Targeting the adult mouse hippocampus using modular protein-only
nanoparticles
Renée R. C. E. Schreurs
Supervisors: C. P. Fitzsimons, J. Domingo-Espín, M. Schouten
ABSTRACT Due to their natural ability to internalize their own genome into target cells, viruses hold considerable appeal as vectors used in gene therapy. However, whenever (part of) a pathogen penetrates the body our immune system is activated, often resulting in damage to our own tissue or to the therapeutic viral vector. This study explores the use of a protein-only nanoparticle (HNRK) to deliver genetic material to target cells in vivo. Twenty-eight mice received bilateral injections in the hippocampus consisting of HNRK loaded with either DNA encoding GFP or microRNA-124 tagged with Cy3. Their brains were extracted and sectioned after either a two-day or a one-week incubation period and examined under a fluorescent microscope. Both the DNA and the microRNA could be successfully delivered in the target tissue, with visible effects after two days. However, the effects were absent after one week. These results call for further experiments in order to establish the safety and functionality of HNRK as a therapeutic vector for gene transfer.
1. Introduction
Viruses hold considerable appeal as vectors used in gene therapy due to their natural ability to internalize their own genome into target cells. Simply put, viral particles are microscopic entities composed of nucleic acids and proteins that allow them to survive in an extracellular environment and mediate their internalization into target cells [1]. The key in using viruses as vectors lies in the exploitation of these biological properties and the substitution of the viral genome with therapeutically relevant nucleic acids before introducing them to a host. However, fighting an infection by pathogens such as viruses is among the key functions of the immune system.
Although viral vectors used in gene therapy have been stripped of their potentially dangerous viral genome, whenever (part of) a pathogen penetrates the body, an immune response is likely to develop [2]. In most cases, the induction of such a response is beneficial to the host. However, an immune response directed against a gene therapy vector may render its therapeutic effects void [2]. Any immune response may involve the production of pro-inflammatory cytokines and chemokines [3] or in the case of intracellular cargo delivery via microbial infection, toll-like receptors are activated that initiate a rapid and innate inflammatory response that can have harmful effects [4].
Circumventing this immune response, while
simultaneously retaining the ability of the vector to deliver therapeutic genes within the target cells, is a major challenge with all viral vector types.
The undersired side effects often observed in viral based gene therapy trials have encouraged the exploration of nonviral strategies. Several attempts are being made at constructing non-viral vectors for gene transfer [5,6]. For example, virus-like particles are being
constructed through the expression of viral structural genes to mimic solely the viral properties that are important for gene therapy [7]. A different approach has been to formulate therapeutic genetic material in protein, lipid, polymer, or peptide mixtures [7-10]. Interestingly, many of the biological properties that make viruses such successful instruments for transferring nucleic acids into host cells can be found in a diverse spectrum of organisms. For example, cell internalization, endosomal escape, uncoating, and replication are not properties exclusive to viral bodies [7]. Protein engineering allows for the combination of the desired biological traits for gene therapy into a single modular molecule by the selection and reorganization of functional domains already existing in nature [7].
Domingo-Espín and colleagues [11] produced a short structural protein (HNRK) in Escherichia coli comprising four functional domains: an integrin-binding motif (derived from the foot-and-mouth disease virus; FMDV), an endosomal escape, a nuclear localization signal (derived from the Simian virus 40; SV40), and a DNA-binding, cationic peptide (Fig. 1a). In the presence of DNA, these protein-DNA complexes form spherical-like nanoparticles (approximately 80 nm in diameter) that rather resemble bacterial inclusion bodies (Fig. 1b/c). These proteins produced in bacteria promote high transgene expression levels when used as artificial viruses in vitro [11]. This study will explore the functionality of the HNRK nanoparticle in vivo, loading the nanoparticle with either DNA or microRNA (miR) and injecting the complexes into the dentate gyrus (DG) of mice.
MiRs are small non-coding RNAs of
approximately 21-23 nucleotides long that regulate target gene expression through translational repression and/or target mRNA degradation [12-15]. Interestingly, miRs can regulate gene expression with a sequence that
is only partially complementary to that of the target mRNA [12], meaning that one miR can target and regulate tens, if not hundreds, of mRNAs. Even so, several types of miR have been shown to play important regulatory roles in the DG, a part of the hippocampus
Fig. 1. Organization, size and morphology of the structural protein (HNRK) (adapted with permission from [11]). (a) Distribution of the functional modules in HNRK. The histidine (His-tag) tail is labeled in blue, the Lysine (K 10) tail is labeled in yellow, the FMDV cell binding (arginine-glycine-aspartic; RGD) in green and the SV40 nuclear localization signal (NLS) in red. (b) Representative transmission electron microscopy image of HNRK alone and (c) of HNRK complexes with DNA.
that is important in learning and memory processes and one of only two locations in the adult brain where new neurons are generated [16,17]. For example, miR-124 has been implied as a regulator of the glucocorticoid receptor (GR) [18] and of the progression of adult neurogenesis, although these functions have not yet been demonstrated in the hippocampus.
