2.1.

Cell Lines

Rat alveolar Ma (8383) and rat glioma cells (F98) were obtained from the American Type Culture Collection (Manassas, Virginia). The ${\mathrm{BT}}_4\mathrm{C}$ rat glioma cell line was obtained from Oslo University Hospital (Oslo, Norway). The cell line was originally derived from transformed fetal BD-IX rat brain cells after exposure to ethyl—nitrosourea. Both F98 and ${\mathrm{BT}}_4\mathrm{C}$ cells were grown as monolayers in Dulbecco’s modified Eagle medium (Thermo Fisher Scientific, Carlsbad, California) with 10% fetal bovine serum, 25 mM HEPES buffer (pH 7.4), penicillin ($100\mathrm{U}{\mathrm{ml}}^{-1}$), and streptomycin ($100\mu \mathrm{g}{\mathrm{ml}}^{-1}$) at 37°C and 5% ${\mathrm{CO}}_2$.

2.2.

Effects of Photodynamic Therapy on F98 Cells

The direct effect of ${\mathrm{AlPcS}}_{2\mathrm{a}}$ PDT on F98 glioma cell viability was first assayed in monolayers. For each light level investigated, eight wells (in a 96-well flat bottomed plate) were seeded with F98 at a density of 5000 cells per well and incubated for 24 h. Cells were aliquoted in every third column to reduce light exposure from neighboring wells under illumination. About $1\mu \mathrm{g}{\mathrm{ml}}^{-1}$ of ${\mathrm{AlPcS}}_{2\mathrm{a}}$ was added to the cells and the incubation was continued for an additional 18 h. The cells were washed twice in fresh medium to remove excess photosensitizer. Light treatment was accomplished with a 670-nm diode laser (Intense, North Brunswick, New Jersey) coupled to a $200\text{-}\mu \mathrm{m}$ diameter optical fiber at an irradiance of $5\mathrm{mW}{\mathrm{cm}}^{-2}$. The cells were irradiated for increasing times resulting in a range of radiant exposures between 0 and $5\mathrm{J}{\mathrm{cm}}^{-2}$. Following light treatment, incubation was continued for 48 h, at which point the culture medium was replaced with fresh clear buffer containing MTS reagents (Promega, Madison, Wisconsin) and incubated for an additional 2 h. The optical density was measured using an ELx 800uv Universal Microplate Reader (BIO-TEK Instruments, Inc., Winooski, Vermont).

2.3.

Flow Cytometry

Flow cytometry was used to determine the fraction of viable, apoptotic, and necrotic cells following PDT. Two different fluorescent labels were used: Annexin V-FITC (Beckton, Dickson and Company, Franklin Lakes, New Jersey) to distinguish apoptotic cells, and propidium iodide (PI; Sigma, St. Louis, Missouri) to label necrotic cells. Unlabeled cells were assumed to be viable. The two labels were added to $1\times {10}^6$ F98 cells with $5\mu \mathrm{L}$ of Annexin V-FITC and $10\mu \mathrm{L}$ of $100\mu \mathrm{g}{\mathrm{ml}}^{-1}$ PI. The solutions were gently agitated and incubated for 15 min at room temperature in the dark. For each experiment, a set of control solutions was also prepared. One control remained unlabeled, the second labeled only with Annexin V-FITC, and the third labeled only with PI. Finally, after 15 min of dark incubation, each tube was analyzed in a flow cytometer (FACS Calibur; Beckton, Dickson and Company) along with CellQuest software.

2.4.

Generation of PDT-Ma Vaccine

F98 monolayers were grown in T25 flasks until subconfluence and were incubated with photosensitizer (${\mathrm{AlPcS}}_{2\mathrm{a}}$; $1\mu \mathrm{g}{\mathrm{ml}}^{-1}$) for 18 h. The cells were detached from the flask by enzyme treatment, washed twice to remove excess photosensitizer, and irradiated with $\lambda =670\mathrm{nm}$ light at an irradiance and radiant exposure of $5\mathrm{mW}{\mathrm{cm}}^{-2}$ and $5\mathrm{J}{\mathrm{cm}}^{-2}$, respectively. The PDT-treated F98 cells ($1\times {10}^6$) were cocultured with $0.5\times {10}^6$ Ma for 24 h. The antigen activated Ma are designated ${\mathrm{Ma}}^{\mathrm{F}98}$ in the text and figures. To evaluate morphological changes of the Ma after exposure to PDT-treated F98 cells, $2\times {10}^4{\mathrm{Ma}}^{\mathrm{F}98}$ cells were placed on a glass bottomed imaging dish and incubated for 18 h. Phase contrast microscopy was carried out both on ${\mathrm{Ma}}^{\mathrm{F}98}$ and “empty” Ma for comparison.

