A Combination of Tri-Leucine and Angiopep-2 Drives a Polyanionic Polymalic Acid Nanodrug Platform Across the Blood–Brain Barrier

Liron L Israel; Oliver Braubach; Anna Galstyan; Antonella Chiechi; Ekaterina S Shatalova; Zachary Grodzinski; Hui Ding; Keith L Black; Julia Y Ljubimova; Eggehard Holler

doi:10.1021/acsnano.8b06437

. Author manuscript; available in PMC: 2020 Nov 4.

Published in final edited form as: ACS Nano. 2019 Jan 16;13(2):1253–1271. doi: 10.1021/acsnano.8b06437

A Combination of Tri-Leucine and Angiopep-2 Drives a Polyanionic Polymalic Acid Nanodrug Platform Across the Blood–Brain Barrier

Liron L Israel ^†,^§, Oliver Braubach ^†,^§, Anna Galstyan ^†, Antonella Chiechi ^†, Ekaterina S Shatalova ^†, Zachary Grodzinski ^†, Hui Ding ^†, Keith L Black ^‡, Julia Y Ljubimova ^†, Eggehard Holler ^†,^*

PMCID: PMC7641102 NIHMSID: NIHMS1054603 PMID: 30633492

Abstract

One of the major problems facing the treatment of neurological disorders is the poor delivery of therapeutic agents into the brain. Our goal is to develop a multifunctional and biodegradable nanodrug delivery system that crosses the blood—brain barrier (BBB) to access brain tissues affected by neurological disease. In this study, we synthesized a biodegradable nontoxic β-poly(L-malic acid) (PMLA or P) as a scaffold to chemically bind the BBB crossing peptides Angiopep-2 (AP2), MiniAp-4 (M4), and the transferrin receptor ligands cTfRL and B6. In addition, a trileucine endosome escape unit (LLL) and a fluorescent marker (rhodamine or rh) were attached to the PMLA backbone. The pharmacokinetics, BBB penetration, and biodistribution of nanoconjugates were studied in different brain regions and at multiple time points via optical imaging. The optimal nanoconjugate, P/LLL/AP2/rh, produced significant fluorescence in the parenchyma of cortical layers II/III, the midbrain colliculi, and the hippocampal CA1–3 cellular layers 30 min after a single intravenous injection; clearance was observed after 4 h. The nanoconjugate variant P/LLL/rh lacking AP2, or the variant P/AP2/rh lacking LLL, showed significantly less BBB penetration. The LLL moiety appeared to stabilize the nanoconjugate, while AP2 enhanced BBB penetration. Finally, nanoconjugates containing the peptides M4, cTfRL, and B6 displayed comparably little and/or inconsistent infiltration of brain parenchyma, likely due to reduced trans-BBB movement. P/LLL/AP2/rh can now be functionalized with intra-brain targeting and drug treatment moieties that are aimed at molecular pathways implicated in neurological disorders.

Keywords: blood—brain barrier, polymalic acid, optical analysis, Angiopep-2, targeting peptide, treatment of neurological disorder

Graphcial Abstract

graphic file with name nihms-1054603-f0009.jpg

Insufficient delivery across the blood—brain barrier (BBB) prevents many preclinical drugs from reaching their intended targets and results in low efficiencies of conventional drug treatments for neurological disorders.^1,2 Drug delivery across the BBB of healthy subjects is especially challenging because an intact BBB, in contrast to the disease-compromised BBB, is largely drug impenetrable. Yet, the early treatment of neurological disease is paramount to the success of drug therapies, given that most diseases have a poor prognosis once they reach advanced stages. Moreover, early drug treatment of neuroinflammation and neurodegeneration may prevent the deterioration of the BBB altogether and could maintain its protective ability of excluding infiltrating cytokines and toxins.³ Our focus was thus directed toward the development of an advanced nanoconjugate drug delivery platform that crosses the BBB of healthy mice.

Certain nanoparticles deliver drugs by encapsulation, but they have large diameters and limited BBB penetration. Such particles also have limited biodegradability and can result in insoluble toxic depositions. In addition, nonspecific drug effects may arise due to spontaneous release of drug cargo, via drug diffusion or nanoparticle dissolution.^4–6 Antibody-based drugs, on the other hand, penetrate the BBB⁷ and have provided promising results in the laboratory⁸ as well as in preclinical treatment of neurological disorders, including Alzheimer’s disease.^9,10 However, antibody-based therapeutics, even when humanized, can trigger systemic immune responses,^11,12 which complicate long-term treatment perspectives. Moreover, antibody molecules are large, limit cargo capacity and deep tissue penetration, and are thus limited in delivery to recipient cells. As an alternative to these drug designs, we chose to develop a mini nanoconjugate around the naturally occurring polymeric scaffold, β-poly(L-malic acid) (PMLA).¹³ PMLA is completely biodegradable to carbon dioxide and water, nontoxic, and non-immunogenic. PMLA also carries abundant carboxyl groups that can be conjugated with multiple targeting and therapeutic moieties, ultimately constituting a nanodrug platform that can carry several diverse functional moieties.^14–18

Certain molecules are transported across the BBB via highly selective endogenous transport mechanisms. For example, the low-density lipoprotein receptor pathway (LRP-1) enables the bidirectional movement of low-density lipoproteins across the BBB.¹⁹‘²⁰ LRP-1-mediated blood-to-brain transport occurs when suitable ligands bind to and become internalized by LRP-1 in the vascular endothelium. After internalization, LRP-1 bound ligands are transcytosed across the BBB and into the brain parenchyma. A synthetic LRP-1 peptide ligand, Angiopep-2, was identified by Demeule et al.²¹ (AP2; TFFYGGSRGKRNNFKTEEY). The transport of AP2 saturates at high concentrations and is inhibited by other LRP-1 ligands,²² confirming transcytosis of AP2 across the BBB. Several laboratories are now using AP2 as a conduit to shuttle nanodrugs across the BBB. The available results are very promising,^22,23 but each laboratory has conjugated AP2 to different platforms, which makes comparisons among different drug transporters difficult. Another class of peptides enhances BBB drug penetration via the transferrin receptor (TfR) pathway. The TfR pathway imports iron into the brain and is critically involved in maintaining cerebral iron homeostasis. TfRs are selectively expressed on endothelial cells of brain capillaries and thus provide a promising conduit for selective drug delivery into the brain.²⁴ An iron-mimic peptide ligand for TfR-mediated drug delivery, cTfRL (CRTIGPSVC-NH2, cyclic, S—S bonded) was isolated via phage display and has been shown to deliver cargo into brain tumors.²⁵ Another TfR ligand, B6 (CGHKAKGPRK), has been described under similar circumstances.^26–28 Finally, MiniAp-4 (M4; H-[Dap] KAPETAL D-NH₂), is a cyclic peptide that was derived from bee venom. This peptide is capable of translocating proteins and nanoparticles across a human cell-based BBB model, but the exact mechanism of how this occurs remains elusive.²⁹ None of the above-mentioned BBB penetrating peptides have been designed to carry reversibly bound cargoes and thus lack inherent therapeutic value(s). It is therefore important to study these peptides as components of cargo delivery molecules and, especially, to study how conjugation with a second peptide or other functional groups influences their BBB penetration abilities.

We present a nanoconjugate that is based on a malic acid polymer and the above-mentioned BBB-penetrating peptides. Our drug platform additionally carries multiple trileucine (LLL) residues, which display pH-responsive lipophilicity and promote endosomal escape of PMLA bound agents once they are internalized and part of the intracellular endosomal pathway.³⁰ Endosomal escape is crucial for intracellular drug treatment since it permits cytoplasmic drug delivery. Rhodamine was conjugated to the drug platform to enable its visualization via optical imaging. The nanoconjugate did not contain further targeting peptides at this stage, given the need to first develop a neutral delivery platform that penetrates the BBB and distributes over multiple brain regions that are potentially affected by neurological disorders.

RESULTS AND DISCUSSION

Synthesis and Characterization of Nanoconjugates.

Thirteen PMLA-based nanoconjugates were synthesized as candidates for trans-BBB drug delivery (Table 1). Conjugates contained 1% rhodamine (rh) unless mentioned otherwise. In nine of the nanoconjugates, 40% of malyl residues carried LLL and 10% mercapto ethylamine (MEA) groups that were attached at pendant carboxylates of the polymer (Figure 1B, C). Conjugates were made via a preconjugate synthesized by the activation of PMLA pendant carboxylates applying the DCC/NHS method (Figure 1A, B), to attach LLL and 2-mercapto ethylamine (cysteamin, MEA, Figure 1B, C) in a follow-up reaction.^31,32 MEA was then used to form thioethers with peptide-PEG-maleimide and rhodamine-maleimide (Figure 1C). Prior to conjugation by amidation, carboxyl groups had been activated via DDC/NHS chemistry (Figure 1A₁, B). Peptide moieties AP2, AP7, and B6 were activated by controlled sulfhydryl alkylation with maleimide-PEG3400 (or PEG2000)-maleimide (Figure 1A₂) and were then loaded onto 2% of the polymer malyl groups through thioether formation engaging the second PEG-maleimide group (Figure 1B, C). Peptides M4 and cTfRL were activated by amidation with an amino-reactive PEG-maleimide linker, maleimide-PEG2000-SCM (maleimide-PEG-succinimidyl carboxyl methyl ester), at their N-terminus. These small exocyclic peptides did not contain a terminal cysteine (unlike AP2, AP7 and B6). PEG3400 or PEG2000 linkers were used to facilitate peptide interactions with biological targets. For one conjugate, a 4% load of AP2 was added by a correspondingly increased amount of activated AP2 with PMLA. Unreacted sulfhydryl on each polymer was masked by a reaction with PDP (3-(2-pyridyldithio) propionic acid.

Table 1.

Nanoconjugate Nomenclature, Formula, ζ Potential, Calculated Molecular Mass, and SEC-HPLC Retention Time

nanoconjugate	components	ζ potential (mV)	grouped by ζ potential	molecular mass (g/mol)	SEC-HPLC retention time (rt, min)
P (nanoconjugate backbone)	PMLA	−22.8	Group 1	50,000	7.52
P/rh	PMLA/rh(1%)	−22.9		56,000	7.23
P/LLL/rh	PMLA/LLL(40%)/rh(1%)	−16.5	Group 2	11,4000	7.10
P/cTfRL/rh	PMLA/cTfRL(2%)/rh(1%)	−15.2		81,000	7.05
P/M4/rh	PMLA/M4(2%)/rh(l%)	−14.6		80,000	7.10
P/AP2/rh	PMLA/AP2(2%)/rh(1%)	− 11.5		106,000	7.18
P/LLL/AP2/rh^a	PMLA/LLL(40%)/AP2(2%)/rh(1%)	− 11.6	Group 3	164000	7.215
P/LLL/M4/rh^c	PMLA/LLL(40%)/M4(2%)/rh(1%)	−10.4		138,000	7.20
P/LLL/cTfRL/rh^d	PMLA/LLL(40%)/cTfRL(2%)/rh(1%)	−9.58		139,000	7.22
P/LLL/AP2/D1/rh^f	PMLA/LLL(40%)/AP2(2%)/D1(2%)/ rh(1%)	−7.44	Group 4	207,000	7.05
P/LLL/B6/rh^e	PMLA/LLL(40%)/B6(2%)/rh(1%)	−6.1		153,000	7.5
P/LLL/AP2/M4/rh	PMLA/LLL(40%)/AP2(2%)/M4(2%)/ rh(1%)	−5.5		188,000	7.05
P/LLL/AP7/rh^b	PMLA/LLL(40%)/AP7(2%)/rh(1%)	−5.48	Group 5	165,000	7.27
P/LLL/AP2(4%)/rh	PMLA/LLL(40%)/AP2(4%)/rh(1%)	−2.2	Group 5	217,000	7.47

Open in a new tab

Peptide sequence: TFFYGGSRGKRNNFKTEEYCNH₂.

Peptide sequence: TFFYGGSRGRRNNFRTEEYCNH₂.

