2.1.

Two-Photon Laser Scanning Microspectrofluorometry (TPLSMF)

Two-photon confocal laser scanning microscopy images were obtained using a commercial CLSM, system model MRC-1024 (Bio-Rad, Richmond, CA), which was attached to an upright microscope (Nikon, model Eclipse E800). A water immersion objective lens (Nikon, Fluor-60X, NA=1.0 ) was used for cell imaging. The system was adapted for two-photon excitation at 820 nm by using dichroic mirrors and filters. A broad band-pass filter (460–600 nm) was used as an emission filter for imaging. The excitation light was provided by a Ti–sapphire laser (Spectra-Physics, Tsunami) pumped by a frequency-doubled diode pumped solid state laser (Spectra-Physics, Millennia). The system is operated at around 90 fs pulse width and 82 MHz repetition rate. The average power at the sample was adjusted to ∼15 mW to avoid photodamage to the cells.

For localized spectrofluorometry, the fluorescence signal was collected from the side port of the confocal microscope using a multimode optical fiber of core diameter 1 mm, and was delivered to a spectrometer (Holospec, Keiser Inc.) equipped with a cooled charge coupled device (CCD) camera (Princeton Instruments) as a detector. The fluorescence signal was directed to the fiber by a dichroic mirror inserted into the sideport of the microscope, before passing through any filters inside the scanhead and without descanning. The system can be easily switched between TPLSM and TPLSMF by moving the dichroic mirror in and out of the beam path. The inherent optical sectioning capability of the two-photon process allowed us to do this collection without confocal aperture. An absorption filter in the infrared range was used inside the spectrograph to cut off the excitation line from the laser. The acquired spectra were corrected for the CCD response as well as for the transmission of the filters and mirrors in the spectral range of interest. Throughout the study, we normalized the obtained spectra with respect to the maximum intensity of the fluorescence spectra, as we were always using the ratio of fluorescence intensity peaks from two different fluorescent tags, rather than the individual intensity of one fluorescent tag.

2.2.

Synthesis of Two-Photon Probe TPR and Conjugation of the Probes to AN-152 and [D-Lys ⁶ ]LH-RH

The chemotherapeutic drug AN-152 and the two-photon probe C625 were synthesized as described previously.^{10 6} [D-Lys ⁶ ]LH-RH (Glp-His-Trp-Ser-Tyr-D-Lys-Leu-Arg-Pro-Gly-NH ₂ ) was obtained from California Peptide Research, Inc. All other reagents were purchased from Aldrich and were used without further purification. The synthesis of the custom designed TPR is illustrated in Figure 1. The electron donor part was synthesized by alkylation of N-methyl aniline with 6-bromocaproic acid ethyl ester. After the Vilsmeier reaction, the carboxylate protective group was removed by hydrolysis. The pyridinium derivative 7 was subsequently linked to amino-substituted aldehyde 6 through base catalyzed aldol condensation. The detailed strategies for the synthesis of this type of pyridinium salt have also been described in our previous work.¹² The final product, designated TPR, was characterized by proton nuclear magnetic resonance (NMR) and MS. [ ¹ H NMR in CDCl ₃ (500 MHz): δ1.29 ( m, 2H), δ1.55 (m, 4H), δ2.17 ( t, 2H), δ2.93 ( s, 3H), δ3.34 ( t, 2H), δ4.02 (s, 3H), δ6.70 ( d, 2H), δ6.77 ( d, H), δ7.40 ( d, 2H), δ7.46 (d, 1H), δ7.66 ( d, 2H), δ 8.21 ( d, 2H); fast bombardment MS: calculated for C ₂₁ H ₂₇ N ₂ O ₂ 339.2 found 339.2]. The preparation of the conjugate of C625 and AN-152 (AN-152:C625 or AC) and the conjugate of C625 and [D-Lys ⁶ ]LH-RH (LH-RH:C625 or LH625) was described in our previous work.¹⁰ The conjugate of TPR and [D-Lys ⁶ ]LH-RH (LH-RH:TPR or LHTPR) was prepared with the same method approach. Purification of the labeled peptide products was carried out by Hewlett Packard Series 1050 HPLC equipped with a C ₁₈ reverse phase analytical column (Phenomenex, Jupiter). All final products were characterized by matrix assisted laser desorption ionization mass spectroscopy.

