (8.46 × 10³), 408 (8.45× 10⁴), 319 (2.2 × 10⁴).

Synthesis of compound 20:

The aldehyde 18 (200 mg, 0.3632 mmol) was dissolved in THF (10 mL). Activated Zn (118.74 mg, 1.816 mmol) and allyl bromide (0.094 mL, 1.089 mmol) are added at 0 °C and stirred for 10 min. To this saturated NH₄Cl solution (5 mL) was added drop wise at 0 °C and the solution was stirred for 3 h at ambient temperature. After completion of the reaction mixture was filtered and extracted with DCM (2 x 15 mL). The organic layer was separated, dried over anhydrous Na₂SO₄, and concentrated in vacuo. The residue on purification by column chromatography (DCM/MeOH, 95:5) afforded pure compound (20), 73% (153.4 mg) yield. ¹H NMR (400 MHz, CDCl₃, δ ppm): 9.62/9.59 (s, 1H), 9.37/9.59 (s, 1H), 8.49/8.47 (s, 1H), 6.17 (t, J = 7.2 Hz, 1H), 5.92-6.60 (m, 1H), 5.28 (d, J = 17.2 Hz, 1H), 5.09 (d, J = 17.2 Hz, 1H), 4.48-5.32 (m, 2H), 4.38-4.46 (td, 1H), 4.16-4.24 (td, 1H), 3.60-3.69 (m, 2H), 3.63/3.62 (s, 3H), 3.58 (s, 3H), 3.37/3.66 (s, 3H), 3.24-3.34 (m, 1H), 3.22/3.21 (s, 3H), 3.09-3.19 (m, 1H), 2.75-2.93 (d, J = 19.20 Hz, 1H), 2.49-2.67 (m, 2H), 2.12-2.35 (m, 2H), 1.76 (t, J = 7.6 Hz, 3H), 1.67 (t, J = 8.0 Hz, 3H), 1.23 (s, 1H), −1.89/−1.92 (s, 1H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 196.20, 173.50/173.48, 171.23/171.20, 160.22/160.17, 154.89, 150.71, 148.88/148.84, 144.88/144.86, 141.10/141.08, 138.95/138.90, 137.70/137.68, 136.16/136.15, 135.35/135.27, 134.42/134.41, 132.17/132.03, 130.34/130.30, 128.29/128.26, 118.67, 105.91/105.85, 103.95/103.94, 98.21/98.05, 92.70, 69.18/69.13, 51.69/51.68, 51.64/51.63, 49.98/49.96, 47.98/47.96, 44.26/44.19, 30.99/30.94, 29.85/29.82, 23.09/23.06, 19.45, 17.40, 11.98, 11.53/11.49, 11.25/11.24. MS (ESI) m/z: 693.34 (M + H⁺).

Synthesis of compound 24:

Heptyl triphenyl phosphonium bromide (45 mg, 0.109 mmol) was dissolved in 5 mL dry THF at 0 oC under Argon atmosphere added 0.05 mL n-BuLi stirred for 5 mins (reaction mixture color was turned to Yellow) at 0°C and pyro- formal 18 (40 mg, 0.0726 mmol) in THF was added. The resulting mixture was stirred for 6 h at room temperature. After completion of the reaction the reaction mixture was quenched with saturated NH₄Cl solution and worked up with DCM/H₂O. The organic layer was separated, dried over anhydrous Na₂SO₄, and concentrated in vacuo. The residue on purification by column chromatography (DCM/Acetone, 95:5) afforded pure compound (24). 68% (31.26 mg) yield. 1H NMR (400 MHz, CDCl3, δ ppm): 9.47/9.46 (s, 1H), 9.33/9.17 (s, 1H), 8.50 (s, 1H), 7.34-7.63 (d, J = Hz, 1H), 6.45-6.74 (dt, J = Hz, 1H), 5.25 (d, J = Hz, 1H), 5.08 (d, J = Hz, 1H), 4.42-4.51 (qd, 1H), 4.24-4.32 (m, 1H), 3.64-3.72 (m, 2H), 3.66 (s, 3H), 3.61 (s, 3H), 3.35/3.24 (s, 3H), 3.23/3.18 (s, 3H), 3.21 (s, 1H), 2.64-2.75 (m, 2H), 2.50-2.61 (m, 1H), 2.22-2.37 (m, 3H), 1.83 (d, J = 7.6 Hz, 3H), 1.69 (t, J = 7.6 Hz, 3H), 1.40-1.58 (m, 4H), 1.05-1.25 (m, 3H), 0.67-1.03 (t, J = 6.8 Hz, 3H), 0.57 (brs, 1H)-1.59/−1.63 (s, 1H). 13C NMR (100 MHz, CDCl3, δ ppm): 196.22/196.21, 173.51/173.50, 171.55/171.50, 160.12/159.95, 155.38/155.31, 150.65/150.61, 149.09/149.05, 145.02/144.97, 142.17/141.99, 140.76, 139.16/137.70, 137.01/136.72, 136.90/136.50, 135.92/132.89, 132.21/130.61, 130.38/130.35, 128.10/128.07, 121.73/120.75, 105.89/105.82, 104.13/104.11, 97.91/97.26, 92.74/92.71, 51.66/51.65, 50.02/50.00, 48.04, 34.52, 31.91/31.58, 30.95/30.93, 29.91/29.89, 29.70/29.38, 29.13/28.90, 23.12/23.10, 22.78/22.48, 19.49, 17.43/17.41, 14.20/13.89, 12.09/12.04, 11.23/11.21. MS (ESI) m/z: 619.38 (M +H⁺).

