Cell-specific Retention and Action of Pheophorbide-based Photosensitizers in Human Lung Cancer Cells

Abstract

This study determined in primary cultures of human lung cancer cells the cell specificity of chlorin-based photosensitizers. Epithelial cells (ECs) preferentially retained 3-[1-hexyloxyethyl]-2-devinylpyropheophorbide-a (HPPH) and related structural variants. Tumor-associated fibroblasts (Fb) differ from EC by a higher efflux rate of HPPH. Immunoblot analyses indicated dimerization of STAT3 as a reliable biomarker of the photoreaction. Compared to mitochondria/ER-localized photoreaction by HPPH, the photoreaction by lysosomally targeted HPPH-lactose showed a trend toward lower STAT3 cross-linking. Lethal consequence of the photoreaction differed between EC and Fb with the latter cells being more resistant. A survey of lung tumor cases indicated a large quantitative range by which EC retains HPPH. The specificity of HPPH retention defined in vitro could be confirmed in vivo in selected cases grown as xenografts. HPPH retention as a function of the tetrapyrrole structure was evaluated by altering side groups on the porphyrin macrocycle. The presence or absence of a carboxylic acid at position 17² proved to be critical. A benzyl group at position 20 enhanced retention in a subset of cancer cells with low HPPH binding. This study indicated experimental tools that are potentially effective in defining the photosensitizer preference and application for individual patient's cancer lesions.

Figure 1.

EC-specific retention of HPPH determined in reconstituted co-cultures of TEC-1 and CSFE-labeled lung (L364) tumor stromal cells. (A and B), Co-cultures were incubated in serum-free RPMI containing 3.2 μM HPPH (A) or HPPH-Lac (B) at 37°C for 30 min and then followed by a chase in fresh RPMI-10%FBS for 4 and 24h. The culture morphology was assessed by phase microscopy followed by imaging CSFE and PS fluorescence. (C) Comparison of cellular retention of HPPH, Foscan and Photofrin II. Co-cultures of TEC-1 and CSFE-stained stromal cells were first incubated for 4 h at 37°C in RPMI-10% FBS containing either 3.2 μM HPPH, 3.2 μM Foscan or 10 μg/ml Photofrin-II followed by a 24 h chase in PS-free medium. The culture and cell-associated PS were imaged by phase and fluorescent microscopy.

Figure 2.

PS-dose dependent photoreactions in primary lung ECs. Confluent monolayers of N-EC and T-EC (L374, passage 3) were treated with increasing concentrations of HPPH or HPPH-Lac for 24 h in RPMI containing 10% FBS followed by a chase for 24 h in PS-free medium. (A) The cell-associated PS fluorescence was determined by microscopy. Only the fluorescent images of T-EC cells at 400X treated with the highest PS concentration are shown to demonstrate the distinct subcellular distribution. (B) The levels of fluorescence in each culture were quantitated and expressed in arbitrary fluorescence units (FU). After 9 min light treatment (3 J/cm²) the products of the PS-mediated photoreaction were determined by immunoblot analyses including the covalent crosslinking of STAT3, the loss of EGFR and the activation of p38 kinase. (C) In parallel cultures, the percentage of PDT-surviving ECs were determined after additional 24 h-incubation of the light-treated cells in RPMI containing10% FBS.

Figure 3.

Compilation of data derived from individual cultures of N-EC (A), T-EC (B) and TEC-1–2 (C) correlating percentage of STAT3 crosslinking with cellular level of HPPH or HPPH-Lac (FU). The treatments of the cultures were the same as applied in Fig. 2. Each circle represents the result of individual cultures derived from a total of 54 separate lung samples.

Figure 4.

Compilation of data derived from individual cultures of N-EC (A), T-EC (B), TEC-1–2 (C), and T-Fb (D) correlating cellular survival with the percentage of STAT3 crosslinking. Duplicate cell cultures were subjected to incubation with HPPH and HPPH-Lac and subsequent light treatment as applied in Fig. 3, but in each set, one of the duplicate cultures was incubated for additional 24 h in RPMI containing 10% FBS for determining the percentage of surviving cells. Each circle represents the result of individual duplicate cultures.

Figure 5.