2.5.

Experimental Animals

Inbred male Fischer rats weighing about 350 g were purchased from Simonsen Laboratories, Inc. (Gilroy, California). Animal care and protocols were in accordance with institutional guidelines. Animal holding rooms were maintained at constant temperature and humidity on a 12-h light and dark schedule at an air exchange rate of $18\text{changes}/\mathrm{h}$. For the surgical procedures, animals were anaesthetized with isoflurane. A postoperative analgesic (buprenorphine: $0.08\mathrm{mg}{\mathrm{kg}}^{-1}$ s.c.) was administered following surgery and twice per day for 3 days thereafter. All animals were euthanized at the end of the study or at the first signs of distress. Euthanasia was accomplished via ${\mathrm{CO}}_2$ inhalation.

2.6.

Tumor Cell Injection

F98 and ${\mathrm{BT}}_4\mathrm{C}$ cells were injected stereotactically into the brains of Fischer rats, as previously described.²² Briefly, anaesthetized rats were fixed in a stereotactic frame. The skin was incised and a 1.0-mm burr hole was made at the following coordinates: 1 mm posterior to the bregma, 2 mm to the right of the midline, and at a depth of 2 mm. The injection device consisted of a 30-G blunt cannula connected through a catheter (Hamilton Co., Reno, Nevada) to an infusion pump (Harvard Apparatus, Holliston, Massachusetts). The cannula was fixed in the electrode holder of the stereotactic frame and then vertically introduced into the brain. A total of ${10}^4$ F98 and/or ${10}^4$ ${\mathrm{BT}}_4\mathrm{C}$ cells in $20\text{-}\mu \mathrm{l}$ PBS were injected into the brain over a period of 2 min. Following injection, the cannula remained in place for 2 min. Closure was done with bone wax and sutures.

2.7.

Experimental Protocol

Animals were divided into five cohort groups. The basic protocols for these five groups are shown in Fig. 1. Each group consisted of three animals. Since this was a proof-of-principle histological study, with no survival arm, the limited number of animals used was considered sufficient. Group 1: animals were injected with F98 cells only. Group 2: animals were injected with F98 cells in the right hemisphere (syngeneic controls) and ${\mathrm{BT}}_4\mathrm{C}$ cells in the left hemisphere (allogeneic controls). Group 3: animals were injected with a mixture of F98 and ${\mathrm{BT}}_4\mathrm{C}$ cells in the right hemisphere. Group 4: animals were injected (i.p.) with a combination of PDT-treated F98 tumor cells and ${\mathrm{Ma}}^{\mathrm{F}98}$. Forty-eight hours following immunization, F98 cells were injected into the right hemisphere. Group 5: animals were injected (i.p.) as in group 4. Forty-eight hours following immunization, a mixture of F98 and ${\mathrm{BT}}_4\mathrm{C}$ cells were injected into the right hemisphere. The animals were followed for 14 days, euthanized, and the brains removed. Histology of removed brains was performed with hematoxylin and eosin (H&E) staining.

Fig. 1

Experimental protocols for all five cohort groups. Group 1: injected i.c. with F98 cells. Group 2: injected i.c. with F98 cells in the right hemisphere (syngeneic controls) and ${\mathrm{BT}}_4\mathrm{C}$ cells in the left hemisphere (allogeneic controls). Group 3: injected with a mixture of F98 and ${\mathrm{BT}}_4\mathrm{C}$ cells in the right hemisphere. Group 4: (1) ${\mathrm{AlPcS}}_{2\mathrm{a}}$ PDT treatment ($5\mathrm{mW}{\mathrm{cm}}^{-2}$; $5\mathrm{J}{\mathrm{cm}}^{-2}$) of F98 glioma cells inducing generation of tumor antigens. (2) 24-h coincubation of Ma and PDT-treated F98 cells. (3) i.p. injection of ${\mathrm{Ma}}^{\mathrm{F}98}$. Forty-eight hours following immunization, F98 cells were injected into the right hemisphere. Group 5: ${\mathrm{Ma}}^{\mathrm{F}98}$ injected i.p. as in group 4. Forty-eight hours following immunization, a mixture of F98 and ${\mathrm{BT}}_4\mathrm{C}$ cells were injected into the right hemisphere. All animals were followed for 14 days, euthanized and the brains removed. Histology of removed brains was performed with H&E staining.