Peptide sequence: H-[Dap] KAPETAL D-NH2, monocyclic lactam bridged.

Peptide sequence: CRTIGPSVC-NH2, monocyclic disulfide bridged.

Peptide sequence: CGHKAKGPRK.

Peptide sequence: qshyrhispaqvc.

Figure 1. — Synthetic route for PMLA/LLL/Angiopep-2/rhodamine (P/LLL/AP2/rh) nanoconjugate production. Biosynthesized PMLA was activated using DCC/NHS chemistry to conjugate LLL and MEA (A). MEA moiety was used to bind AP2 peptide conjugated to a PEG linker (B) *via* a maleimide-thiol reaction. Rhodamine was attached in the same manner prior to capping with PDP (C).

Both LLL and MEA reactions were monitored during conjugation using a ninhydrin-color assay on TLC, and thiols were quantified using Ellman’s reagent (see Methods). Typically, after polymer activation with 2-mercapto ethylamine, active SH groups were detected at the level of 8.2–10% malic acids of the polymer. Malic acid content was analyzed via a malic acid dehydrogenase assay.¹⁵ This analysis was corrected by accounting for 15% of L-malic acid content converted to nonsubstrate D-isomer during sample preparation. Constituents of nanoconjugates and reagents were subjected further to analysis by FTIR, SEC-HPLC elution profile, ζ potential, and ¹H NMR. Formula, calculated molecular mass, ζ potential, and SEC-HPLC retention times are summarized in Table 1. Rhodamine-free conjugates were also characterized by their hydrodynamic diameter using dynamic light scattering (DLS; Table 2).

Table 2.

Hydrodynamic Diameter and PDI for Selected Non-Fluorescent Nanoconjugates Measured by DLS (Volume Mode)

nanomolecule	hydrodynamic diameter (nm)	PDI
PMLA	3.68	0.89
P/LLL	2.68	0.50
P/AP2	5.93	0.79
P/LLL/AP2	4.45	0.39

Open in a new tab

Grouping of Conjugates with Different ξ-Potentials.

Table 1 reveals important structure–function relationships, in which nanoconjugates display similar ξ potentials based on their chemical formulas. In Group 1, P and P/rh have negative ξ potentials of −22.8 mV and −22.9 mV, which are the lowest potentials listed in Table 1. This highlights the negative charge of PMLA and indicates that the binding of PMLA with rhodamine does not change the ξ potential. In Group 2, the conjugates P/LLL/rh, P/cTfRL/rh, P/M4/rh, and P/AP2/rh have ξ potentials from −16.5 mV to −11.5 mV. Addition of LLL or a single peptide moiety thus decreases the negative ξ potential. Conjugates in Group 3 were P/LLL/AP2/rh, P/ LLL/M4/rh, and P/LLL/cTfRL/rh and had intermediate ξ potentials from −11.6 mV to −9.58 mV. Except for P/LLL/ AP2/rh, Group 3 ξ potentials are considerably less negative than those of Groups 1–2, indicating that the addition of a LLL and a peptide moiety further decreased the nanoconjugate negative ξ potential. This observation continues in Group 4, consisting of P/LLL/AP2/D1/rh, P/LLL/B6/rh and P/LLL/ AP2/M4/rh, with ξ potentials ranging from −7.44 mV to −5.5 mV. Finally, P/LLL/AP7/rh and P/LLL/AP2(4%)/rh were the least negatively charged conjugates with ξ potentials from −5.48 mV to −2.2 mV. Thus, nanoconjugate composition correlates with molecular charge, and with some exceptions, it appears that additional LLL or peptide moietie increased the conjugates ξ potential. As will be shown below, Group 3–4 conjugates and P/LLL/B6/rh showed good BBB penetration, suggesting that conjugates with intermediate ξ potentials and chemical compositions may be most suited for trans-BBB movement.

Composition.

The compositions of rhodamine-labeled peptide nanoconjugates were verified by diode array detection (DAD) monitored SEC-HPLC, quantitative photometry, FTIR analysis, and ¹H NMR. The SEC-HPLC profiles of constituents and intermediates in the synthesis of the nanoconjugate P/LLL/AP2/rh are shown at wavelength 220 nm in Figure S1: monitoring the elution of PMLA (Figure S1A), the elution of AP2 (pink) and AP2-PEG3400-maleimide (blue in Figure S1B), the elution of AP2-PEG3400-maleimide (Figure S1B), and of preconjugate (P/LLL/MEA; Figure S1C). These profiles are depicted together with the profiles in the DAD mode in the figure insets. Figure S2A shows the elution of P/LLL/AP2/rh in the DAD mode (retention time, nm wavelength, absorbance). The absorbance profile depicts the ultraviolet spectrum of the conjugate P/LLL/AP2/rh residue in the 200–300 nm wavelength region and an additional absorbance at 570 nm for rhodamine. The absorbances of the conjugate and rhodamine are in the same time elution position, demonstrating the physical unity of P/ LLL/AP2/rh. The absorbance ratios at 220 nm wavelength and 570 nm wavelength, compared with the absorbance ratios measured separately for nanoconjugate and rhodamine, confirm the composition of the nanoconjugate-rhodamine. The composition of P/LLL/AP2/rh was also verified following the consumption of AP2-PEG3400-maleimide during the synthesis of the conjugate. The completion of the nanoconjugate is evident from Figure S2B, showing the profile of P/ LLL/AP2/rh and an arrow in the would-be position in case of incompletely reacted AP2-PEG3400-maleimide. This reaction was stoichiometrically followed by measuring the consumption of a solution standardized for activated AP2 (AP2-PEG3400-maleimide standard curve shown in Figure S2C). Further evidence for the authenticity of P/LLL/AP2 was obtained via FTIR spectra. The spectrum in Figure S2D contains several distinctive peaks (see arrows) which are attributed to both the preconjugate (blue) and the pristine AP2 peptide (green), while some peaks were shifted or decreased in intensity. A prominent shift of the peak at 3050 cm⁻¹ in the preconjugate (P/LLL/MEA) spectrum to 3057 cm⁻¹ in the P/LLL/AP2 spectrum is seen as well as other changes at the lower frequencies of 1040, 1104, and 950 cm⁻¹ (see arrows).

We extended our structural validation studies via ¹H NMR analysis of PMLA, the preconjugate (P/LLL/MEA), maleimide-PEG3400-maleimide, the AP2 peptide, and P/LLL/AP2 (Figures S3–5). PMLA ¹H NMR (Figure S3A) shows a broad acid OH peak at 8.5 ppm and two main peaks at 3–3.1 ppm attributed to the CH₂ hydrogens in α position to the backbone carbonyl (marked a) as well as in the region 5.5–5.6 ppm attributed to CH hydrogens in β position to the carbonyl (marked b).^{33 1}H NMR of P/LLL/MEA (Figure S3B) shows that these peaks changed: A prominent new peak was the methylene signal of the tripeptide LLL in the region 0.8–0.86 ppm (marked c). Each LLL unit has 18 protons in that position (not resolved). If the integral of the peak marked a for hydrogen in α position to the carbonyl (2 protons per malic acid unit, now shifted to 2.8 ppm, marked a) was calculated and divided by the integral of peak c (normalized to the number of protons per unit), one obtains the ratio 2.9 which suggests a 87% yield for the LLL conjugation with PMLA. Furthermore, the thiol hydrogen (marked d) and multiple amide hydrogens (marked e, NMR not shown) could be identified. Figure S4 depicts the ¹H NMR spectrum of the AP2 peptide (Figure S4A, 0.4–5.6 ppm and Figure S4B, 6.0–12.2 ppm). As the AP2 concentration on the nanoconjugate is low (2% of the malic acid content), only some prominent peaks could be identified, that is, the aromatic phenylalanine peak (AP2 contains 3 phenylalanine residues) at 7.2–7.4 ppm (Figure S4B; marked f-f′). Furthermore, the ethylene ¹H NMR peak of PEG3400 in maleimide-PEG3400-maleimide is identified in Figure S5A. The ethylene hydrogen peak is seen at 3.5 ppm. The ¹H NMR spectrum of P/LLL/AP2 (Figure S5B) contains a large peak at 3.5 ppm that can be attributed to the maleimide-PEG3400 linker ethylene hydrogens of PEG3400 (peak g). The spectrum also contains a small peak at 7.2–7.4 ppm that is attributed to AP2 phenylalanine (peak f-f′) and a trileucine methylene peak (peak c). Hence, our SEC-HPLC and ¹H NMR analyses confirm the design of the nanoconjugates AP2-PEG3400-maleimide, P/LLL/MEA, P/ LLL/AP2, and P/LLL/AP2/rh.

Finally, to ensure that the different conjugate moieties did not affect the intensity of the rhodamine signal, that is, via electrochemical and electrostatic forces, we measured the fluorescence emission of the nanoconjugates P/AP2/rh, P/LLL/AP2/rh, P/LLL/cTfRL/rh, and P/LLL/AP7/rh in solution. We observed a 20–30% higher fluorescence intensity for the LLL-containing nanoconjugates in comparison with P/AP2/rh. We assume that this effect reflected the hydrophobicity of LLL side chains, but we rule out that this affected the outcome of fluorescence measurements in brain tissues (as below).

Biodistribution and Drug Toxicity Studies.

We first examined the biodistribution and potential toxicity of P/LLL/ AP2/rh following i.v. injections into normal mice. Figure S6 shows the biodistribution of P/LLL/AP2/rh at two hours (A) and four hours (B) after i.v. injections of 0.068 μmol/kg drug. The P/LLL/AP2/rh accumulated primarily in the kidney and liver at both time points, and nanoconjugate fluorescence in the brain was considerably lower than in either of those organs. However, a preferential uptake of P/LLL/AP2/rh into the brain was evident when we compared nanoconjugate fluorescence to the heart, lung and spleen (Figure S6A₁, B₁).

Potential toxicity of our nanoconjugate was studied in an experiment in which we injected mice four consecutive times with P/LLL/AP2/rh at 0.548 μmol/kg over the course of 11 days. The 0.548 μmol/kg drug concentration was the highest dose used in our study, and we reasoned that potential toxic effects should be maximally visible if the conjugate is administered at this concentration. As shown in Figure S7A, we found no indication of toxicity based on animal weight. A two-way repeated measures ANOVA failed to detect a treatment effect (F(1,4)=0.1887, p = 0.686), and we therefore conclude that no abnormal weight loss occurred in mice receiving drug injections. Moreover, behavioral abnormalities were not observed. Mice displayed normal levels of activity, normal posture, regular respiration, and reacted to handling as would be expected. This is in accordance with our previous work, which examined toxicity of PMLA-based conjugates in more detail.¹⁶ Taken together, we conclude that our drug is nontoxic and that it distributes as expected from prior studies.³⁴

Characterization of Nanoconjugate Fluorescence in Brain Parenchyma and Vascular Labeling.

We studied the BBB penetration and brain distribution of our nanoconjugates via optical imaging of fluorescence emitted by their rhodamine moiety. All imaging was conducted in fixed cryosections that were obtained from mice at various times after systemic i.v. injections. Two distinct patterns of fluorescence were observed, however only one could be ascribed to our nanoconjugate. Specifically, diffuse staining near the vasculature and in the perivascular space was exclusively visible using a rhodamine fluorescence imaging filter set (asterisks in Figure 2A₁_–₂). This type of labeling was not visible in PBS-injected animals, nor if other fluorescence excitation/emission filter sets were used. Diffuse perivascular staining therefore represents the rhodamine-labeled nanoconjugate(s). We also observed subcellular-sized particulate fluorescence in close vicinity of DAPI-stained nuclei. Examples of this kind of fluorescence are indicated with arrows in Figure 2A₁_–₂. Such particulate fluorescence was reproducible during imaging with multiple fluorescence excitation/emission filter sets, suggesting that it does not specifically represent the rhodamine signal from our nanoconjugate. Furthermore, particulate fluorescence was present in brain tissue of animals that were injected with PBS, as is shown in images of the hippocampal CA1 cellular layer (Figure 2B) and layers II/III of the cerebral cortex (Figure 2C). We thus rule out that particulate staining represents our nanoconjugate. Instead, we hypothesize that the fluorescent particles shown in Figure 2A₁–C are lipofuscin, which are intracellular metabolite and waste deposits in neurons.^35–37 Yet, we do not rule out that nanoconjugate fluorescence contributes to the lipofuscin signal (i.e. via degradation and accumulation of rhodamine in intracellular organelles), but we exclude this type of fluorescence from our analysis. A distinction between diffuse nanoconjugate fluorescence and lipofuscin has, to our knowledge, not been made, even though several studies have shown lipofuscin-like particulate staining patterns.^23,38 We believe that this distinction is a crucial precondition to obtaining accurate and reliable optical measurements of nanoconjugate fluorescence.