On a cellular level, glucocorticoids bind to one of two types of nuclear receptors that are both abundantly expressed in the DG: the mineralocorticoid receptor and the GR [19,20]. The binding of glucocorticoids to the GR allows these receptors to regulate the development of subgranular zone (SGZ) stem cells on the level of proliferation and of transit amplifying cells transforming into neuroblasts (Fig. 2) [19,21-23].
Glucocorticoids are secreted by the adrenal glands in response to stress [24,25]. Increased levels of activated GR, resulting from an increase in the release of these “stress hormones” can easily disturb the delicate process of adult hippocampal neurogenesis (AHN) [26]. Studies have shown that increased levels of GR result in decreased levels of AHN [27,28]. Since miR-124 is thought to be involved in the regulation of GR protein levels, presumably through the translational repression/degradation of possibly several mRNAs, an increase in miR-124 during stressful events may aid in restoring activated GR to normal levels [29]. Due to its implied role in the progression of adult neurogenesis, an increase in miR-124 in the DG, as a result of nanoparticle delivery, may also cause altered neurogenesis under non-stressful circumstances [13].
In this study, HNRK complexes will be loaded with either DNA or miR. Green fluorescent protein (GFP) will be used as a reporter of transgene expression of HNRK/DNA complexes. The HNRK/miR complexes will consist of miR-124, tagged with Cyanine dye 3 (Cy3). Non-complexed DNA with GFP and non-targeting (NT) Cy3-miR in saline will serve as negative controls. Here we report that HNRK complexes can be successfully delivered and expressed transgenically in the DG of mice.
2. Materials and Methods
2.1 Animals
Animal experiments were performed in accordance with the laws and regulations set by the European Union and were approved by the committee of Animal Welfare at the University of Amsterdam (DEC# 236). For the purpose of these experiments, 38 male BALB/c mice (Harlan Laboratories, Inc.), weighing 25-30 grams, were used. The animals arrived at age 6-8 weeks and were housed in groups (maximum 8/cage) for at least one week to allow for acclimatization. 10
Fig. 2. The regulation of AHN. Stem cells in the SGZ proliferate and transform into transit amplifying cells which then form neuroblasts that migrate through the hippocampus. MiR-124 downregulates the GR, which hampers both the transformation of stem cells into transit amplifiers and of transit amplifiers into neuroblasts.
animals were used to determine the correct dorsoventral coordinates for the injection site. These 10 animals were euthanized before surgery using a lethal dose of pentobarbital (60 mg/ml) injected intraperitoneally. The remaining mice received bilateral injections into the dentate gyrus, following the coordinates established with the first group. All animals were kept in a temperature and humidity controlled room and under a 12 hour light-and-dark cycle (lights on at 08:00 AM) and had access to food and water ad libitum. The well-being of the animals was monitored daily and documented by the responsible researcher.
2.2 Determination of coordinates for injection
In order to minimize animal suffering, 10 animals were euthanized before surgery using an intraperitoneally injected lethal dose of pentobarbital (60 mg/ml). Injections were then performed as described under heading 2.3. Mediolateral (ML) coordinates (-1.5/1.5 mm relative to Bregma) and anteroposterior (AP) coordinates (2.0 mm relative to Bregma) were determined using existing literature [30]. To determine the correct dorsoventral coordinates (DV), 1 µl bromophenolblue (2 mg/ml) was injected bilaterally into the dentate gyrus at five different depths (2 animals per coordinate), namely -2.3, -2.2, -2.1, -2.0, and -1.9 mm relative to Bregma.