Figure 1

Synthetic pathway of the two-photon probe TPR.

2.3.

Cell Culture

MCF-7, human breast cancer cells (American Type Culture Collection) were maintained at 37 °C in a humidified environment with 5 CO ₂ . Cells were cultured in improved minimum essential medium with 10 fetal bovine serum. Cells were subcultured in 30 mm petri dishes for 48–72 h (80 confluent). prior to incubation with labeled reagents.

3.1.

Results

A recently developed two-photon probe, C625, can be successfully used in real-time optical tracking using TPLSM.^{10 11} The design of a new dye (TPR) with red color provides a nice probe with a different spectral window, even though it shows a lower two-photon fluorescence efficiency compared to C625. The absorption and emission spectra of TPR and C625 are shown in Figure 2. The excitation wavelength of 820 nm was selected to obtain reasonable excitation efficiency with both C625 and TPR. We used these probes to study the entry and transport of AN-152 and its carrier peptide in living cells in vitro.

Figure 2

Absorption and emission spectra of the probes TPR and C625 (in DMSO).

In the control experiments, the cells treated with [D-Lys ⁶ ]LH-RH (but without addition of labeled molecules) were scanned for 1 h to check for autofluorescence and no significant emission was detected. To observe the entry ability of TPR itself, we directly added TPR (10 μM) and AC (0.6 μM) to the medium of cells. As AC can enter MCF-7 cells and even stain the nucleus within 1 h,^{11 10} its high fluorescence efficiency allows for an easier identification of the nuclei inside the cells from fluorescence images. Figure 3(a) shows the localized spectra inside the nucleus, cytoplasm, and outside the cells, 90 min after staining. The white and black spots in the center of each image in Figure 3(b) correspond approximately to the size of the focus point when the spectra were being taken. From Figure 3, it can be seen that the localized spectra inside the cells (nucleus and cytoplasm) are dramatically different from the one obtained outside the cell.

Figure 3

(a) Localized spectra from the nucleus, cytoplasm, and medium from an AC&TPR treated MCF-7 breast cancer cell. TPR (10 μM) and AC (0.6 μM) were added to the medium of cells and localized spectra were acquired 90 min later. All the acquired spectra were normalized with respect to the maximum fluorescence intensity. (b) Two-photon confocal microscope images showing the positions where the spectra were acquired.

To investigate the effect of the local environment on the fluorescence efficiency of AC and TPR, we recorded their fluorescence spectra in water and after adding membrane model dimyristoyl-L-α-phosphatidylcholine (DMPC). Figure 4(a) shows the emission spectra of AC in water before and after adding DMPC. The dramatic improvement of fluorescence from AC after it is associated with the model membrane can clearly be seen. In contrast, the emission of TPR changed little after adding DMPC to the solution [Figure 4(b)].

Figure 4

Emission spectra of AC (a) and LHTPR (b) in water and while bound to lipid vesicles [dimyristoyl-L-α-phosphatidylcholine (DMPC)]. AC shows much stronger fluorescence when it is associated with membranes (more hydrophobic environment) while TPR shows almost same fluorescence efficiency with and without model membranes.

To study the effects of intracellular environment on the fluorescence efficiency of the probes C625 and TPR, we coupled them to [D-Lys ⁶ ]LH-RH to form LH625 and LHTPR, respectively. LH625 and LHTPR were then added to the medium of MCF7 cells and the fluorescence spectra from the nucleus, the cytoplasm, and the media recorded (Figure 5). The peaks from LH625 and LHTPR in the medium are readily observed. The calculated ratiometric spectral profiles of LHTPR and LH625 were similar in the nucleus and the cytoplasm.

Figure 5

Localized spectra from the nucleus and cytoplasm of an LH625 and LHTPR treated MCF-7 breast cancer cell and its background spectra from the media. LH625 (0.6 μM) and LHTPR (3.8 μM) were added to the medium of cells and localized spectra were acquired 90 min later. All the acquired spectra were normalized with respect to the maximum fluorescence intensity.