Figure 2. — Distinction of nanoconjugate vs lipofuscin fluorescence. Fluorescence in cryosectioned tissue from mice injected with either P/LLL/AP2/rh (A) or PBS (B, C). (A₁) Injection of P/LLL/AP2/rh produces diffuse fluorescence in the vascular and perivascular space (asterisks). This fluorescence is only visible if a rhodamine-carrying nanoconjugate is injected and if a rhodamine filter set is used for imaging. (A₂) Particulate fluorescence in proximity to DAPI-labeled nuclei is visible in the brain of P/LLL/AP2/rh injected mice (arrows). (B, C) Particulate staining is visible in the brains of PBS injected animals, indicating that particulate fluorescence is related to auto fluorescence (red; lipofuscin). No diffuse perivascular staining is observed in PBS injected mice. (D) Average diameters of lectin-labeled blood vessels for cortical layers II/III (green), the superior/inferior colliculi of the midbrain (red), and the hippocampal CA1-CA3 cellular layers (blue). Data are means and their standard errors from 20 randomly sampled vessels in 4 animals for each brain region. The vessel diameters were measured as the shortest distance between the luminal vessel walls and were 4–5 μm in every brain region.

We injected our mice with fluorescently labeled tomatolectin minutes before the termination of each experiment. This led to widespread fluorescent labeling of the cerebral vasculature. Tomato-lectin binds to glycoproteins on the luminal vascular endothelium³⁹ and allows visualization of the blood vessel luminal space. In the case of our experiment, lectin labeling allowed us to discriminate the inside vs outside of the vasculature, and to see if a nanconjugate had entered the brain: examples of a brain-permeating nanconjugate are indicated with asterisks in Figure 2A₁_–₂. Furthermore, tomato-lectin binds to glycoproteins that are preferentially located in the endothelium of capillary vessels,⁴⁰ and indeed, by measuring the average diameters of lectin-labeled blood vessels in multiple brain regions, we confirmed their average diameters to be within 5 μm (Figure 2D); this is within the previously reported size range of the cerebral microvascula-ture.⁴¹ We therefore conclude that tomato-lectin labeled vessels represent capillaries, and that any nanoconjugate outside of these capillaries had crossed the BBB to enter the brain parenchyma.

Concentration-dependent BBB Penetration of P/LLL/AP2/rh.

Table 1 lists 13 nanoconjugates that were studied for their ability to penetrate the BBB and distribute in the brain parenchyma. Of the 13 nanoconjugates, P/LLL/AP2/rh had the best BBB penetration. Figure 3A shows whole-brain epifluorescence imaging data from mice that were i.v. injected with PBS vs P/LLL/AP2/rh and sacrificed 120 min post-injection. Fluorescence in the PBS group was absent (Figure 3A₁), while incremental fluorescence was evident for brains injected with P/LLL/AP2/rh at 0.068 μmol/kg (Figure 3A₂) and 0.548 μmol/kg (Figure 3A₃). Figure 3B₁_–₄ are optical imaging data from cryosectioned brain tissue. The drug concentration is listed as the total concentration of each injected nanoconjugate, where conjugates contained 40% LLL, 2% peptide, and 1% rhodamine (Table 1). Tissue shown in Figure 3B₁_–₄ was counterstained with tomato-lectin to show the vasculature (red); the nanoconjugate is shown in gray. Injections of P/LLL/AP2/rh at increasing drug concentrations produced visibly more fluorescence in the brain parenchyma, as is shown for mice injected with 0.068 μmol/kg (Figure 3B₁), 0.137 μmol/kg (Figure 3B₂), 0.274 μmol/kg (Figure 3B₃), and 0.548 μmol/kg (Figure 3B₄).

The brain tissue permeation of the nanoconjugate was not uniform, and most of the nanoconjugate fluorescence was concentrated in the perivascular space, between 5 and 20 μm from the blood vessel wall. This is visible in Figure 3B₄ as strong nanoconjugate fluorescence (gray) near the blood vessels, but diminished fluorescence further away from the blood vessels. Figure 3C explores this relationship in a plot from all our measurements (for each condition: 4 mice, 4 sections with 20 random measurements each). All fluorescence intensity measurements were conducted with 10 μm²-sized regions of interest placed outside of tomato-lectin stained blood vessels (ROI in Figure 3B₁); the positions of these ROIs were then measured against the location of the nearest blood vessel wall to produce the scatterplot in Figure 3C. Fitting our data with a linear regression, indicated a fluorescence intensity decrease (slope) of −0.72 ± 0.15 for the 0.548 μmol/kg drug injection condition and −0.272 ± 0.07 for the 0.274 μmol/kg drug injection condition. This confirms that nanoconjugate tissue permeation is not uniform and that the drug concentration decreases with distance from the vasculature. However, based on significantly different y-intercepts, we can confirm more BBB penetration of P/LLL/AP2/rh following injections at higher drug concentrations: the y-intercept for the 0.548 yMmol/kg drug injection condition was 34.07 ± 2.3, 17.49 ± 0.8 for the 0.274 μmol/kg drug injection, and 6.342 ± 0.34 for drug injected at 0.068 μmol/kg.

The results described above are applicable to layers II/III of the cerebral cortex (Figure 3D₁), the superior/inferior colliculi of the midbrain (Figure 3D₂), and the CA1-CA3 cellular layers of the hippocampus (Figure 3D₃). Data shown in Figure 3B₁_–₃ are average nanoconjugate fluorescence intensity values and their standard errors: These were obtained from randomly sampled ROIs, irrespective of their location and distance from the vasculature (4 mice in each condition). Notably, Figure 3D₃ shows that fluorescence measurements in the cellular layers of the hippocampus were consistently lower than those in cortical and midbrain regions. The hippocampus is linked to the formation and maintenance of memories, is affected by neurodegenerative disease, and is thus a crucially important target for potential nanoconjugate therapies.^42–44 We hypothesize that the lower nanoconjugate fluorescence in the hippocampus is due to the comparatively small amount of vascular perfusion of this brain region. This result is presented in Figure S8, where we show that the area covered by blood vessels is less in the hippocampal CA1 region than in layers II/ III of the cortex or the midbrain colliculi (Figure S8B1). This results in an intervessel distance in the hippocampus of 59 μm, which is almost twice that of the cortex (32 μm) and midbrain colliculi (30 μm; Figure S8B2). Considering that P/LLL/AP2/ rh distributes preferentially within ~30 μm from the microvasculature (i.e., Figure 3C), we can argue that the reduced vascular access in the hippocampus may be responsible for its reduced drug perfusion. This issue can be partially resolved through drug injections at higher concentrations, as is shown by a significant dose-dependent increase of hippocampal nanoconjugate fluorescence in Figure 3D₃.

BBB Penetration Depends on Nanoconjugate Composition.

Having determined that P/LLL/AP2/rh enters the brain parenchyma in a concentration-dependent manner, we next turned our attention to the effects of LLL and AP2 on BBB penetration, whereby the concentrations of remaining LLL (40%), AP2 (2%) and rhodamine (1%) were held constant in each case. We first removed the LLL moiety, which resulted in P/AP2 (with 0% LLL). Figure 4A₁ vs 4A₂ show that P/LLL/AP2/rh penetrated the brain parenchyma better than P/AP2/rh. This is especially apparent in the perivascular space where much of the diffuse gray nanoconjugate fluorescence can be seen in the P/LLL/AP2/rh but not the P/AP2/rh condition. Corresponding fluorescence measurements from cortical layers II/III are summarized in Figure 4B₁ (red vs green data) and were significantly brighter for P/LLL/AP2/rh vs P/AP2/rh injected at 0.068 μmol/kg (Tukey: p < 0.0001), 0.137 μmol/kg (Tukey: p < 0.0001), and 0.274 μmol/kg (Tukey: p < 0.0001). Indeed, the fluorescence associated with P/AP2/rh was invariably lower across all imaged tissues. The same observations were made in the midbrain colliculi (Figure 4B₂) and the hippocampal CA1–3 cellular layers (Figure 4B₃). We thus conclude that P/AP2/rh owns little potential for BBB penetration.

Figure 4. — Nanoconjugate composition determines degree and locus of BBB penetration. (A_1–3) Optical imaging data showing nanoconjugate permeation of the cerebral cortex: Nanoconjugate fluorescence is gray, and the vasculature is red. Different nanoconjugates are indicated with different colors. (B) Average nanoconjugate fluorescence in layers II/III of the somatosensory cortex (B₁), the midbrain colliculi (B₂), and the hippocampal CA1–3 cell layers (B₃) as a function of nanoconjugate composition and concentration: P/LLL/AP2/rh is shown in red, P/AP2/rh in green, and P/LLL/rh in blue. Average nanoconjugate fluorescence measurements were obtained from 20 randomly sampled ROIs explicitly outside of the cerebral vasculature (4 mice with 4 images each, for each measurement). Statistical tests were conducted between nanoconjugate types (*e.g*., red vs green) within different concentrations. The results are indicated with asterisks where * = p < 0.01, ** = p < 0.001, and *** = p < 0.0001; the red lines show the concentration of P/LLL/AP2/rh against which each comparison was made.

We next examined if the removal of the AP2 moiety affected the BBB penetration of our nanoconjugate. P/LLL/rh (with 0% AP2) generated less fluorescence in brain parenchyma than P/LLL/AP2/rh (Figure 4A₁ vs 4A₃) at all concentrations tested (Figure 4B₁; red vs blue data). However, brain tissues from mice injected with P/LLL/rh were significantly more fluorescent than tissues from mice injected with P/AP2/rh (green vs blue in Figure 4B₁): More cortical fluorescence was associated with P/LLL/rh vs P/AP2/rh at 0.068 μmol/kg (Tukey: p < 0.01), 0.137 μmol/kg (Tukey: p < 0.0001), and 0.274 μmol/kg (Tukey: p < 0.0001). This observation was also made in the midbrain colliculi (Figure 4B₂) and in the hippocampal CA1–3 layers (Figure 4B₃). Thus, P/LLL/rh penetrates the BBB even without a peptide moiety. The addition of the AP2 peptide increases BBB penetration and, combined with LLL, produces the optimal formula P/LLL/ AP2/rh.