2.3 Stereotactic surgery
Surgery was performed in sterile conditions and in anesthetized animals using a modified version of Cetin
et al.’s [31] protocol. General anesthesia was induced by isoflurane inhalation (3% isofluorane mixed in pure oxygen) in a Plexiglas induction chamber padded with tissue. Anesthesia was maintained with an airflow-system, using a 1.5-2% isoflurane mixture that was supplied to the animals through a breathing mask fitted over the nose. When the animals failed to show any reflexes, their scalp was shaved and they were placed on a heatpad (37°C) to prevent loss of body temperature. The animals were placed in a stereotactical device using ear bars. The eyes were kept moist with eye drops. The scalp was cleaned with 70% ethanol after which a small incision was made anteroposteriorally across the skull until both Lambda and Bregma were exposed. Lidocaine (10%, Astra Zeneca) was applied immediately afterwards and the skull was kept moist with sterile saline at all times. The coordinates of Lambda and Bregma were determined and the position of the skull was adjusted so that the difference between the DV coordinates at these two anatomical locations was never greater than 0.2 mm. When the skull was in the correct position, the drill and injection coordinates relative to Bregma were calculated and 2 small holes were drilled in the skull. An infusion pump (Harvard Apparatus,
PHD 2000) using glass pipette pulled needles and a Hamilton syringe were used to inject 1 µl (0.2 µl/min) of fluid (see Table 1 for a detailed surgery schedule) into each hemisphere. Three minutes after the infusion pump had stopped, the needle was slowly taken out of the brain. The animals then received two sutures to close the wound and were given a subcutaneous buprenorfine (0.075 mg/kg per 1 ml Ringer solution) injection. After surgery the animals were housed individually and animals showing extreme signs of weight loss (>15%) during recovery (did not occur), would have been euthanized using an intraperitoneal injection of pentobarbital (60 mg/ml).
Table 1
Stereotaxic injection schedule
Animals Injection fluid Incubation time 1-10 Bromophenolblue - 11-13 pmaxGFP 2 days 14-18 HNRK/pmaxGFP 2 days 19-21 Cy3-miR (NT) 2 days 22-26 HNRK/Cy3-miR-124 2 days 27-29 pmaxGFP 1 week 30-32 HNRK/pmaxGFP 1 week 33-35 Cy3-miR (NT) 1 week 36-38 HNRK/Cy3-miR-124 1 week 2.4 HNRK complex preparation
The HNRK protein complex was produced and purified according to the protocol described by Domingo-Espín et al. [11]. For the purpose of these experiments, the following concentrations of stock were used: 4300 ng/µl of protein, 500 ng/µl of pmax GFP, and 50µM of either Cy3-miR-124 or Cy3-miR (NT). Ten minutes prior to each injection, 3 µl of HNRK/pmaxGFP or HNRK/Cy3-miR-124 was prepared using a final protein concentration of 1.57 µg/µl and 0.32 µg/µl of pmaxGFP or 3.47 µM of Cy3-miR-124. In the case of injecting non-complexed pmaxGFP and cy3-miR (NT), similar volumes of sterile saline were used as a substitute for the protein [see also 32].
2.5 Brain slice cultures
All animals not used for DV determination were euthanized using a pentobarbital (60 mg/ml) injection and perfused using sterile saline and 4% paraformaldehyde (PFA) in phosphate buffered saline (PBS, pH 7.4). Brains were extracted and fully submerged in 4% PFA in PBS (pH 7.4) for 24 hours and then in 30% sucrose in phosphate buffer (PB) for 48 hours. The brains were cut using a sliding microtome (Jung, Germany) and were frozen to the microtome table
using 30% sucrose in PB and dry ice. The dentate gyrus was located and the brains were cut into 30 µm thick sections (the brains for coordinate determination and the one-week miR and HNRK/miR brains were cut at 40 µm to prevent the miR from washing out). Brain sections were transferredfrom the microtome to PB or to 0.01% sodiumazide in 0.1 M PB when further processing was not immediate.
2.6 Immunohistochemistry
NeuN, a neuronal nuclear antigen, staining was performed using a 1:200 dilution of monoclonal mouse primary antibody (Chemicon) and a 1:800 Alexa 594 conjugated goat anti mouse secondary (Invitrogen) in all pmaxGFP and HNRK/pmaxGFP samples and a 1:800 Alexa 488 rabbit anti mouse secondary (Invitrogen) in all miR and HNRK/miR samples. Furthermore, GFP staining in all pmaxGFP and pmaxGFP/HNRK samples were performed using a 1:200 dilution of polyclonal rabbit primary antibody (Invitrogen) and a 1:800 Alexa 488 donkey anti rabbit secondary (Invitrogen). Doublecortin (DCX), an antigen expressed by neuronal precursor cells and immature neurons, staining in miR and HNRK/miR samples was performed using a 1:1000 dilution of polyclonal anti rabbit primary (Abcam) and a 1:800 Alexa 488 donkey anti rabbit secondary
(Invitrogen). All samples used for
immunohistochemistry were further stained with a 1:25000 Hoechst 33342, which binds to DNA, in TBS solution for one minute. Some samples from the one-week miR and HNRK/miR were mounted immediately after sectioning inmounting medium containing DAPI, which binds to DNA. DCX and NEUN positive cells and colocalization of GFP, NEUN and Hoechst, and of cy3 and DAPI was analyzed using a fluorescent microscope (Leica DMI3000B) and camera (LeicaDFC310 FX). Fluorescent images were modified for figures using ImageJ software.