In other experiments, we added AC and LHTPR to MCF-7 cells and studied the spectroscopic profiles inside the cells. Figure 6(a) shows the localized spectra from the nucleus, the cytoplasm of an MCF-7 cell. The cell was pretreated with EGF to upregulate the number of LH-RH receptors on the membrane of cells,¹¹ and then treated with AC (0.6 μM) and LHTPR (3.8 μM). LHTPR concentration used in this case is considerably lower than that of TPR (10 μM) in the control experiment. Figure 6(b) shows the spectrum taken from the background without any cells, where we can clearly see that the emission from LHTPR is dominating.

Figure 6

Localized spectra from the nucleus and cytoplasm of an AC and LHTPR treated MCF-7 breast cancer cell and its background spectra from the media. AC (0.6 μM) and LHTPR (3.8 μM) were added to the medium of cells and 90 min later, localized spectra were acquired from different compartments of the cell and from the medium. All the acquired spectra were normalized with respect to the maximum fluorescence intensity.

This experiment was repeated in different cultures of MCF-7 cells and the spectra were taken in several cells of each culture. The spectroscopic profiles of the nucleus and the cytoplasm were reproducible. Occasionally, the fluorescence intensity of AC and LHTPR in the cytoplasm was found to be similar, but in all cases the ratio of AC to LHTPR in the nucleus (2.46±0.53) was significantly higher than that in the cytoplasm (0.88±0.45) (as calculated from the peak fluorescence intensities of the fluorescent tags C625 and TPR, respectively).

In another experiment, we intentionally induced photodamage to the cell by increasing the laser power. The dynamic change of the spectral profile in the nucleus can be clearly observed by TPLSMF [Figure 7(a)]. Figure 7(b) shows the fluorescence images of the same cell before and after photodamage, respectively, which could be easily identified through the observation of morphology changes by transmission and fluorescence microscopy. Our procedure was to first acquire the localized spectra (Δ) from the nucleus of the cell, and then induce photodamage to the cell by increasing the power of the laser. At that time, another spectra (○) was acquired from the nucleus by shifting the focus a small distance away to avoid the photobleached region. A dramatic difference in profiles before and after cell damage was observed. While the emission intensity from AC was much higher than that of LHTPR in the nucleus of the living cell, the emission from LHTPR virtually dominated in the case of the photodamaged cell.

Figure 7

(a) Localized spectra from the nucleus of an AC and LHTPR treated MCF-7 breast cancer cell before and after photodamage. After acquiring the localized spectra(Δ) from the nucleus of the cell, we intentionally induced photodamage to the cell by increasing the power of the laser. The other spectra (○) were acquired from the nucleus by shifting the focus a small distance to avoid the photobleached region. (b) Fluorescence images of live cell (a) and dead cell (b). The white arrows in the picture indicate the cell from which the spectra were recorded.

We have also used a single-photon confocal microscopy to image the entry of unlabeled Dox into the MCF-7 cells.¹¹ Dox shows some fluorescence with an excitation wavelength of 488 nm. Its entry into cells was monitored by one-photon confocal microscopy using the same equipment as described above, but equipped with an emission filter for 580 nm (band-pass filter 580/32). Figure 8 shows a single photon image of an MCF-7 cell treated with Dox (1 μM).

Figure 8

Single photon confocal microscopy imaging of Doxorubicin in MCF-7 breast cancer cells using a 488 nm line from a Kr:Ar laser as an excitation source. A 580/30 emission filter was used in front of the fluorescence detector. The cells were incubated with Doxorubicin for 30 min before imaging: (a) reflection mode image and (b) fluorescence mode image.

3.2.

Discussion

In MCF-7 cells treated with TPR (10 μM) and AC, the spectra of the nucleus and the cytoplasm are very similar and the peak positions are close to the emission spectrum of C625 in chloroform (Figure 3). There is no spectroscopic indication of the presence of TPR inside the cells or at least, the entry of TPR into the cells was so negligible that their spectrum was overridden by the emission of AC, even though the concentration of TPR (10 μM) used in this experiment was much higher than the concentration of LHTPR (3.8 μM) used in other experiments. In contrast, the spectra obtained from the medium showed no emission peak from AC, while we could clearly see the peak corresponding to the emission of TPR. The lack of signals from C625 in the medium can be due to the low fluorescence efficiency of AC in water, and also to a high degree of internalization. The strong emission signals for AC inside the cells could be a result of the dramatic increase in its fluorescence efficiency during the interaction of the lipophilic C625 moiety with cellular membranes as demonstrated with the model membrane, DMPC (Figure 4). The lipophilic property of the C625 moiety of AC actually aids in understanding the intracellular environment of the drug.