The original purpose of the LLL moiety was to act as an endosomal escape unit, which would eventually permit the release of nanoconjugate moieties during intracellular endosomolytic drug targeting.³⁰ Our present results clearly indicate that the LLL moiety, in conjugation with PMLA, also contributes to BBB penetration of the PMLA molecule, without the need of a BBB penetrating peptide. This mechanism may involve synergistic contributions of PMLA and LLL moieties to introduce a specific hydrophobic/ hydrophilic amphiphilic conjugate, which breaks the BBB. The interaction of P/LLL copolymers with membranes including their penetration has already been investigated in our laboratory.⁴⁵

An energy calculation in Figure S9B represents a short-segmented view of PMLA containing 16 malic acid residues and 4 LLL residues (red arrows Figure S9A), two hexa ethylene glycol oligomers conjugated via maleimide, and two propionyl disulfide moieties. The calculation shows that LLL-LLL groups on the polymer that locate next to each tend to form clusters in vacuum (arrows in Figure S9B) and would do so even more in an aqueous milieu under the driving force of hydrophobicity. Such intra-polymer hydrophobicity-driven associations are prone to impact the conformations of PMLA conjugates (compare Figure S9 with Figure S10), which could ultimately affect the associations of BBB-penetrating peptides with biological substrates. Even though our energy calculations are modeled in the absence of water, they already showed evidence for the clustering. The calculation in Figure S10B of a similar segment, however in the absence of conjugated LLL, does not favor clustering. Such structural effects of LLL substitutions are in accord with the following considerations: (1) Because of the hydrophilic nature and high negative charge (Table 1: Z potential of P/rh is −22.9 mV), PMLA conjugates with peptides do not readily approach and interact with negatively charged membranes on their way to penetrate BBB. The massive number of LLL-conjugations partially “neutralizes” the negative charges (Table 1, ξ-potential of P/LLL/rh = − 16.5 mV) and infers hydrophobicity through introducing the LLL-side chains. Both effects could promote the approach and interaction with cell membranes. (2) In the absence of LLL residues occupying extended stretches of PMLA, the negatively charged polymer is highly flexible and invites peptides for electrostatic interactions and hydrogen bonding with pendant carboxylates. Positively charged side chains of AP2 would favor association along the PMLA backbone, and important side chains would be unavailable for receptors binding in the BBB transcytosis pathway. Multiple LLL residues bound to carboxylates eliminate most of the negative charges next to the backbone and sterically inhibit backfolding so that important groups of the peptides could interact with receptors of the pathway. (3) Three-dimensional structures calculated for minimized energy (Figures S9 and S10) suggest that LLL-LLL cluster formation occurs and compacts nanoconjugate structure, hydrodynamic diameter and volume, number(s) of interchangeable conformations and promotes exposure of peptide residues. The results in Table 2 are in agreement with a reduction of the hydrodynamic diameter after LLL conjugation that is exemplified in the case of P/LLL, 2.68 nm vs PMLA, 3.68 nm and P/LLL/AP2, 4.45 nm vs P/AP2, 5.93 nm. (4) Furthermore, the polydispersity index (PDI) in Table 2 decreases concomitantly with LLL conjugation in the sequence PMLA (0.89) > P/AP2 (0.79) > P/LLL (0.5) > P/LLL/AP2 (0.39). In summary, LLL conjugation with PMLA favors BBB permeation by (l) optimizing the interactions of targeting peptides with receptors of transcytosis pathways, (2) reducing the diameter and volume of permeating nanoconjugates, and (3) increasing the conformational rigidity of the nanoconjugate.

Screening BBB-Penetrating Peptide Moieties.

The effect of LLL on BBB permeation was compared for the following peptides known to have BBB penetration activity: AP2,²² M4,²⁹ B6,^27,28 and cTfRL.²⁵ The results with respect to different brain regions are summarized in Figure 5. The nanoconjugate with optimal BBB penetration had the formula P/LLL/AP2/rh (see above). Replacing AP2 with M4 (Figure 5A₁_–₂; P/LLL/M4/rh) resulted in similar levels of nanoconjugate fluorescence if mice were injected with 0.068 μmol/ kg conjugate (red vs orange in Figure 5B; Sidak: p = 0.5749). However, P/LLL/M4/rh injected at 0.274 μmol/kg produced significantly less cortical fluorescence than P/LLL/AP2/rh (Figure 5B; Sidak: p < 0.0001). Yet, we measured essentially identical levels of P/LLL/M4/rh and P/LLL/AP2/rh fluorescence in both the midbrain colliculi and the hippocampal CA1–3 layers, regardless of the injected drug concentrations (red vs orange in Figure 5C, D). Hence, P/ LLL/M4/rh and P/LLL/AP2/rh appear to permeate the brain tissue with similar efficacies, but P/LLL/M4/rh shows regional selectivity and poor permeation of layers II/III of the cerebral cortex.

Figure 5. — Nanoconjugate peptide moiety screen. P/LLL/rh was equipped with different peptides to assess their role in BBB penetration. (A_1–3) Optical imaging data showing rhodamine-labeled nanoconjugate permeation of the cerebral cortex by P/LLL/rh conjugated to AP2 (A₁), M4 (A₂), and B6 (A3). Nanoconjugate fluorescence is gray and the vasculature is red. Average nanoconjugate fluorescence in layers II/ III of the (B) cerebral cortex, (C) the midbrain colliculi, and (D) hippocampal CA1–3 layers. (E) Nanoconjugate fluorescence measurements in cortical layers II/III for peptide combinations and altered loads of AP2 and rhodamine. Peptide identity is color coded and indicated on the side of the histogram. Statistical tests were conducted against each of the different concentrations of P/LLL/AP2/rh in each histogram and are indicated with asterisks where * = p < 0.01, ** = p < 0.001, and *** = p < 0.0001; the red lines show the concentration of P/LLL/ AP2/rh against which each comparison was made.

Brain tissue fluorescence resulting from injections of TfR ligand-conjugates were generally less than those obtained from injections with P/LLL/AP2/rh. P/LLL/B6/rh fluorescence was almost always less when compared to P/LLL/AP2/rh in the same brain region (red vs magenta in Figure 5B–D). The only exception was P/LLL/B6/rh associated fluorescence in the midbrain colliculi, which was similar to that measured for P/LLL/AP2/rh injected at 0.274 μmol/kg (compare red vs magenta in Figure 5C; Sidak: p = 0.2499). The midbrain colliculi contain the highest density of cerebral microvasculature (Figure S8B₁), which may facilitate drug entry into the brain tissue. This could also explain why P/LLL/AP2/ rh, P/LLL/M4/rh, and P/LLL/B6/rh showed essentially the same levels of nanoconjugate fluorescence in the midbrain colliculi if injected at a high enough concentration (0.274 μmol/kg).

A nanoconjugate containing the transferrin (Tf) ligand cTfRL, when injected at 0.068 jMmol/kg (P/LLL/cTfRL/rh), produced fluorescence intensity measurements comparable to B6 (Figure 5B–D). Because results for P/LLL/cTfRL/rh were redundant with P/LLL/B6/rh, we dismissed the nanoconjugate from further experiments. Finally, and as an additional control to our experiments in Figure 4, we synthesized P/M4/rh, P/cTfRL/rh (i.e., different peptides but no LLL), and P/rh. All three conjugates, P/M4/rh, P/cTfRL/rh, and P/rh had poor BBB penetration and produced extremely low fluorescence in brain parenchyma (not shown). These results confirm our observation that LLL is required for BBB penetration, regardless of which peptide the conjugate carries.

In another set of experiments, we asked if nanoconjugates with peptide combinations and modified peptide loads traverse the BBB more efficiently (Figure 5E). A nanoconjugate carrying a combination of AP2 and M4 (P/LLL/AP2/M4/ rh), each of which was promising on its own, permeated the cortex slightly more than nanoconjugates that contained a single peptide. This is shown in Figure 5E, where P/LLL/AP2/M4/rh injected at 0.137 μmol/kg produced slightly, but not significantly more fluorescence than P/LLL/AP2/rh at the same concentration (red vs turquoise; Sidak: p = 0.0617). P/LLL/AP2/M4/rh therefore failed to display a significant sum of BBB-crossing effects by each peptide. Moreover, P/LLL/AP2/M4/rh has a reduced cargo capacity due to higher occupancy of the polymer platform by peptides; this limits its future development and loading, and we decided against further development of P/LLL/AP2/M4/rh.

We also asked if an increase in the same peptide load on the nanoconjugate could lead to enhanced BBB penetration. Thus, far, all conjugates carried 2% peptide content. In Figure 5E, we show that a doubling of the peptide load, P/LLL/AP2(4%)/rh, resulted in decreased BBB penetration (red vs beige; Sidak: p < 0.0154). While unexpected, this result may be explained in the following way. An increased peptide load means that the conjugate carried more peptides on the same chain. This leaves the peptides in closer proximity to each other (~17 loading units for 4% vs ~8 units for 2%), which could facilitate intramolecular interactions between peptides, and ultimately reduce their steric availability and interaction with biological receptors. This “crowding effect” would reduce BBB penetration, and per our results, we therefore conclude that 2% AP2 peptide is the optimal load for our nanoconjugate delivery system.

It could be argued that residues LLL and AP2 (or the other peptide moieties) in the conjugate recognized independent routes of BBB permeation and that the observed BBB penetration efficacy was the sum of these independent contributions. Although the results in Figures 4 and 5 cannot be reconciled with sums of independent contributions of LLL and peptide specific pathways, we measured the results of injected P/LLL/AP7/rh as a control. AP7 differs from AP2 by the replacement of two lysine groups with arginine (TFFYGGSRGRRNNFRTEEYCNH₂), which reportedly impairs peptide interactions with endothelial LRP-1 receptors.^21,22 P/LLL/AP7/rh permeated cortical brain tissue but produced significantly less fluorescence than P/LLL/AP2/rh, both injected at 0.137 μmol/kg (red vs gray in Figure 5E; Sidak: p < 0.0001). This result confirms a substantial role for authentic AP2 to enable trans-BBB movement of the nanoconjugate. Together with our other findings, we demonstrate that nanoconjugate transport through the BBB depends on peptide identity, peptide load, and interaction with other nanoconjugate moieties (i.e., LLL).

Multiple BBB crossing peptides have been studied as conduits to deliver nanodrugs across the BBB. The most promising peptides appear to be those that interact with LRP-1 or TfR transport mechanisms, while others rely on other/unknown transport systems. Each peptide has its own merits and applications, but thus far, there had not been much systematic comparison of peptide BBB-crossing efficacies when attached to a nanocarrier. Reports about a BBB crossing peptide focused on variation of carriers, and this makes it difficult to compare peptide behaviors in a standardized setting (but see ref 46). Furthermore, peptides recently developed de novo were compared in their performance with AP7, which appears to have little BBB-crossing ability (see above). Our study offers a systematic comparison between different BBB-crossing peptides conjugated to the same carrier molecules, polymalic acid, and P/LLL. We conclude that AP2, M4, and B6 enhance BBB-penetration of P/LLL/rh.

Our results apply to the brain of healthy mice. It is instructive to consider that the performance of certain peptides may differ in pathological conditions in which the BBB is altered. For instance, the TfR route may be effective for drug delivery into brain tumors. Gliomas overexpress TfR in their vascular endothelium, and this may aid drug-tumor penetration and delivery via enhanced TfR transport.⁴⁷ In contrast, the LRP-1 route is critically linked to amyloid β (Aβ) protein homeostasis and clearance in normal brains, but is aberrant in Alzheimer’s disease.⁴⁸ LRP-1 receptors are downregulated in Alzheimer’s disease,^49–51 which may not only impair Aβ protein clearance but may also reduce LRP-1-dependent drug transport across the BBB. While this issue remains unexplored, it underscores the need to carefully design BBB penetrating drugs appropriately for each disease. Our results provide a comprehensive database on the effects of different BBB penetrating peptides on P/LLL/rh transport into the disease-free brain.

Nanoconjugate Pharmacokinetics in Blood and Brain.

We administered fluorescent nanoconjugates via i.v. injections and drew blood at 15 to 480 min to measure the decay and pharmacokinetics of P/LLL/AP2/rh, P/LLL/rh and P/AP2/rh in serum. The nanoconjugate serum concentrations, as shown in Figure 6A, were calculated from calibration curves that were derived from fluorescence measurements of nanoconjugates with known concentrations. P/LLL/AP2/rh and P/ LLL/rh had serum half-lives of 76.7 and 119 min, respectively. The half-lives were determined by fitting serum fluorescence measurements with single exponential decay functions: The fluorescence decay of P/LLL/AP2/rh provided a good fit of r² = 0.9361, while the decay of P/LLL/rh was fit at r² = 0.715. The decay functions differed significantly (extra sum of squares F-test: F = 8.281; p = 0.0002), which confirms distinct serum clearances for P/LLL/AP2/rh and P/LLL/rh.