3. Results
3.1 Experiment 1: Coordinate determination
In order to establish the adjusted coordinates of the hilus of the mouse DG, bromophenolblue dye (2 mg/ml) was injected using five different DV coordinates (-2.3 to -1.9 mm relative to Bregma) and the predetermined coordinates: ML +/- 1.5 mm and AP 2.0 mm relative to Bregma. The brains were then extracted, sliced into sections of 40 µm thick and investigated under a light microscope. Upon examination it was clear that the correct coordinates for these stereotactic surgeries were ML +/- 1.5, AP 2.0, and DV -2.1 mm relative to Bregma (Fig. 3). The bromophenolblue dye is clearly visible in the infrapyramidal layer of the DG at this coordinate.
3.1 Experiment 2: GFP-expression after two days
Once the correct coordinates had been determined, the first set of surgeries was performed. During these stereotactic infusions either non-complexed pmaxGFP in saline, or HNRK/pmaxGFP complexes were injected bilaterally
into the DG of eight mice and left to incubate for two days before the animals were perfused and their brains
Fig. 3. DV injection coordinate for stereotactic surgery. The bromophenolblue (in blue) is clearly visible in the infrapyramidal layer of the DG.
were sectioned. In the HNRK/pmaxGFP group, investigation of the tissue samples after anti-GFP stainings showed evident but modest GFP expression in the DG, as revealed by the presence of green GFP-positive neurons in this region (Fig. 4). No GFP-GFP-positive cells were found in the non-complexed pmaxGFP injected animals (data not shown).
3.3 Experiment 3: GFP-expression after one week
Since GFP is only modestly expressed after an incubation time of two days, the experiment was repeated with a one-week incubation period. Six animals were used for these surgeries and received bilateral injections of either non-complexed pmaxGFP in saline or HNRK/pmaxGFP complexes. However, no GFP-positive cells were found in the DG of any of the animals (Fig 5a). Moreover, in all tissue samples, just one GFP-positive cell was found; in the subventricular zone near the hippocampus in one animal (Fig 5b). Although this is the only other area where adult neurogenesis occurs, it is not at all where the GFP complexes were intended to be delivered. However, it provides the evidence that the fluorescent anti-GFP staining was successful and that the absence of GFP-positive cells in the DG is most likely
not the result of a malfunctioning
3.4 Experiment 4: Cy3-miR-124 after two days
Another set of surgeries involved the injection of either non-complexed Cy3-miR (NT) in saline or HNRK/Cy3- miR-124 with an incubation period of two days before extraction. Fluorescent microscopic images of the acquired tissue samples shows Cy3-positive and complex-like entities around areas of tissue damage close to but not exactly in the DG (Fig. 6a). Even though the complexes are only visible outside the DG, this does not necessarily indicate that the injections hit the wrong area as one Cy3-positive cell was found in the infrapyramidal layer of the DG. This finding suggests successful transgenic expression in at least one HNRK/Cy3-miR-124 complex-injected animals (Fig. 6b). It must be noted that this cell was found in a section that was mounted onto a glass slide during preliminary investigations, straight from the microtome and without any washing or immunohistochemistry procedures.
3.5 Experiment 5: Cy3-miR-124 after one week
The presence of one Cy3-positive cell encouraged the repetition of the experiment using a one-week incubation period comparable to the experiments using HNRK/DNA complexes. It appears that complex-like entities can also be found around tears in these samples, although in far fewer numbers (Fig. 7). Many samples were mounted immediately after slicing and without immunohistochemistry protocols in order to find more Cy3-positive cells as occurred in the two-day sample. However, no Cy3-positive cells were found (data not shown).
All the brains of the animals injected with HNRK/Cy3-miR-124 were examined more closely for
changes in AHN. As was explained above,
overexpression of miR-124 in the DG may alter the quantity or morphology of newborn neurons in the granular cell layer. Therefore, all tissue samples (two
Fig. 6. HNRK-mediated delivery of Cy3-miR-124 in the mouse brain after two days. (a) Each image is representative of the five animals independently injected. Complex-like Cy3-positive entities can be found around tears in the tissue close to the DG. Panels on the left show Hoechst staining, panels in red show Cy3-positive complexes, panels in green show autofluorescence not co-occurring with Cy3-positiveity, and panels on the right show a merged image of the staining. (b) One Cy3-positive neuron (in red) was found in the DG of one animal. The panel in green shows autofluorescence not co-occurring with the Cy3-positive neuron, the panel in the middle shows one Cy3-positive cell (in red), and the panel on the right represents the boxed area in the middle panel.