3.1.

PDT Response of F98 Cells

The direct effects of ${\mathrm{AlPcS}}_{2\mathrm{a}}$ PDT on F98 cell monolayers are shown in Fig. 2(a). At a radiant exposure of $5\mathrm{J}{\mathrm{cm}}^{-2}$, less than 3% of the cells were viable. This radiant exposure was therefore used in the subsequent experiments. The results of flow cytometry for two PDT radiant exposures are shown in Fig. 2(b). At $0.5\mathrm{J}{\mathrm{cm}}^{-2}$, most of the cells were viable as was also seen in the MTS assay [Fig. 2(a)]. At $5\mathrm{J}{\mathrm{cm}}^{-2}$, only a small fraction of F98 cells were viable as most of the cells died via apoptosis. Figure 2 shows flow plots of apoptosis of untreated (c) or PDT-treated (d) F98 cells stained with annexin V-FITC and PI. Early apoptotic cells were defined as annexin V-FITC+ but PI−. These findings are in good agreement with previous results showing that apoptosis is the primary mode of cell death following low irradiance PDT.^23–25

Fig. 2

Generation of immunogenic apoptotic cells and effect on Ma. (a) Cell viability of F98 glioma cells following PDT. (b) Flow cytometry results of PDT-treated F98 cells and flow cytometry scatter plots of (c) nontreated and (d) PDT-treated F98 cells. Cells were incubated in $1\mu \mathrm{g}{\mathrm{ml}}^{-1}$ ${\mathrm{AlPcS}}_{2\mathrm{a}}$ for 18 h and irradiated with 670-nm light at an irradiance of $5\mathrm{mW}{\mathrm{cm}}^{-2}$. (e) Phase contrast micrograph of nonactivated Ma. (f) Altered Ma morphology following 24 h coincubation of PDT F98 and Ma (${\mathrm{Ma}}^{\mathrm{F}98}$). Arrow denotes apoptotic F98 cell and the scale bar represents $10\mu \mathrm{m}$. PDT experiments were performed in triplicate and error bars denote standard deviations.

Several reports have demonstrated that apoptotic cells are superior to necrotic cells in inducing antitumor immunity.^26–30 Recent research has led to the concept of immunogenic cell death, which refers to an immunogenic form of apoptosis or necrosis. Cells undergoing immunogenic apoptosis are more potent inducers of antitumor immune responses than cells dying via necrosis. Recent work by Ji et al.³¹ has shown that ALA-PDT is capable of generating immunogenic apoptotic cells and that these cells can activate immature DCs. Vaccines using these activated DCs inhibited the growth of squamous cell carcinoma in mice.

3.2.

Morphology of Ma and Ma^F98

Coincubation of Ma with PDT-treated F98 cells led to pronounced morphological changes of the ${\mathrm{Ma}}^{\mathrm{F}98}$ compared to nonstimulated Ma. As shown in the phase contrast micrographs [Fig. 2(e)], 8383 rat Ma are round, $\sim 10$ to $15\mu \mathrm{m}$ in diameter and are composed of an equal population of both adherent and floating cells in vitro. By contrast, ${\mathrm{Ma}}^{\mathrm{F}98}$ [Fig. 2(f)] were significantly larger and irregular in shape with increased intracellular inclusions. Additionally, the majority of the cells were adherent.

3.3.

Effects of Implanting Syngeneic or Allogeneic Glioma Cells in Nonimmunized Animals

Control animals implanted with $2\times {10}^4$ F98 cells into the brain developed large tumors 14 days postinjection (Fig. 3). The lack of encapsulation and significant infiltration of F98 cells into normal brain has been observed previously.³² F98 glioma cells have been used in numerous experimental brain tumor studies since the tumors share many fundamental traits with human GBM, including infiltration, rapid growth, extensive neovascularization, absence of encapsulation, and weak immunogenicity.³³ The weak immunogenicity of the F98 model is a significant advantage in evaluating the effects of vaccines compared to other rat glioma models (e.g., C6 and 9L gliosarcoma), which can be highly immunogenic and can therefore evoke intense immune responses on their own.