Having established the serum pharmacokinetics of our nanoconjugates, we next asked if these measurements could be replicated via optical imaging of brain slices. To do this, we took optical measurements of vascular P/LLL/AP2/rh fluorescence in a large cerebral blood vessel, the sagittal sinus, from 30 to 240 min after i.v. injection (Figure 6B; ex vivo data from separate specimens). The images in the top row show the nanoconjugate (red) and the sinus vasculature (green); images in the bottom row show only the P/LLL/AP2/rh -associated fluorescence (gray). Figure 6C shows the fluorescence intensity profile for the sagittal sinus and adjacent brain tissue, as indicated with a yellow line in Figure 6B. The fluorescent nanoconjugate is clearly concentrated in the sinus at 30 min post i.v. injection, while subsequent time points show a progressive loss of vascular fluorescence (Figure 6C). We next calculated the “vascular fluorescence” by measuring the difference between fluorescence peaks and the fluorescence in surrounding parenchyma (see dashed lines in Figure 6C) and then plotted the resulting vascular fluorescence alongside our actual serum measurements (Figure 6A black curve). The resulting “optical” measurements were converted to μmol/mL units via normalization to one time point of serum P/LLL/ AP2/rh at 120 min; the remaining time points were then converted using the same ratio (0.267 nmol/mL serum concentration for the 0.068 jMmol/kg injection at 120 min). Surprisingly, we obtained almost the exact same half-life for the optically measured P/LLL/AP2/rh vasculature clearance (optical half-life = 73.2 min), and no difference was detected between optical and serum-fitted functions (extra sum of squares F-test: F = 0.3327; p = 0.8017). This result confirms the validity of optical imaging to understand nanoconjugate pharmacokinetics in the brain.

The decay of nanoconjugate-associated fluorescence in the parenchyma of cortical layers II/III is summarized in Figure 6D. P/LLL/AP2/rh fluorescence was maximal at 30 min after i.v. injections and decreased only slightly by 120 min (Figure 6D; ANOVA: F = 531.6; p < 0.0001), despite a significant decrease of serum drug (Figure 6A). At 240 min after i.v. injection, we could not distinguish nanoconjugate fluorescence from background fluorescence, suggesting that P/LLL/AP2/rh is eliminated from the brain 4 h after administration (Figure 6D). The same observations were made in the midbrain colliculi and hippocampal CA1–3 cell layers (not shown). P/ LLL/rh associated fluorescence in the cortex was lower than that of P/LLL/AP2/rh throughout the 30—480 min time period (ANOVA: F = 268.5; p < 0.0001), but the overall fluorescence buildup and clearance followed the same pattern as seen with P/LLL/AP2/rh (Figure 6D; blue). Finally, P/AP2/rh associated fluorescence was lower than that of other nanoconjugates, but again followed a similar trajectory of fluorescence decay (Figure 6D; green).

Our analysis shows that fluorescent P/LLL/AP2/rh disappears from the serum and brain tissue beginning at 4 h after i.v. injection. This is a beneficial drug property and indicates that the nanoconjugate is broken down or expelled within a predictable time frame. Breakdown, for example, by spontaneous degradation and expulsion via retrograde diffusion through the BBB minimizes potential nonspecific interactions with biological tissues that could be harmful and toxic. Retrograde clearance through the BBB into the vascular system is consistent with the reported reversibility and homeostatic function of the LRP-1 BBB transport. In this sense, acute experiments or diagnostic imaging applications generally prefer agents that have a rapid onset and clearance, unless they bind and/or are taken up by targeted extracellular material (e.g., Aβ plaques) or brain cells. Our nanoconjugate platform could produce agents of this category. On the contrary, an exaggerated instantaneous breakdown and/or retrograde diffusion back through the BBB could render our nanoconjugate inefficient and “too-short” acting to be considered as an agent for drug therapy. Finally, it is important to note that our pharmacokinetic measurements were obtained from tissues of mice injected with 0.068 μmol/kg nanoconjugate concentration. The clearance of drugs injected at higher concentrations was not studied but could be slower; this is likely, considering that we observe more drug accumulation in the parenchyma after administering high drug concentrations (see Figure 3).

Estimating Drug Concentration in the Brain Parenchyma Based on Optical Ratio Measurements.

We used our imaging results to estimate the actual concentration of P/ LLL/AP2/rh in the parenchyma of cortical layers II/III at 120 min after the drug injection. This was accomplished by measuring P/LLL/AP2/rh fluorescence in the cerebral vasculature and surrounding parenchyma with identical ROI, followed by a calculation of the vessel/parenchyma fluorescence ratio (Figure 7A: 4 mice, 4 sections with 10 measurements for each condition). The images in Figure 7A₁_–₂ (bottom panel) demonstrate this procedure in tissue from mice injected with 0.068 μmol/kg and 0.274 μmol/kg of P/LLL/AP2/rh, respectively. The fluorescence ratios that resulted from our measurements are summarized in Figure 7B. A significant reduction in the vasculature/parenchyma fluorescence ratio was observed for P/LLL/AP2/rh injections of 0.274 μmol/kg (ANOVA: F = 11.36; p < 0.0001; Tukey: p < 0.0001) and 0.548 μmol/kg (Tukey: p = 0.0018) when both concentrations were compared to 0.068 μmol/kg. The result indicates reduced blood-to-brain transport at high concentrations of P/LLL/AP2/rh, presumably due to LRP-1 pathway saturation.

Figure 7. — Estimation of the nanoconjugate concentration in nmol/ mL of i.v. injected P/LLL/AP2/rh in the parenchyma of the cerebral cortex. (A_1–2) Imaging data showing cortical layer II/III tissue from mice injected with P/LLL/AP2/rh at 0.068 μmol/kg (A₁) and 0.274 μmol/kg (A₂). The top images show cell nuclei (red), vasculature (green) and P/LLL/AP2/rh (gray). The lower panels show only P/LLL/AP2/rh-associated fluorescence. Yellow and orange ROI indicate how fluorescence was measured in the blood vessels vs the parenchyma: The selected ROIs were close to each other but not ultimately in regions of highest nanoconjugate staining. (B) Fluorescence ratios in vasculature/brain parenchyma. Asterisks indicate statistical significance in Tukey test conducted for the 0.068 μmol/kg drug injection condition, where ** = p < 0.001 and *** = p < 0.0001. (C) Estimated P/LLL/AP2/rh concentration in cortical brain parenchyma. Asterisks indicate statistical significance in Tukey test conducted against the 0.068 μmol/kg drug injection condition, where ** = p < 0.001 and *** = p < 0.0001.

We next estimated the P/LLL/AP2/rh concentration in brain parenchyma by multiplying each of our vasculature/ parenchyma ratio measurements with known serum drug concentrations at 120 min post-injection (0.267 nmol/mL for the 0.068 μmol/kg injection as per Figure 6A). The resulting drug parenchyma concentrations are plotted in Figure 7C. A significant overall increase in the P/LLL/AP2/rh concentration was observed in the cortical layer II/III parenchyma (ANOVA: F = 166.3; p < 0.0001); the lowest concentration was estimated at 0.114 ± 0.001 nmol/mL for the 0.068 μmol/ kg injection; the highest parenchyma concentration was 0.74 ± 0.01 nmol/mL for the 0.548 μmol/kg injection. Based on these estimates, we conclude that P/LLL/AP2/rh traverses the BBB efficiently and that upward of 40% of serum drug can be detected in the brain within 120 min after i.v. administration. The parenchymal drug concentration is highest when measurements are made proximal to the vasculature; it decreases with increasing distance from the vasculature.

Our concentration estimates reflect the passage of drug from the vasculature to the brain and vice versa, if reversibility in the LRP-1 pathway is assumed. On that basis, we could tentatively assume that vascular and proximal parenchymal concentrations are similar (40% and higher vascular concentrations as reference). The similar concentrations could indicate that for P/LLL/AP2/rh, the BBB does not function as a very efficient barrier, at least in the relatively low concentration range that we investigated here. This speculative assumption needs to be additionally confirmed, however, we believe that our optical estimate is accurate and indicative of the good BBB penetrating ability of our nanoconjugate P/LLL/AP2/rh.

BBB Penetration of a D1 Carrying Version of P/LLL/ AP2/rh.

We next asked if P/LLL/AP2/rh can be equipped with a bioactive drug, and if so, if the BBB penetration behavior of the resulting nanoconjugate changes. To do this, we conjugated P/LLL/AP2/rh to a small Aβ₄₂-binding D-amino acid peptide, D1 (qshyrhispaqvc), resulting in P/LLL/ AP2/D1/rh. D1 was identified by mirror image phage display selection with Aβ₄₂ as target;⁵² the peptide binds Â₄₂ in vitro and in vivo^53,54 and thus qualifies as a potential diagnostic or drug treatment compound for Alzheimer’s disease-associated amyloidosis. The BBB penetration of fluorescently labeled P/LLL/AP2/rh and P/LLL/AP/D1/rh was nearly indistinguishable from one another. This is shown in Figure 8A₁_–₂, which is imaging data from brain tissue loaded with 0.274 μmol/kg of each conjugate and obtained 2 h after i.v. drug injection. As above, the vasculature is shown in red and the nanoconjugate in gray. Figure 8B₁ shows that P/LLL/AP2/rh and P/LLL/AP2/D1/rh penetrated the parenchyma of cortical layers II/III almost identically, with no statistical difference observed between conditions (two-tailed t test: t = 1.871; p = 0.0621). In contrast, P/LLL/AP2/D1/rh associated fluorescence was comparatively brighter in the midbrain colliculi (Figure 8B₂: two-tailed t test: t = 4.203; p < 0.0001), while fluorescence intensities in the hippocampus were overall lower but not significantly different between drug injection conditions (Figure 8B₃: two-tailed t test: t = 0.6793; p = 0.4978). We thus demonstrate the P/LLL/AP2/rh can cross the BBB even after the addition of a drug cargo; this is an important precondition for further development of this nanoconjugate as a potential therapeutic agent.

Figure 8. — BBB penetration of a nanoconjugate carrying D1 peptide. P/LLL/AP2/rh was loaded with a D1 peptide drug load to assess potential effects on BBB penetration. (A_1–2) Optical imaging data showing nanoconjugate permeation of the cerebral cortex for P/LLL/ AP2/rh (A₁) and P/LLL/AP2/D1/rh (A₂). Comparison of the average fluorescence intensity of the two nanoconjugates in layers II/III of the cerebral cortex (B₁), the midbrain colliculi (B₂), and the hippocampal CA1–3 layers (B₃). In each brain region, the intensities are similar for the two conjugates and even significantly increased for P/LLL/AP2/D1/rh in the midbrain colliculi (B₂). Statistical results are indicated as *** = p < 0.0001.

P/LLL/AP2/rh Is Not Internalized by Neurons or Glia.

We have previously shown that a trileucine carrying PMLA conjugate is internalized and released into the cytoplasm of cancer cells in vivo.³¹ Seeking a similar result, we analyzed the potential uptake of P/LLL/AP2/rh in neurons and glia. Figure S11A₁, B₁ depicts cortical layers II/III from animals that were injected with PBS and P/LLL/AP2/rh, respectively. The injected PBS or drug is shown in red and separately in Figure S11A₂, B₂. Neurons were counterstained with anti-Neun and are shown in yellow; the vasculature is shown in green in S11B₁. The asterisks in Figures S11B₁_–₂ show the locations of multiple pairs of neurons, and inspection of the drug only image in Figure S11B₂ indicates that no drug accumulated near these neurons (but see presumed lipofuscin particles in Figure S11B₂). Likewise, the arrows in Figures S11B₁_–₂ point toward an area adjacent to a blood vessel: This area has a high P/LLL/AP2/rh concentration, but a nearby neuron appears to be devoid of P/LLL/AP2/rh labeling. Similar observations were made for P/LLL/AP2/D1/rh, and we conclude that our nanoconjugate is not internalized by neurons at levels that are detectable with fluorescence microscopy.