Fig. 7. HNRK-mediated delivery of Cy3-miR-124 in the mouse brain after one week. In the top panel, a few Cy3-positive (in red, indicated by arrows) and complex-like bodies can be seen. In the bottom panel (autofluorescence), no such bodies are visible.
Fig. 8. DCX-positive cells (in green) in the DG after HNRK/Cy3-miR-124 delivery. (a) Schematic representation of the GCL in the mouse DG. Proliferating and migrating young mature neurons can be marked by DCX, while the more mature neurons can be marked by NeuN (ML = molecular layer). (b) DCX-positive cells in the SGZ of the DG in control animals (top panel), two days (middle panel) and one week (bottom panel) after HNRK/Cy3-miR-124 delivery. It appears as though more DCX-positive cells are present one week after delivery than in either control or two-days-after-delivery animals. (c) One week after delivery of the HNRK/Cy3-miR-124 complexes, some DCX-positive cells (indicated by the arrow) can be found in the outer GCL, a place where normally no DCX-positive cells are present.
Fig. 9. Cortical and hippocampal damage due to the injections. Each image is representative of the type of damage most often seen following bilateral injections into the hippocampus. The top panel shows cortical damage around the topical site of injection. The bottom panel represents the boxed area of the top panel and shows a DG with a severely damaged suprapyramidal layer. Tissue is stained with NeuN.
days and one-week incubations) of the mice injected with HNRK/Cy3-miR-124 or non-complexed Cy3-miR (NT) in saline were stained with DCX and NeuN, both markers for different stages of neuogenesis (Fig. 8a).
Although extensive damage in many brains (Fig. 9) did not allow for accurate quantification, it appears that there is at least a difference in the number of DCX-positive cells between the one-week samples and the two-day and control tissue; the one-week tissue samples appear to have more DCX-positive cells in the SGZ than either the two-day or the control tissue (Fig. 8b). Interestingly, some DCX positive cells were found in the outer granular cell layer (Fig. 8c). Future experiments are necessary in order to establish the existence of a quantitative difference. No quantitative or qualitative difference between the samples stained against NeuN appears to exist (data not shown).
4. Discussion
This study has demonstrated that HNRK-protein particles can be successfully delivered and transgenically expressed in the DG of mice. The construction of the HNRK-nanoparticles was optimized in vitro, after which they were loaded with either DNA or miR-124 in order to determine their functionality in vivo.
The results of the HNRK/DNA injections show that GFP-positive neurons can be found in the DG two days, but not one week after infusion of the nanoparticles. The absence of GFP-positive cells after one week is not the result of a malfunctioning antibody or immunohistochemistry-protocol, as one GFP-positive cell was found in the subventricular zone near the hippocampus in one of the brains. Even though this is not an area that was intentionally targeted, it indicated that the GFP transgene can still be expressed one week after HNRK/DNA injections.
The most plausible explanation for finding a GFP-positive cell outside the injection-area is simply a faulty injection or a bent needle. In some instances, the thin glass needle tip was not completely centered in the drill hole, causing the needle shaft to hit the edge of the drill hole, bending it upon entry into the brain. The subventricular zone is fairly close to the hippocampus, so it cannot be excluded that the needle tip ended up in this location accidentally. However, the needle tract and tissue damage caused by the pressure of the infusion are invisible in this particular section, making it difficult to establish exactly where the nanoparticles were delivered. That said, a faulty injection in one hemisphere does not explain the absence of GFP-positive cells in the other sections.
As was mentioned before, the infusions can wreck havoc on the GCL. The DG is a highly encapsulated structure, allowing it to resist great deals of pressure. Although it is not the case in all the injected brains, it often happened that the pressure in the infusion system builds up without releasing any fluid, before the pressure suddenly becomes too much and a large amount of the it is released at once and with force. In many cases this may cause one of the DG blades to burst and consequentially causes the nanoparticle-mixture to leak out, thus possibly explaining the absence of GFP-positive neurons. However, the same is true for the brains of the animals that were incubated with HNRK/DNA for only two days and these do show GFP-positive neurons. Either way, the possibility that the technique used is this destructive to the area intended for delivery is worth discussing. If the nanoparticles are to become successful vectors for gene transfer, the method of delivery will need to be thoroughly adjusted. It may be as simple as replacing parts of the infusion system,
such as the Hamilton syringe. However, circumventing surgical procedures altogether would be the optimum solution. Perhaps advancements in the field will one day allow for brain-targeting gene therapy with safer alternatives such as capsules or simple subcutaneous injections.