Fig. 3

Photomicrographs of H&E-stained sections showing F98 cells injected stereotactically into the brains of Fischer rats. (a)–(c) The three animals were euthanized 14 days posttumor implantation.

Group two animals received F98 cells in their right hemisphere and allogeneic ${\mathrm{BT}}_4\mathrm{C}$ cells in their left hemisphere. As illustrated in Fig. 4, all F98-cells developed into tumors while no tumors were observed in the ${\mathrm{BT}}_4\mathrm{C}$-injected hemispheres. This clearly demonstrates the rejection of allogeneic glioma cells in the brain. By contrast, ${\mathrm{BT}}_4\mathrm{C}$ cells readily formed tumors when injected into the brains of syngeneic BD-IX rats.³⁴

Fig. 4

H&E-stained sections showing F98 and BT4C cells injected stereotactically into the brains of two Fischer rats (a and b). All animals developed F98 tumors while none developed ${\mathrm{BT}}_4\mathrm{C}$ tumors.

Interestingly, the syngeneic F98 tumors in these animals were much smaller in comparison to the F98-only tumors implanted in group one animals. Injecting a mixture of F98 and ${\mathrm{BT}}_4\mathrm{C}$ cells into the same hemisphere (group 3) also resulted in the development of tumors (Fig. 5) similar in size to those seen in the group 2 animals (Fig. 4). Taken together, these findings indicated that the rejection reaction against the ${\mathrm{BT}}_4\mathrm{C}$ cells could slow tumor progression but was insufficient to completely prevent F98 tumor formation. Although the underlying mechanism of the antitumor immunity observed in these experiments remains to be determined, allogeneic cells likely contain antigen determinants shared with the syngeneic tumor, leading to the observed reduction in tumor growth. This hypothesis is in agreement with the findings of Stathopoulos et al.,³⁵ who reported that Fischer rats, after initially rejecting subcutaneous C6 allogeneic tumors, failed to develop tumors after subcutaneous injection of syngeneic 9L glioma cells.

Fig. 5

H&E-stained sections showing mixtures of F98 and ${\mathrm{BT}}_4\mathrm{C}$ cells injected in the same hemisphere of two animals (a and b). All animals developed tumors, indicating the rejection reaction against the ${\mathrm{BT}}_4\mathrm{C}$ cells was insufficient to prevent F98 tumor growth.

3.4.

Tumor Development in Immunized Animals

Ma were coincubated with PDT-treated apoptotic F98 (${\mathrm{Ma}}^{\mathrm{F}98}$) cells for 24 h. A new cohort of animals (Group 4) were inoculated (i.p.) with the resulting mixture of the two cell types. Forty-eight hours following immunization, F98 cells were injected into the right hemisphere. Fourteen days later, the brains were removed and sectioned. The resulting histological sections are shown in Fig. 6. Although tumors developed in all of these ${\mathrm{Ma}}^{\mathrm{F}98}$ immunized animals, the tumors were significantly smaller than in the nontreated controls (group 1) as well as in the groups receiving both F98 and ${\mathrm{BT}}_4\mathrm{C}$ intracranial injection (groups 2 and 3).

Fig. 6

H&E-stained sections showing limited tumor development in 3 vaccinated animals (a, b and c). Animals were inoculated i.p. with PDT-treated F98 loaded Ma (${\mathrm{Ma}}^{\mathrm{F}98}$) 2 days prior to stereotactic i.c. implantation of $2\times {10}^4$ F98 tumor cells and sacrificed 14 days later.

In the final cohort (group 5), allogeneic ${\mathrm{BT}}_4\mathrm{C}$ cells were injected along with the F98 glioma cells into ${\mathrm{Ma}}^{\mathrm{F}98}$ immunized hosts. As illustrated in Fig. 7, histological sections revealed that there was no evidence of tumor development in any of the animals. Although small remnants of the initial cell inoculate could be seen [Figs. 7(a) and 7(b)], no tumor growth or infiltration was evident. The addition of an allogeneic immune response therefore appeared to enhance the efficacy of the ${\mathrm{Ma}}^{\mathrm{F}98}$ vaccine.