In addition, we could not unambiguously detect internalized P/LLL/AP2/rh in glial cells. These data are shown in Figure S12A₁, B₁, which show glia (green) in the subiculum of mice injected with PBS and P/LLL/AP2/rh, respectively. As above, PBS or drug is shown in red and separately in Figure S12A₂, B₂. The vasculature is visible based on glial staining in S12B₁. Each set of arrows in Figure S12A₁, B₁ points toward glial cells and their processes. In Figure S12A₂, B₂, these same arrows point toward weak fluorescence that is visible in the rhodamine imaging channel. Crucially, this weak fluorescence was visible in both, PBS and drug injected animals, and we found no difference in the fluorescence intensity between these conditions (not shown). We therefore rule out that glia internalized appreciable levels of P/LLL/AP/rh under the given conditions. We made the same observation for P/LLL/AP2/D1/rh.

We conclude that P/LLL/AP2/rh and P/LLL/AP/D1/rh are not internalized by neurons or glia but emphasize that our observation applies to the experimental conditions described here. Certain modifications to our experimental protocol may, in the future, lead to another result. For example, the data described above were obtained from mice that were sacrificed 2h after the drug injection. This is a relatively short time, and it is possible that longer incubation of the drug and/or repeat administration may produce measurable accumulations of drug in neurons and/or glia. Furthermore, injected drug concentrations will almost certainly affect drug internalization. Finally, by equipping our nanoconjugates with specific targeting molecules, we should, in principle, be able to selectively target neurons and glia, and subtypes thereof. This is the next step, for example, by conjugation with specific ligands having high neuron and/or glia affinities in our efforts to develop P/LLL/ peptide as a drug delivery platform.

CONCLUSION

We present a macromolecular nanoconjugate platform for trans-BBB drug delivery. Our strategy builds on previously published peptides to design PMLA-based drug platforms, which offer simultaneous delivery of a variety of targeting and pharmacologically active components. We show that PMLA/LLL/peptide/rh interactions critically determine BBB passage, and we investigate in detail how our nanoconjugates distribute in the brain. In addition, we reveal that a moiety of inherent hydrophobic structure, LLL, critically influences and enhances brain delivery, especially in areas with high blood vessel density such as the midbrain colliculi. Although speculative, our results suggest that the BBB in healthy mice, for our lead nanoconjugate and under applied conditions, may not constitute an impenetrable barrier and that it can deliver μmolar and higher concentrations of covalently bound drug for pharmaceutical treatment.

Neurological disorders affect brain regions differently, and almost every disease can be attributed to specific malfunctions in a brain region. A detailed knowledge of nanoconjugate behavior in different brain regions is crucial for drug development, and such information is provided here. With P/LLL/AP2/rh, we functionalized only 50% of the carboxylic acids, leaving us with additional sites to further equip the nanoconjugate with targeting and drug treatment moieties. Such research is currently under way in our laboratory.

METHODS

Materials.

Highly purified poly(β-L-malic acid; 50000 g/mol) was prepared from the culture broth of Physarum polycephalum as previously described.¹⁵ The peptides Angiopep-2-cys (AP2) (TFFYGGSRGKRNNFKTEEYCNH2) containing an added C-terminal cysteine, Angiopep-7-cys (AP7) (TFFYGGSRGRRNNFR-TEEYCNH₂) containing an added C-terminal cysteine, B6 (CGHKAKGPRK), M4 (H-[Dap] KAPETAL D-NH₂; monocyclic lactam bridged), cTfRL (CRTIGPSVC-NH₂; monoc²yclic disulfide bridged), and D1 (qshyrhispaqvc) were custom synthesized by AnaSpec (Fremont, CA, USA). Rhodamine-maleimide (Rhodamine Red C2 maleimide) was purchased from ThermoFisher Scientific (Canoga Park, CA, USA). Maleimide-PEG3400-maleimide and maleimide-PEG-succinimidyl carboxyl methyl ester (maleimide-PEG2000-SCM) were purchased from Creative PEGWorks (Durham, NC, USA). Trileucine (LLL) was ordered from Bachem (Torrance, CA, USA). N,N’-Dicyclohexylcarbodiimide (DCC), N-hydroxysuccinimide (NHS), trifluoroacetic acid (TFA), cysteamine (or 2-mercaptoethylamine, MEA), dithiothreitol (DTT), deuterated acetone and dimethylsulfoxide (DMSO), dimethylformamide (DMF), 3-(2-pyridyldithio)propionic acid (PDP), triethylamine (NEt₃), and 5,5′-disulfanediylbis(2-nitrobenzoic acid) (DTNB) were obtained from Sigma-Aldrich (St. Louis, MO, USA). PD-10 columns were purchased from GE Healthcare (Chicago, IL, USA).

Calculation of Molecular Masses.

The molar mass of PMLA has been routinely measured by SEC-HPLC in phosphate buffered saline pH 7.0 using polystyrenesulfonate as molecular mass standards. The standards resemble the structure of PMLA only remotely, but reflect the polyanionic character at pH 7.0, and otherwise fail when comparing PMLA and chemical derivatives which differ in number and kind of structural modifications. This is mirrored by the polymer hydrodynamic radius after ligand substitutions that could not be standardized.¹⁶

The molecular mass of nanoconjugates was calculated on the basis of their constituents’ masses and stoichiometry. For example, the molar mass of PMLA/LLL(40%)/MEA(l0%) is composed of (l) the molar mass of PMLA at 50,000 g/mol corresponding to the number of 50,000/116 = 431 (100%) malyl residues. (2) The molar mass of tripeptides LLL (40%) attached to 40% malyl residues corresponds to 0.40 X 431 X 357.5 = 61,633 g/mol. (3) The molar mass of 2-mercapto ethylamine (MEA, 10%) attached to 10% malyl residues corresponds to 0.10 × 431 × 77 = 3319 g/mol. The resulting molar mass of the conjugate is 50,000 g/mol + 61,633 g/mol + 3319 g/mol = 114,952 g/mol. After correction for water liberated by the condensation corresponding to 0.5 × 431 × 16 = 3448 g/mol, the molar mass of P/LLL/MEA is calculated as 111,504 g/mol.

Synthesis of PMLA/LLL(40%)/MEA(10%) (LLL Containing Pre-conjugate).

Hereafter, we refer to the percent loading of LLL and other substituents with reference to the total content of malic acid residues in the polymer PMLA. The polymer (40 mg of 116 g/mol monomer, 0.345 μmol) was dissolved in 400 μL acetone at ambient temperature. A mixture of N-hydroxy succinimide (NHS, 115 g/mol, 40 mg, 0.345 μmol) and N,N’-dicyclohexylcarbodiimide (DCC, 206 g/mol, 74 mg, 0.36 μmol) dissolved in 400 μL of DMF was added dropwise. After 2 h, a mixture of trileucine (LLL, 357.4 g/mol, 49.3 mg, 0.138 μmol) and trifluoroacetic acid (TFA, 114 g/mol, d = 1.489 g/mL, 12.7 μL) in 200 μL DMF was added (in portions of 20, 25, 30, 35, 40, 45, and 50 μL in 10 min intervals). Every addition was followed by triethylamine (NEt₃) in DMF (101.2 g/mol, d = 0.73 g/ mL, i.e., 26.65 μL in total of 200 μL DMF given in portions of 15, 20, 25, 30, 35, 40, and 45 μL). The reaction extent was monitored using TLC (n-BuOH/Ô/AcOH 5/1/1) and ninhydrin reaction. After reaction termination, dithiothreitol (DTT, 7 mg, 154.25 g/mol, 0.045 μmol) in 50 μL of DMF was added, followed by 2-mercapto ethylamine (MEA, 113.61 g/mol, 3.92 mg, 0.035 μmol, in 10.8 μL DMF) and NEt₃ (4.8 μL, 1 equiv to MEA). The reaction was monitored using TLC and ninhydrin reaction. After reaction termination, 1.2 mL (identical to reaction volume) of sodium phosphate buffer (150 nM, pH 6.8) was added, and the mixture stirred for 2 h and centrifuged to separate from precipitation. The product was purified over a PD-10 column and characterized by SEC-HPLC (7.2 min retention time, 220 nm wavelength). After lyophilization, the product was stored as a white fibrous material.

Synthesis of PMLA/MEA(10%) (Preconjugate Not Containing LLL).

PMLA (10 mg of 116 g/mol monomer, 0.164 mmol) in 300 μL acetone received dropwise a mixture of NHS, 115 g/mol, 9.6 mg, 0.083 mmol, 50% of PMLA malic acid content and N,N-DCC, 206 g/ mol, 17.7 mg, 0.086 mmol 50% of PMLA malic acid content dissolved in 500 μL of DMF, followed by 15 mg of DTT, 154.25 g/mol, 0.097 mmol in 38 μL of DMF, then MEA, 113.61 g/mol, 1.9 mg, 0.017 mmol, in 7.8 μL DMF, and NEt₃, 2.3 μL, 1 equiv to MEA. The reaction was monitored using TLC (n-BuOH/H₂O/AcOH 5/1/1, ninhydrine reaction). After reaction termination, 0.8 mL (identical to reaction volume) of sodium phosphate buffer (150 nM, pH 6.8) was added, the mixture stirred for an additional 2 h and was centrifuged to discard precipitated salts, and the liquid phase purified over a PD-10 column followed by SEC-HPLC analysis (retention time 7.1 min at 220 nm). After lyophilization, white fibrous material was obtained.

General Procedure for Thiol Quantification using Ellman’s Reagent.

Five mg of DTNB (Ellman’s Reagent: 5,5-dithio-bis(2-nitrobenzoic acid)) was dissolved in 0.5 mL ethanol to a concentration of 10 mg/mL. Preconjugate (2 mg of either P/LLL/ MEA or P/MEA) was dissolved in 0.4 mL ethanol to a concentration of 5 mg/mL. Preconjugate solution, 50 μL, was added to 50 μL of the DTNB solution, followed by 900 μL phosphate buffer at pH 8.5. The mixture was stirred for 30 min and A₄₁₂ measured using a photometer (Flexstation, Molecular Devices, Sunnyvale, CA, USA). DTNB, 50 μL, diluted in 950 μL phosphate buffer pH 8.5 was used as reference. The read-out A₄₁₂ was then converted to concentration using the coefficient 14,150 (M cm)⁻¹. The result for experimental loading of P/LLL/MEA and P/MEA with MEA was 8.2–10% (% refers to total malic acid of the polymer).

Synthesis of Angiopep-2-PEG3400-maleimide.

At ambient temperature, maleimide-PEG3400-maleimide (3400 g/mol, 7.4 mg, 2.2 μmol, 1.05 equiv) was dissolved in 500 μL of phosphate buffer 100 nM including 2 mM EDTA at pH 6.3. Cysteine modified Angiopep-2 (TFFYGGSRGKRNNFKTEEYC), 2403.7 g/mol, 5 mg, 2.08 μmol, 1 equiv, dissolved in 500 μL phosphate buffer pH 6.3 was added dropwise. The reaction monitored by HPLC was completed after 1 h. The lyophilized product, dissolved in 10 mg/mL phosphate buffer (pH 6.3), was used for the reaction with PMLA preconjugate, followed by SEC-HPLC analysis (retention time 8.2 min at 220 nm wavelength). Angiopep-7-PEG3400-maleimide (SEC-HPLC retention 8.25 min at 220 nm), B6-PEG-maleimide (SEC-HPLC retention 7.92 min at 220 nm), and D1-PEG-maleimide (SEC-HPLC retention 8.0 min at 220 nm) were synthesized in the same manner.

Synthesis of Peptide-PEG2000-maleimide.

At ambient temperature, maleimide-PEG2000-SCM (maleimide-PEG-succinimidyl carboxyl methyl ester, 2000 g/mol, 3.5 mg, 1.75 μmol, 1.05 equiv) was dissolved in 250 μL of DMF. TfR ligand (932 g/mol, 1.5 mg, 1 equiv, 1.6 μmol) in 250 μL DMF was added, followed by 0.34 μL of NEt₃ (101.2 g/mol, d = 0.73 g/mL, 2.4 μmol, 1.5 equiv). The reaction was monitored using SEC-HPLC (usually overnight), and 0.1 μL of NEt₃ was added in case the reaction was not progressing. The reaction was purified on PD-10 column, analyzed by SEC-HPLC, and lyophilized. MiniAp-4-PEG2000-maleimide was synthesized in the same manner, using the N-terminus and the succinimidyl carboxyl methyl ester reaction for attachment.