A more plausible explanation for the absence of GFP-positive cells is that the site of infusion is so severely damaged that an innate immune response is triggered and the cells in the area discard any GFP-positive neurons in the process. GFP may have been expressed in the first few days of the one-week incubation period (as in the two-day incubations), but were cleared before the brains were sectioned. This would also explain the survival of the GFP-positive cell in the subventricular zone, since it may be much less resistant to pressure and thus less prone to damage upon infusion.
There is another possible explanation why no GFP-positive neurons were found in the DG after one week. Perhaps the nanoparticle releases some type of pathogenic signal that prompts the brain to get rid of any cell expressing its load. Two days may not be enough time to clear the DG of all GFP-expressing cells, but one week is. Since the purpose of the HNRK-protein is to deliver nucleic acids safely and result in transgenic expression for a period longer than two days (if it is to be therapeutically beneficial), this possibility will need to be thoroughly investigated.
The results of the HNRK/miR-124 injections show that after two days incubation, a fair amount of Cy3-positive spherical and complex-like structures can be found around tissue tears in the sections close to but not actually in the DG. The observation of Cy3-positive complexes in the tissue next to nuclei is as expected. Why they are only visible in damaged areas is not entirely known, an explanation for this finding has yet to be found, but it could be the result of pressure bursts during infusions. Especially since poor perfusions but mainly pressure may be the cause of tissue disruption. Furthermore, the edges of the tears are ragged and tangled; possibly more capable of capturing the complexes after sectioning than the intact tissue.
It is important to note that the HNRK-proteins coupled with miRs is do not crosslink during perfusions so chances of the complexes washing out soon after the brains are sliced are realistic. In order to hamper the nanoparticles from washing out of the tissue, the brains of the one-week incubation animals were cut 10 µm thicker and a decently sized sample was mounted straight after slicing and without any washing. Again, some Cy3-positive complex-like structures were visible, but far fewer in number and in torn areas only.
There are at least three possible explanations why so few of the HNRK/miR-124 complexes are
visible after an incubation period of one week. One, the nanoparticles’ lifespan is not long enough and they simply disintegrate before the brains are extracted. Two, as was speculated about the HNRK/DNA-complexes, the HNRK-nanoparticles may release some type of pathogenic signal that prompts the body to destroy it. Three, the miRs may have been degraded by cellular systems after one week.
All three explanations are plausible and with the current data, not one can be excluded. It makes sense that the complexes and their content degrade somehow, as they are not transgenically expressed like GFP, but are merely present in the cytoplasm, exposed to numerous volatile cellular systems. However, there seems to be one exception. The data clearly shows one red cell in the infra pyramidal layer of the DG after an incubation period of two days. Contrary to all other Cy3-positive areas, this cell does not appear just to have a nanoparticle inside, it appears Cy3-positive in its entirety. How this cell came to be Cy3-positive is somewhat of a mystery and future studies should try to replicate this result.
Lastly, neurogenesis in the DG was examined two days and one week after injecting HNRK/Cy3-miR-124. MiR-124, though it has many possible targets, is one of the factors that is believed to play a role in downregulating the GR, which if correct could dampen the GR’s inhibiting effect on hippocampal neural stem cell proliferation and maturation. MiR-124 is naturally present in the DG, but overexpressing it via HNRK-complex injections was speculated to be capable of altering neurogenesis.
From the viable samples it appears as though there are more DCX-positive neurons in the SGZ of the DG one week after injecting the complexes. However, the infusions caused such extensive damage in many of the brains that there weren’t enough samples to accurately quantify this possible difference; In many of the samples one of the DG blades is destroyed and with it the cells that would have been DCX-positive. In other samples where both DG blades are intact, the infusion may have caused enough damage to allow for an assembly of highly autofluorescent microglia that completely overshadows the presence of any DCX-positive neurons. Nevertheless, there seems to be a difference and replication of this experiment with a larger sample and should allow for quantification.
An increase in the number of DCX-positive cells in the DG one week after HNRK/miR-124 injections is not the only alteration (though unproven) in AHN. There appear to be quite a few DCX-positive neurons located in the outer granular cell layer. SGZ stem cells migrate in the direction of the molecular layer as they mature and DCX also stains migrating cells, but usually when the neurons have nearly reached the molecular layer they are
too mature to be expressing DCX.