HPPH dose-dependent and cell type-specific PDT response. A co-culture of TEC-1–2 (passage 37) and CSFE-labeled T-Fb (L323) was incubated for 24 h with 800nM HPPH in DMEM-10% FBS and then chased for 24 h in HPPH-free medium. Cellular level of retained HPPH was determined by fluorescent microscopy prior to treatment with therapeutic light (Day 0). The culture was imaged daily for 3 days post PDT to record the reorganization of CSFE-labeled stromal cells and fate of TEC-1–2 cell clusters. HPPH is stably retained by photo-damaged cells and is detectable by fluorescence present in aggregates of killed TEC-1–2 cells. Surviving TEC-1–2 clusters are recognized by the low to non-detectable HPPH fluorescence and physical exclusion of stromal cells from the collagen-support matrix.

Figure 6.

Specificity of pheophorbide retention by human lung tumor cells. Primary cultures of lung T-ECs were grown to confluence in KSFM. A standard PS treatment was applied for every cell preparation: Uptake of the PSs indicated at the right was carried by incubating the cultures for 24 h in RPMI containing10%FBS and 1600 nM PSs followed by 24 h chase in PS-free medium. The cell-associated PS fluorescence was imaged by microscopy at 100X magnification using 2 sec exposures. All images were identically processed. Examples of PS retention patterns are shown. For comparison, the PS retention pattern of N-Fb is included at the left and the two most distinct patterns detected in cancer cells derived from PDX tissue (TEC-1–2 and TEC-21, still including murine stromal cells) are shown on the right.

Figure 7.

PS-specific photoreaction in lung T-EC. Confluent cultures of TEC-1–2 cells (passage 12) were treated for 24 h in RPMI containing 10% FBS and 1600 nM of the PSs indicated at the top, followed by 24 h chase in PS-free medium. (A) The cellular level of PS was determined by imaging and expressed as AU. The cultures were subjected to PDT and immediately analyzed for the level of crosslinked STAT3, EGFR and phosphorylated p38 MAPK. (B) Duplicate cultures were identically treated and PDT-surviving cells determined after 24 h culture.

Figure 8.

Uptake and retention of HPPH by organs in TEC-1 xenograft-bearing SCID mice. Aliquots of TEC-1 xenograft tissue were implanted into two separate groups of 12 SCID mice and grown for 4 weeks to a size ranging from 8 to 10 mm diameter. HPPH (3 μmole/kg) was injected into the tail vein. After 6, 24 and 48 h, sets of four mice were euthanized and the indicated organs collected, embedded into OCT and 5-μm cryosections prepared. The sections were imaged by fluorescent microscope at 100X magnification with capture of the images by 3 sec exposure. (A) The fluorescent images of a representative set of the organ sections are reproduced. (B) The net HPPH fluorescence densities were determined for each organ section (using 4 to 6 separate areas) and expressed in arbitrary units (AU). The average AU values for HPPH fluorescence were used as measure of HPPH concentration in each organ of individual mice. Mean and S.D. of the AU values for both sets of animals (N=8) are shown.

Figure 9.

Comparison of HPPH retention by lung tumor cells grown in vitro and in vivo. Duplicate sets of SCID mice bearing either a TEC-1 or TEC-21 xenograft were generated. (A) From one set of animals, the tumor was collected and processed for tissue culture of the cancer cells along with residual murine stromal cells using DMEM-10% FBS as growth medium (passage 0). After 3 weeks, the cultures contained clusters of tumor cells within a background monolayer of stroma cells. These cultures were treated for 24 h with medium containing 3 μM HPPH followed by 24 h chase period. The cellular retention of HPPH was determined by fluorescent microscopy using identical imaging setting. (B) A second set of animals was injected with 3 umole/kg HPPH and after 24 h, the tumor and liver were collected. The relative level of HPPH in these organs was determined by fluorescent microscopy of 5-μm cryosections. The images of HPPH fluorescence for each section were processed using identical conditions. (C) A cryosection of TEC-21 tumor was stained with hematoxylin and eosin (H&E) and compared to the HPPH fluorescent image (red colorized) of the same area on an adjacent section. The white bar shown at the top of the right panel indicates 250 μm.