Synthesis of PMLA/LLL(40%)/peptide(2%)/rh(1%).

PMLA/LLL-(40%)/MEA(10%) 4 mg of (260 g/mol, 15 μmol preconjugate monomer) was dissolved in 350 μL of phosphate buffer pH 6.3 and placed in a glass vial with a magnetic stirrer at ambient temperature. In order to achieve 2% loading, 1.78 mg of Angiopep-2-PEG3400-maleimide (5802.7 g/mol), or 2.07 mg of Angiopep-7-PEG3400-maleimide (5858.8 g/mol), or 0.87 mg cTfRL-PEG-maleimide (2817 g/mol), or 0.86 mg MiniAp-4-PEG2000-maleimide (2796 g/ mol), or 1.33 mg B6-PEG2000-maleimide (4480 g/mol) were each dissolved in phosphate buffer pH 6.3 to a 10 mg/mL concentration and added dropwise. After 1 h, the reactions monitored using SEC-HPLC (220 nm) were completed. Rhodamine-maleimide (0.104 mg for 1% loading, 680.79 g/mol, 0.149 μmol, 52 μL of 2 mg/mL solution in DMF) was loaded forming thioethers with the PMLA platform at pendant MEA-SH. The reaction was conducted in the dark and was monitored using SEC-HPLC. Extent of the conjugation was determined via rhodamine absorbance in the PMLA conjugate elution peak. After stirring for a further 1–2 h, 15 μL of PDP (10 mg/ mL solution in DMF) was added to cap the free SH groups. After stirring for an additional hour, the product was purified over a PD-10 column, analyzed, lyophilized, and stored at −20 °C.

Synthesis of PMLA/LLL/AP2(2%)/M4(2%)/rh(1%).

PMLA/LLL-(40%)/MEA(10%) 1 mg of (260 g/mol, 0.00375 mmol) was dissolved in 300 μL of degassed phosphate buffer (pH 6.3) and placed in a glass vial with a magnetic stirrer at ambient temperature. 2.3% (0.0862 μmol) or 0.512 mg of Angiopep-2-PEG3400-maleimide (5803 g/mol) peptide-PEG-maleimide was added dissolved in phosphate buffer pH 6.3 at 10 mg/mL concentration. The reaction was monitored using HPLC, typically for 1 h. Then, 0.215 mg (21.5 μL) of MiniAp-4-PEG2000-maleimide (2796 g/mol) was added. The reaction mixture was monitored using SEC-HPLC (typically 1 h reaction time), and, once completed, the glass vial was covered with aluminum foil, and rhodamine C2 was added (0.026 mg for 1% loading, 680.79 g/mol, 0.153 μmol, 13.05 μL of 2 mg/mL solution in DMF) and stirred for 1 h. Then, 15 μL of PDP (10 mg/mL solution in DMF) was added to cap the free SH groups. The reaction was stirred for an additional hour before purification using a PD-10 column (H2O as solvent), HPLC analysis, and lyophilization.

Preparation of PMLA/LLL/Angiopep-2(2%)/D1 (2%)/rh(1%).

Three mg of (260.7 g/mol monomer, 0.0115 mmol) was dissolved in 700 of phosphate buffer pH 6.3 and placed in a glass vial with magnetic stirrer at ambient temperature. Dl-PEG-maleimide was added (4924.8 g/mol, 1.13 mg/2%, 2.3 μmol, 10 mg/mL solution in phosphate buffer 6.3). The reaction was monitored using SEC-HPLC and typically lasted for 1 h. Then, 2% (0.23 μmol) or 1.33 mg of Angiopep-2-PEG-maleimide (5802.7 g/mol) was added dropwise (dissolved in phosphate buffer pH 6.3 to a 10 mg/mL concentration). The reaction was monitored using SEC-HPLC. Once completed, the glass vial was covered with aluminum foil, rhodamine-maleimide was added (0.0786 mg for 1% loading, 680.79 g/mol, amount added 0.115 μmol, 39.3 μL of a 2 mg/mL solution in DMF), and the reaction was monitored again using SEC-HPLC. Diode array detection (DAD) was used to detect rhodamine at 570 nm on the PMLA peak. Typically, the reaction was stirred for 1 h. Then, 15 μL of 3-(2-pyridyldithiopropionic acid, 10 mg/mL solution in DMF) was added to cap the free SH groups. The reaction was stirred for an additional hour before purification using PD-10 column, SEC-HPLC analysis, and lyophilization. ζ-Potentials, calculated molecular mass, and the SEC-HPLC retention times (rt, min) are listed in Table 1.

Calibration Curve Preparation for Peptide-PEG-maleimide.

Typically, 0.2, 1, 2, and 4 mg/mL solutions of peptide-PEG-maleimide were prepared from stock solutions of 10 mg/mL in phosphate buffer pH 6.3. Twenty μL of each sample was injected to the SEC-HPLC at the conditions described for the synthesis above. Then, the area under the curve (AUC) was measured for each concentration. The results were plotted using Microsoft Excel (Microsoft, Redmond, Washington, USA) and combined by a line with best fit.

Synthesis of PMLA/Peptide(2%)/rh(1%) Conjugate.

The reaction was conducted in the same way as PMLA/LLL/peptide(2%)/rh using PMLA/MEA(10%) conjugate (2 mg, 127.36 g/mol monomer, 0.0157 mmol) and either 1.82 mg, 3.14 μmol, or 5802.7 g/mol of AP2-PEG-maleimide, or 0.88 mg cTfRL-PEG-Mal, 2817 g/mol, or 0.88 mg M4-PEG-Mal, 2796 g/mol; 0.107 mg, 680.79 g/mol, 0.153 μmol of rhodamine-maleimide at 1% loading.

SEC-HPLC Analysis.

The analysis was performed using a Hitachi L-2130 pump with a Hitachi L-2455 detector with EZChrome Software. The SEC-HPLC column was Polysep 4000, at 1 mL/min flow rate and PBS (pH 7.4).

Rhodamine Quantification of the Final Nanoconjugates.

Prior to lyophilization, 10 μL sample of rhodamine-labeled nanoconjugate was diluted with 990 μL PBS pH 7.4. Absorbance was scanned at wavelength 570 nm (Flexstation, Molecular Devices, Sunnyvale, CA, USA). The dye concentration was calculated from A570 measurements using the molar absorbance coefficient 119,000 M⁻¹cm⁻¹. In addition, fluorescence scans (excitation 570 nm/emission 600 nm, cutoff 590 nm) confirmed the presence of rhodamine in the samples.

ζ Potential.

Synthesized conjugates were characterized with respect to their ζ potential using a Zetasizer Nano ZS90 (Malvern Instruments, Malvern, UK). Ten μL aliquots of nanoconjugate samples were diluted in 0.99 mL PBS, and the voltage applied was 150 mV. Data represent the mean of three measurements ± their standard deviation.

Dynamic Light Scattering.

The hydrodynamic diameter of synthesized conjugates, which did not contain rhodamine (these were excluded because of fluorescence effects), and PDI were obtained using Zetasizer Nano ZS90 (Malvern Instruments, Malvern, UK). Ten μL aliquots of nanoconjugate samples were diluted in 0.99 mL PBS and recorded in the volume modus. Data are given as means of three measurements.

Chemical Characterization.

Copolymers were subjected to hydrolytic cleavage in sealed ampules containing 2 M HCl for 12 h at 100 C. Malic acid in the hydrolysate was quantified by colorimetric malate dehydrogenase assay.^15,55

FTIR Measurements.

A dry sample of the materials was added to KBr powder and scanned using a Bruker Alpha instrument with a DRIFT module (Bruker, Billerica, Ma, USA). KBr alone was used for the background scan.

¹H NMR Measurements.

For each measurement, a minimum of 5 mg sample was applied. PMLA was dissolved in deuterated acetone and the other materials in deuterated DMSO. Samples were scanned 128 times at RT using a Varian “Mercury” 400 Hz instrument, and data were analyzed by MestReNova software (Mestrelab Research S.L., Santiago de Compostela, Spain).

Energy Calculations and 3D Images.

Calculations followed software from Chem3D Pro 11.0 (CambridgeSoft, Wellesley, MA, USA).

Animal Drug Injections.

Eight to nine-week-old BALB/C mice were obtained from Charles River Laboratories (Wilmington, MA, USA). Mouse maintenance and experimental procedures followed the guidelines established by the Cedars Sinai Institutional Animal Care and Use Committee (IACUC Protocol #7416). Three to four mice of each sex were used for each experiment. A total of126 mice were used to produce the data shown in this publication.

Nanoconjugates were dissolved freshly in PBS (pH 7.4) prior to each experiment and injected intravenously (i.v.) into the lateral tail vein. Mice were anesthetized with isoflurane beforehand, and their tails were briefly warmed to facilitate injections. All conjugates were administered as a single dose, at final concentrations ranging from 0.068 to 0.548 μmol of total nanoconjugate per kg body weight, or as indicated for each experiment. The drug injection volume was kept constant at 150 μL. After each injection, mice were promptly returned to their cages.

To determine drug toxicity, P/LLL/AP2/rh was injected Monday and Thursday, 2 weeks in a row. Animals were weighed after each injection and promptly returned to their cages. Mice were euthanized 2 h after the final injection, and tissue was obtained as described below.

Retroorbital Blood Collection and Tissue Collection.

Blood was drawn from the retroorbital sinus at 30 to 480 min to measure the drug concentration in the serum; relevant time points are indicated for each experiment. Blood was collected with a microhematocrit capillary tube (I.D. 1.1 mm; Chase Scientific Glass, Rockwood, TN, USA), and 150 μL blood was collected per mouse into a BD Microtainer SST, stored at room temperature for 45 min, and then centrifuged at 6000 rpm for 5 min. The serum was then transferred into fresh tubes and stored at −80 °C.

Immediately following blood collection, mice were euthanized. Euthanasia was conducted by spinal dislocation of deeply anesthetized animals. The brain, spleen, liver, heart, lungs, and kidneys were promptly removed, flash frozen, and stored at −80 °C. All tissue used for microscopic analysis was embedded in Optimal Cutting Temperature compound (OCT; Sakura, Torrance, CA, USA) and placed on dry ice for freezing.

Pharmacokinetic (PK) Measurements Using Serum.

Fluorescently labeled nanoconjugates with known concentrations (μmol/mL in PBS) were used to obtain standard fluorescence intensity calibration curves, which were then used to convert raw fluorescent data into μmol/mL serum units. To do this, 20 μL of serum containing fluorescent conjugates were placed in 96-well white opaque plates, and fluorescence intensities were measured using a fluorometer at 570/600 nm excitation/emission wavelengths with a 590 nm cutoff (Flexstation, Molecular Devices, Sunnyvale, CA, USA). Results were converted to μmol nanoconjugate/mL using the calibration curve and plotted as a function of time. The data were fitted with single expotential decay curves, and PK half-life values were calculated from the best fit in Prism (Graphpad, LaJolla, CA, USA).

Optical drug clearance measurements were performed for the brain sagittal sinus blood vessel and surrounding parenchyma for mice that were sacrificed at 30 to 480 min after drug injections (e.g., Figure 6B). Vascular fluorescence was defined as the difference between fluorescent peaks and shoulders in a linear profile that was drawn perpendicularly across the blood vessel (see Figure 6B, C). The sequential decrease in fluorescence was then converted to μmol/mL concentration via the previously determined plasma drug concentration in Figure 6A.

Biodistribution and Organ Imaging Experiment.

Extracted organs were imaged to determine the biodistribution of P/LLL/AP2/rh (Figure S6). Likewise, whole brain fluorescence imaging (Figure 3A_1–3) was performed with an IVIS Lumina XR Optical Imaging System (PerkinElmer, Richmond, CA, USA) using DsRed filters sets. Organ and brain fluorescence was simultaneously measured in all tissue samples from an experiment in order to attain consistent imaging parameters. Imaging data were analyzed with Living Image (PerkinElmer) where the average fluorescence for each organ was determined with hand drawn ROI. Fluorescence values were further analyzed in Prism.