If miR-124 indeed exerts a dampening effect on the GR, an increase in the number of DCX-positive cells could be explained by an increase in stem cell proliferation as a result of downregulated GR. However, since miR-124 potentially has many targets, changes in GR expression, stem cell proliferation and/or number of DCX-positive cells may be due to any of these possible targets. The same is true for the DCX-positive cells found in the outer molecular layer. A direct cause cannot (yet) be given, but it is plausible that a crowded SGZ filled with immature neurons pushes the maturing neurons to migrate either faster or sooner in order to make room for more proliferating stem cells. Whichever explanation is closer to the truth, it is a result that needs further investigation as do many other results discussed in this paper.
The HNRK nanoparticle is a clever construction that takes from nature and viral bodies only those characteristics that could potentially make it a highly successful and safe vector for gene transfer capable of targeting many different tissues. However, as this paper indicates, it is also a vector that is in much need of further investigation. Though the research on the use of viral vectors for gene transfer is still continuous, the potential for non-viral particles to safely replace the use of viruses is evident and worthy of investigation.
5. Conclusion
With only a few positive cells it is too soon to suggest that the HNRK-protein complex holds the potential to replace viral vectors for the delivery of therapeutic nucleic acids in gene therapy. However, its flexibility and adaptability could allow it to target many different tissues and deliver many different therapeutically beneficial materials. The results press for much-needed future experiments in order to optimize the construction, delivery, lifespan, and safety of this protein-only nanoparticle.
Acknowledgement
The author would like to thank Dr. C. P. Fitzsimons, Dr. J. Domingo-Espín, and M. Schouten for their extended supervision during this project. Special appreciation is also expressed to G. Meerhof for his excellent technical assistance and to Dr. A. Korosi for acting as co-supervisor and UvA representative.
References
[1] M. Giacca, S. Zacchigna. Virus-mediated gene delivery for human gene therapy. Journal of Controlled Release 161 (2012) 377-388.
[2] N. Bessis, F.J. Garcia-Cozar, M-C. Boissier. Immune responses to gene therapy vectors: influence on vector function and effector mechanisms. Gene Therapy 11 (2004) S10-S17. [3] M.L. Edelstein, M.R. Abedi, L. Wixon. Gene
therapy clinical trials worldwide to 2007 an update. Journal of Gene Medicine 9 (10) (2007) 833-842. [4] E.P. Brown. Regulation of B-cell responses by
Toll-like receptors. Immunology 136 (2012) 370-379. [5] N. Shi, W.M. Pardridge. Noninvasive gene targeting
to the brain. PNAS 97 (13) (2000) 7567-7572. [6] M. Wu, T. Sherwin, W.L. Brown, P.G. Stockley.
Delivery of antisense oligonucleotides to leukemia cells by RNA bacteriophage capsids. Nanomedicine 1 (1) (2005) 67-76.
[7] A. Arís, A. Villaverde. Modular protein engineering for non-viral gene therapy. TRENDS in Biotechnology 22 (7) (2004) 371-377.
[8] A. Arís, J.X. Feliu, A. Knight, C. Coutelle, A. Villaverde. Exploiting viral cell-targeting abilities in a single polypeptide, non-infectious, recombinant vehicle for integrin-mediated DNA delivery and gene expression. Biology and Bioengineering 68 (6) (2000) 689-696.
[9] A. Arís, A. Villaverde. Molecular organization of protein-DNA complexes for cell-targeted DNA delivery. Biochemical and Biophysical Research Communications 278 (2000) 455-461.
[10]E. Mahon, A. Salvati, F. B. Bombelli, I. Lynch, K.
A. Dawson. Designing the
nanoparticle-biomolecule interface for “targeting and therapeutic delivery”. Journal of Controlled Release 161 (2012) 164-174.
[11]J. Domingo-Espín, E. Vazquez, J. Ganz, O. Conchillo, E. García-Fruitós, J. Cedano, U. Unzueta, V. Petegnief, N. Gonzalez-Montalbán, A.M. Planas, X. Daura, H. Peluffo, N. Ferrer-Miralles, A. Villaverde. Nanoparticulate architecture of protein-based artificial viruses is
supported by protein-DNA interactions.
Nanomedicine 6 (6) (2011) 1047-1061.
[12] D.P. Bartel. MicroRNAs: Target recognition and regulatory functions. Cell 136 (2009) 215-233.
[13] J. Yu, K. Chung, M. Deo, R.C. Thompson, D.L. Turner. MicroRNA miR-124 regulates neurite outgrowth during neuronal differentiation. Experimental Cell Research 314 (2008) 2618-2633. [14] W. Filipowicz, S.N. Bhattacharyya, N. Sonenberg.
Mechanisms of post-transcriptional regulation by microRNAs: Are the answers right? Nature 9 (2008) 102-114.