Primary cultures of tumor cells and xenografts:

Lung and Head and neck tumor tissue were obtained from Tissue Procurement Services at RPCCC under the terms of IRB-approved protocol. Each specimen was labeled with laboratory code “L-#” or “HN-#”, respectively . Tissue was dissociation by treatment with trypsin and cell preparations were plated unto bovine collagen-1 matrix (PureCol, Advanced BioMatrix, Carlsbad, CA) for selection of proliferating epithelial (termed T-EC) and stromal fibroblasts (termed T-Fb) as described^10,16. Reconstituted co-culture for evaluating cell type-specific PS retention were generated by the addition of CFSE-stained fibroblasts to 24 h-old cultures of T-EC and maintained these for additional 3 to 7 days until formation of confluent cultures with segregated tumor cell clusters.

Patient-derived tumor tissue pieces were implanted s.c. on the dorsal side of the left hindleg of SCID mice (C.B-lgh-lb/lcrTac-PrMscid/Ros; Department of Laboratory Animal Resources, RPCCC). Establish xenografts were maintained by serial transplantation. Animal experiments were approved by IACUC protocol. Tongue tumor xenograft-derived cells labeled as HNT1 and lung tumor xenograft labeled as TEC-1 (squamous carcinoma) and TEC-21 (adenocarcinoma) were placed into tissue culture using DMEM containing 10% FBS. These cultures were used as references for evaluating PS-binding specificity determined for primary tumor cell cultures.

PS treatment and photoreaction

Binding of PS to surfaces of cultured cells was measured by incubating cell cultures with serum-free medium containing 1600 nM PS for 30 min on ice. Diffusion of surface-bound PS into cells was evaluated by raising the temperature of the cell cultures to 37°C for 30 min. Uptake and intracellular accumulation of PS was determined by incubating cells in RPMI containing 10% FBS and 1600 nM PS for periods between 4 to 24h. Retention was assessed by continued culturing PS-treated cells in PS-free RPMI medium containing 10% FBS for up to 120 h (termed “Chase”). To test uptake of PS in vivo, HPPH or PS-8 were diluted in Tween80-glucose-water and injected intravenously into xenograft-bearing SCID mice (3 μmole per Kg). After 24 h, the systemic distribution of PS was verified by imaging of the animals (IVIS, PerkinElmer) prior to euthanasia and dissection of organs. Part of liver and tumor were embedded into OCT, frozen on liquid nitrogen and subjected to 10 μm cryosection (Leica cryotome). Binding of PS to tumor cryosection was determined after fixation of the section for 4h in buffered 10% formalin by incubating in PBS containing 10% FBS and 1.6 μM HPPH or PS-8 for 24 h at 37°C, followed by chase for 24 h in PS-free PBS containing 10% FBS (Tracy et al. in preparation).

Cell- or tissue section-associated PSs were imaged with a Zeiss inverted fluorescence microscope (Axiovert 200) using a Q-imaging camera. All images were recording in monochrome format (16 bits per channel, 2.9 Mb per frame). Images were changed to 600 dpi resolution and unaltered in size integrated into composite figures. For demonstrating greater details of the cellular staining patterns, 5% sections of the microscopic images were integrated into the composite Figures 2, 5A, and 8. For illustrating cell-specific association of PSs in co-culture models, the monochrome images were colorized in Photoshop. The fluorescence signals of full frame monochrome images were quantitated using ImageQuant program and expressed in arbitrary units (net pixel intensity per image area unit). Tumor tissue sections were stained with eosin and hematoxylin after fluorescence imaging and photographed using a Nikon Eclipse E200 microscope.

PS-treated cell cultures were exposed to 665 nm diode laser light (6 mW/cm²) for 9 min within a 37°C incubator. The cells were immediately extracted with RIPA buffer. Aliquots of lysates (10 μg protein) were analyzed by western blotting for STAT3 crosslinking and level of EGFR and phosphorylated p38 MAPK as previously describe¹⁰. Lethal response to PDT was determined in cultures 24 h after therapeutic light treatment by release of the cells by trypsin and counting viable cells using trypan-blue exclusion. Survival was expressed relative to control cultures not incubated with PSs (defined as 100% survival).

ABBREVIATIONS

PDT

photodynamic therapy

PS

photosensitizer

HPPH,

3-(1’-hexyloxy)ethyl-3-devinyl pyropheophorbide-a

HNT

head & neck tumor

TEC

tumor epithelial cells