In the case of cells treated with AC and LHTPR, although the spectra obtained from the media and the cells differed, all the spectra taken inside the cells, including the nucleus, showed the emissions from both AC and LHTPR. These are dramatically different from the spectra in Figure 3, in which TPR (10 μM) instead of LHTPR (3.8 μM) was used together with AC. The spectroscopic exclusion of TPR from MCF-7 cells made it an excellent probe to track the entry of LHTPR. This result supports our previous observation that LH-RH could also enter the nuclei of LH-RH receptor positive cells.^{10 11} Recently the presence of nuclear binding sites for LH-RH in the nuclear fraction of the LH-RH receptor-positive OV-1063 human ovarian cancers was also demonstrated by radioreceptor assays.¹³ On the other hand, we found different intensity ratio profiles between AC and LHTPR emission in different compartments of cells. The spectra in Figure 6 correspond to the emission from the points shown in Figure 4(b), which were identified by us with transmission and fluorescence microscopy as the nucleus or cytoplasm. Our results indicate that the ratio of the peak fluorescence values between AC and LHTPR inside the nucleus is consistently greater than in any other part of the cell.

This ratiometric difference cannot be due to the changes in fluorescence efficiency of the two probes, in different cellular environment, as the cells showed similar ratiometric spectral profiles after the treatment with LHTPR and LH625 (Figure 5). The study on the intracellular spectral patterns of LHC625 and LHTPR also ruled out another possibility that probes may cause the distribution difference of AC and LHTPR in MCF-7 cells.

In contrast to living cells, the fluorescence from LHTPR in the nucleus of dead (photodamaged) MCF-7 cells was much higher than that from AC, as shown in Figure 7. This dynamic change may be due to the “leaky” membranes of the damaged cell. After the photodamage occurs, the nucleic membranes become more permeable and less selective to the molecules outside. Because there are no membrane structures inside the nucleus like those in cytoplasm and AC has much weaker fluorescence in a hydrophilic environment, the nucleic fluorescence from LHTPR would become dominant. Conversely, when the cells are healthy, it could be much more difficult for LHTPR than for AC to enter the nucleus.

As we do not know the fluorescence efficiency of C625 and TPR inside the cells, we cannot quantitatively correlate their intensities with the local concentrations. However, at the same local environment, we can still characterize their dynamic changes in relative concentrations from the ratio of their spectral intensities. The differences in the emission profiles in the nucleus and the cytoplasm imply a different ability of AN-152 and [D-Lys ⁶ ]LH-RH to enter the nucleus. AN-152 is a cytotoxic conjugate of Dox and [D-Lys ⁶ ]LH-RH. As one of the most widely used anticancer drugs, Dox has been shown to have a broad spectrum of antitumor effect.¹⁴ Even though Dox has been used in the clinics for about 30 yr, its mechanism of action is still not completely clear and has been the subject of intense investigations.⁵ Present concepts about the toxic effect of Dox are that it could initiate DNA damage through inhibition of topoisomerase II, after intercalation into the DNA, which could happen only inside the nucleus. The mechanism of action of the hormone conjugate AN-152 should also be elucidated. Our previous research has shown its cellular pathway into the nuclei of LH-RH receptor-positive cells, but it is not certain whether this entry was caused by the Dox component or the peptide. From the localized spectroscopic profiles obtained by TPLSMF, it appears that AN-152 can accumulate in the nucleus more readily than the carrier itself.

Furthermore, our previous study¹¹ using single photon confocal microscopy, has shown that Dox without any carrier can accumulate in the nuclei of MCF-7 cells. From Figure 8 in this study, it is clear that Dox is present in the nucleus at much higher concentrations than in the cytoplasm of MCF-7 cells. Therefore, it is reasonable to assume that the Dox part of AC is responsible for the higher relative concentration of AC in the nucleus. This study provides the first visual evidence that AN-152 may have a cytotoxic mechanism of action similar to that of Dox.