[15] L.P. Lim, N.C. Lau, P. Garrett-Engele, A. Grimson, J.M. Schelter, J. Castle, D.P. Bartel, P.S. linsley, J.M. Johnson. Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature 433 (2005) 769-773.
[16] N.A. Datson, N. Speksnijder, J.L. Mayer, P.J. Steenbergen, O. Korobko, J. Goeman, E.R. de Kloet, M. Joëls, P.J. Lucassen. The transcriptional response to chronic stress and glucocorticoid receptor blockade in the hippocampal dentate gyrus. Hippocampus 22 (2012) 359-371.
[17] M. Schouten, M.R. Buijink, P.J. Lucassen, C.P. Fitzsimons. New neurons in aging brains: Molecular control by small non-coding RNAs. Frontiers in Neuroscience 6 (25) (2012).
[18] E. Vreugdenhil, C.S.L. Verissimo, R. Mariman, J.T. Kamphorst, J.S. Barbosa, T. Zweers, D.L. Champagne, T. Schouten, O.C. Meijer, E. R. de Kloet, C.P. Fitzsimons. MicroRNA 18 and 124a down-regulate the glucocorticoid receptor: Implications for glucocorticoid responsiveness in the brain. Endocrinology 150 (5) (2009) 2220-2228. [19] A. Garcia, B. Steiner, G. Kronenberg, A.
Bick-Sander, G. Kempermann. Age-dependent
expression of glucocorticoid- and mineralocorticoid receptors on neural precursor cell populations in the adult murine hippocampus. Aging Cell 3 (2004) 363-371.
[20] N.R. Nichols, M. Zieba, N. Bye. Do glucocorticoids contribute to brain aging? Brain Research Reviews 37 (2001) 273-286.
[21] G. Kempermann, S. Jessberger, B. Steiner, G. Kronenberg. Milestones of neuroral development in the adult hippocampus. TRENDS in Neurosciences 27 (8) (2004) 447-452.
[22] C.P. Fitzsimons et al. Molecular Psychiatry (2012). [23] M. Sundberg, S. Savola, A. Hienola, L. Korhonen,
D. Lindholm. Glucocorticoid hormones decrease proliferation of embryonic neural stem cells through ubiquitin-mediated degradation of cyclin D. The Journal of Neuroscience 26 (20) (2006) 5402-5410. [24] A. Korosi, E.F.G. Naninck, C.A. Oomen, H.
Krugers, C.P. Fitzsimons, P.J. Lucassen. Early-life stress mediated modulation of adult neurogenesis and behavior. Behavioral Brain Research 227 (2012) 400-409.
[25] J.S. Snyder, A. Soumier, M. Brewer, J. Pickel, H.A. Cameron. Adult hippocampal neurogenesis buffers stress responses and depressive behavior. Nature 476 (2011) 458-462.
[26] R.J.A.H. Eendebak, P.J. Lucassen, C.P. Fitzsimons. Nuclear receptors and microRNAs: Who regulates the regulators in neural stem cells? FEBS Letters 585 (2011) 717-722.
[27] G. Kronenberg, I. Kirste, D. Inta, S. Chourbaji, I. Heuser, M. Endres, P. Gass. Reduced hippocampal neurogenesis in the GR+/- genetic mouse model of depression. European Archives of Psychiatry & Clinical Neuroscience 259 (2009) 499-504.
[28] C.K. Barha, S. Brummelte, S.E. Lieblich, L.A.M. Galea. Chronic restraint stress in adolescence differentially influences hypothalamic-pituitary-adrenal axis function and adult hippocampal neurogenesis in male and female rats. Hippocampus 21 (2011) 1216-1227.
[29] V.M. Heine, S. Maslam, J. Zareno, M. Joëls, P.J. Lucassen. Suppresed proliferation and apoptotic changes in the dentate gyrus after acute and chronic stress are reversible. European Journal of Neuroscience 19 (2004) 131-144.
[30]G. Paxinos, K.B.J. Franklin, The Mouse Brain in Stereotaxic Coordinates, second ed., Academic Press, San Diego, 2001.
[31]A. Cetin, S. Komai, M. Eliava, P.H. Seeburg, P. Osten. Stereotaxic gene delivery in the rodent brain. Nature Protocols 1 (6) (2006) 3166-3173.
[32] H. Peluffo, L. Acarin, A. Arís, P. Gonzalez, A. Villaverde, B. Castellano, B. González. Neuroprotection from NMDA excitotoxic lesion by Cu/Zn superoxide dismutase gene delivery to the postnatal rat brain by a modular protein vector. BMC Neuroscience 7 (35) (2006).