Fig. 7

H&E-stained sections from three animals inoculated i.p. with ${\mathrm{Ma}}^{\mathrm{F}98}$ 2 days prior to intracranial implantation of $2\times {10}^4$ allogeneic ${\mathrm{BT}}_4\mathrm{C}$ cells together with an equal number of syngeneic F98 cells. (a) and (b) Remnants of the implanted cells but no tumor development in any of the animals are shown. (c) No implanted cells are evident.

High magnification ($250\times$) H&E histological sections of tumors removed from nontreated controls (group 1), $\mathrm{F}98+{\mathrm{BT}}_4\mathrm{C}$ implants (group 3), ${\mathrm{Ma}}^{\mathrm{F}98}$ immunized animals (group 4), and ${\mathrm{Ma}}^{\mathrm{F}98}$ immunized $\text{animals}+{\mathrm{BT}}_4\mathrm{C}$ (group 5) are shown in Figs 8(a)–8(d), respectively. Control tumors were extremely cell rich and compact [Fig. 8(a)], whereas the tumors from the immunized animals demonstrated a much lower cell density [Fig. 8(c)] with areas in the tumor that contained almost no visible tumor cells. Sections from group 5 animals showed only small remnants of the initial cell inoculate, as shown in Fig. 8(d).

Fig. 8

Magnified ($250\times$) H&E sections of (a) F98 injected control animals (group 1), (b) $\mathrm{F}98+\mathrm{BT}4\mathrm{C}$ i.c. (group 3), (c) ${\mathrm{Ma}}^{\mathrm{F}98}$ vaccine treated animals (group 4), and (d) ${\mathrm{Ma}}^{\mathrm{F}98}$ vaccine treated $\text{animals}+\mathrm{BT}4\mathrm{C}$ i.c. (group 5).

The generation of vaccines using PDT has been attempted by a number of groups. For example, Shixiang et al.³⁶ generated DC vaccines using PDT-treated C6 glioma cell antigenic peptides. Vaccine efficacy was assessed ex vivo by DC-induced cytotoxic T lymphocyte mediated lysis of C6 target cells. They concluded that PDT of C6 cells significantly enhanced tumor cell immunogenicity compared to freeze-thawed C6 cells.

The use of established allogeneic cell lines to generate glioma vaccines based on HLA typing of individual patients was proposed by Zang et al.,³⁷ who suggested that HLA typing can predict which combination of cell lines is capable of producing a customized allogeneic vaccine containing the relevant antigens to stimulate the patient’s immune system. Alternatively, this approach would allow for autologous DC vaccines to be generated using lysates containing these relevant antigens. Based on previous studies, Stathopoulos et al.³⁸ reported a prototype brain cancer vaccine against glioma cells containing multivalent antigens derived from both allogeneic and syngeneic cells and lysates. In a subcutaneous tumor model, they demonstrated arrested progression of tumor growth when the vaccine was codelivered with costimulatory agents. The results of a clinical phase I trial on nine recurrent GBM patients have recently been reported.³⁹ The immunization approach, consisting of combined administration of multiple allogeneic and autologous tumor-isolated antigens derived from the patient’s surgically removed tumor tissue and glioma tumor tissue from three allogeneic donor patients, was based on the preclinical approach described in CNS-1 Lewis rats.³⁸ A phase II trial (NCT01903330) employing this approach is currently ongoing.

Although the strategy used in the present proof-of-concept study reported here (i.e., Ma vaccine generated by PDT treatment of the tumor cells combined with direct intracranial alloimmune stimulation) has proven promising, work is in progress aimed at inducing more efficient immunization in a larger number of animals. In the present protocol, the immunizing loaded ${\mathrm{Ma}}^{\mathrm{F}98}$ were injected i.p., however, direct injection into the lymph nodes, where T cell priming occurs, is likely a better approach. Specifically, studies have shown that priming of T cells by APCs in the cervical lymph nodes can induce a homing pattern toward locations within the brain.⁴⁰ Additionally, live allogeneic cells were injected directly into the brain and, as such, this approach is not translatable to the clinic. Developing an Ma vaccine using PDT treatment of both the syngeneic and allogeneic cells, to prime the Ma, as well as a more realistic postoperative brain tumor model are currently being explored.