Tissue Processing and Staining.

The cerebral vasculature was stained in every experiment to differentiate blood vessels from brain parenchyma. In experiments in Figures 2–5, 7, 8, and S8, DyLight 488 tomato-lectin (DL-1174; Vector Laboratories, Burlingame, CA) was injected as a 150 μL bolus at a 1:2 dilution in saline, 15 min prior to euthanasia. This led to widespread and optimal staining of the vasculature. Immunohistochemical staining of the vasculature was performed for tissue shown in Figures 6, S11, and S12. This was accomplished in 8—14 μm-thick cryosections that were air-dried at room temperature, fixed with 1% paraformaldehyde for 5 min, and then rinsed with PBS. The sections were then incubated in a humid chamber with blocking buffer (5% normal BSA and 0.1% Triton X-100 in PBS) for 1 h. Sections were stained with antivon Willebrand Factor (vWF, Abcam, Cambridge, UK) conjugated to AlexaFluor 488 (Thermo-fisher scientific, Canoga Park, CA, USA). After washing, the sections were mounted between coverslips in Fluoromount-G with DAPI (Invitrogen, Carlsbad, CA, USA).

Neuronal and Glial Counterstaining.

In Figures S11 and S12, we counterstained brain tissue from drug-injected mice to determine if our nanoconjugates are taken up by neurons or glia. Cryosections (8–14 μm thickness) were air-dried for 10 min, fixed with 2% paraformaldehyde for 5 min, and then rinsed three times with PBS. The sections were then incubated in a humid chamber with blocking buffer containing 5% normal BSA, 0.25% Triton X-100, 2% DMSO, and 1% normal goat serum (all from Sigma) in PBS (PBS-T) for 1–2 h at room temperature. The antibodies anti-Neun (Abcam, Cambridge, MA, USA; AB104225) and anti-GFAP (Neuromics, Edina, MN, USA; CH22102) were then diluted 1:500 in PBS-T, and tissue sections incubated simultaneously with antibody solutions overnight at 4 °C in a humid chamber. Tissue sections were then washed five times with PBS-T, incubated in appropriate secondary antibody solutions, and diluted 1:250 in PBS-T, for 2–4 h. After five washes in PBS at room temperature, sections were mounted between coverslips in Fluoromount-G with DAPI.

Image Acquisition and Optical Analysis.

Imaging was performed with a Lecia DM 6000B epifluorescence microscope (Leica Microsystems, Wetzlar, Germany). Rhodamine-labeled nanoconjugates were visualized with a 534—558 nm excitation and 560–640 nm emission filter set, viewed with a 20X Leica HC Plan Apo 0.70 NA. and a 40X Leica HCX Plan Apo 0.85 N.A. lens, and recorded with a Leica DFC 360 FX camera. The camera was controlled with Leica LAS X software, and images were acquired with 4.5 s + 2.0 gain exposures with the 20X lens and 3.5 s + 2.0 gain exposures with a 40X lens. These parameters were held constant to enable image-to-image comparisons across specimens. Other fluorophores (DAPI, tomato-lectin, secondary antibodies) were viewed using complementary filter sets, and imaging parameters were also held consistent across experimental trials.

Analysis of Optical Data.

Optical imaging data analysis was performed with ImageJ FIJI.⁵⁶ To determine if our nanoconjugates entered the brain parenchyma, we performed an image intensity analysis in regions that did not contain vasculature (see yellow boxes in Figure 3B₁). In this analysis, 20 X 10 μm² sized ROI (ROI) were randomly overlaid on images showing the vasculature, explicitly avoiding blood vessels. Intensity measurements and positions were then obtained for each ROI after it was separately overlaid on the image showing nanoconjugate fluorescence. Fluorescence measurements were thus based on the anatomy of the cerebral vasculature rather than nanoconjugate labeling and were therefore unbiased with regards to nanoconjugate load. Overall levels of nanoconjugate labeling are shown as means and standard errors for 20 measurements from four separate images of the hippocampal CA1–3 layers, layers II/ III of the somatosensory or visual cortices, and the superior and inferior midbrain colliculi (4 mice for each conjugate and brain region). To determine how nanoconjugate-associated fluorescence relates to the anatomy of the vasculature, we manually measured the distance of each analyzed ROI from the nearest blood vessel wall using the line tool in FIJI. Intensity values were then plotted against the location of the blood vessel wall and summarized in scatterplots (Figure 3C).

All nanoconjugate fluorescence measurements in brain parenchyma (e.g., Figures 3D_1–3) are presented as relative fluorescence intensities. Relative fluorescence measurements were obtained by subtracting from each nanoconjugate fluorescence measurement the average intensity of autofluorescence, obtained via imaging a corresponding brain region in mice injected with PBS (total 6 mice and 28 images for each brain region). Relative fluorescence measurements thus represent total fluorescence minus autofluorescence. Data plots and statistical analysis were conducted in Prism. Unless indicated otherwise, fluorescence measurements were compared via a one-way ANOVA combined with pairwise posthoc comparisons of individual data points; exact parameters and tests are indicated for each result.

Supplementary Material

ACS Nano

NIHMS1054603-supplement-ACS_Nano.pdf^{(1.2MB, pdf)}

ACKNOWLEDGMENTS

This work was supported by the following grants: R01 CA188743 (JYL), R01 CA206220 (JYL), R01 CA230858-01, R01 CA209921 (EH), and Health Effects of Air Pollution Foundation Agreement No. BTAP011 and BTAP013 (KLB).

Footnotes

ASSOCIATED CONTENT

Supporting Information

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.8b06437.

Additional figures and data pertaining to nanoconjugate characterization and biodistribution. Data include SEC-HPLC, FTIR (Figures S1–S2), and NMR (Figures S3–S5) nanoconjugate characterizations, whole-animal biodistribution (Figure S6) and drug toxicity data (Figure S7), a morphological analysis of vascular content in different brain regions (Figure S8), Chem3D energy analyses of nanoconjugate analogues (Figures S9–S10), and morphological data that demonstrates a lack of nonconjugate internalization by neurons and glia (Figures S11–S12) (PDF)

The authors declare no competing financial interest.

REFERENCES

(1).Dréan A; Goldwirt L; Verreault M; Canney M; Schmitt C; Guehennec J; Delattre J-Y; Carpentier A; Idbaih A Blood-Brain Barrier, Cytotoxic Chemotherapies and Glioblastoma. Expert Rev. Neurother. 2016, 16, 1285–1300. [DOI] [PubMed] [Google Scholar]
(2).Abbott NJ Blood–Brain Barrier Structure and Function and the Challenges for CNS Drug Delivery. J. Inherited Metab. Dis. 2013, 36, 437–449. [DOI] [PubMed] [Google Scholar]
(3).Alyautdin R; Khalin I; Nafeeza MI; Haron MH; Kuznetsov D Nanoscale Drug Delivery Systems and the Blood— Brain Barrier. Int. J. Nanomed. 2014, 9, 795–811. [DOI] [PMC free article] [PubMed] [Google Scholar]
(4).Lopes S; Ribeiro F; Wojnarowicz J; Ƚojkowski W; Jurkschat K; Crossley A; Soares AM; Loureiro S Zinc Oxide Nanoparticles Toxicity to Daphnia Magna: Size-Dependent Effects and Dissolution. Environ. Toxicol. Chem. 2014, 33, 190–198. [DOI] [PubMed] [Google Scholar]
(5).Shang L; Nienhaus K; Nienhaus GU Engineered Nanoparticles Interacting with Cells: Size Matters. J. Nanobiotechnol. 2014, 12, 5. [DOI] [PMC free article] [PubMed] [Google Scholar]
(6).Elnegaard MP; List M; Christiansen H; Schmidt S; Mollenhauer J; Block I Protein-Based Nanotoxicology Assessment Strategy. Nanomedicine 2017, 13, 1229–1233. [DOI] [PubMed] [Google Scholar]
(7).Loureiro JA; Gomes B; Coelho MA; Carmo Pereira M. d.; Rocha S Targeting Nanoparticles Across the Blood–Brain Barrier with Monoclonal Antibodies. Nanomedicine 2014, 9, 709–722. [DOI] [PubMed] [Google Scholar]
(8).Baruch K; Deczkowska A; Rosenzweig N; Tsitsou-Kampeli A; Sharif AM; Matcovitch-Natan O; Kertser A; David E; Amit I; Schwartz M PD-1 Immune Checkpoint Blockade Reduces Pathology and Improves Memory in Mouse Models of Alzheimer’s Disease. Nat. Med. 2016, 22, 135–137. [DOI] [PubMed] [Google Scholar]
(9).Sevigny J; Chiao P; Williams L; Chen T; Ling Y; O’Gorman J; Hock C; Nitsch RM; Sandrock A Aducanumab (BIIB037), an Anti-Amyloid ß Monoclonal Antibody, in Patients with Prodromal or Mild Alzheimer’s Disease: Interim Results of a Randomized, Double-Blind, Placebo-Controlled, Phase 1b Study. Alzheimer’s Dementia 2015, 11, P277. [Google Scholar]
(10).Sevigny J; Chiao P; Bussiere T; Weinreb PH; Williams L; Maier M; Dunstan R; Salloway S; Chen T; Ling Y; et al. The Antibody Aducanumab Reduces Aß Plaques in Alzheimer’s Disease. Nature 2016, 537, 50–56. [DOI] [PubMed] [Google Scholar]
(11).Laubli H; Balmelli C; Bossard M; Pfister O; Glatz K; Zippelius A Acute Heart Failure due to Autoimmune Myocarditis under Pembrolizumab Treatment for Metastatic Melanoma. J. Immunother. Cancer. 2015, 3, 11. [DOI] [PMC free article] [PubMed] [Google Scholar]
(12).Borlak J; Langer F; Spanel R; Schondorfer G; Dittrich C Immune-Mediated Liver Injury of the Cancer Therapeutic Antibody Catumaxomab Targeting EpCAM, CD3 and Fcγ Receptors. Oncotarget 2016, 7, 28059–28074. [DOI] [PMC free article] [PubMed] [Google Scholar]
(13).Lee B-S; Fujita M; Khazenzon NM; Wawrowsky KA; Wachsmann-Hogiu S; Farkas DL; Black KL; Ljubimova JY; Holler E Polycefin, a New Prototype of a Multifunctional Nanoconjugate Based on Poly(ß-l-malic acid) for Drug Delivery. Bioconjugate Chem. 2006, 17, 317–326. [DOI] [PMC free article] [PubMed] [Google Scholar]
(14).Ljubimova J; Fujita M; Lee B; Khazenzon N; Wachsmann-Hogiu S; Farkas D; Black K; Holler E In Nanoconjugates of Poly (Malic Acid) with Functional Modules for Drug Delivery; NSTI-Nanotech: Austin, TX, 2006; pp 354–357. [Google Scholar]
(15).Ljubimova JY; Ding H; Portilla-Arias J; Patil R; Gangalum PR; Chesnokova A; Inoue S; Rekechenetskiy A; Nassoura T; Black KL Polymalic Acid-Based Nano Biopolymers for Targeting of Multiple Tumor Markers: an Opportunity for Personalized Medicine? J. Visualized Exp. 2014, 88, e50668. [DOI] [PMC free article] [PubMed] [Google Scholar]
(16).Ljubimova JY; Portilla-Arias J; Patil R; Ding H; Inoue S; Markman JL; Rekechenetskiy A; Konda B; Gangalum PR; Chesnokova A; Ljubimov AV; Black KL; Holler E Toxicity and Efficacy Evaluation of Multiple Targeted Polymalic Acid Conjugates for Triple-Negative Breast Cancer Treatment. J. Drug Targeting 2013, 21, 956–967. [DOI] [PMC free article] [PubMed] [Google Scholar]
(17).Ljubimova JY; Fujita M; Ljubimov AV; Torchilin VP; Black KL; Holler E Poly(Malic Acid) Nanoconjugates Containing Various Antibodies and Oligonucleotides for Multitargeting Drug Delivery. Nanomedicine 2008, 3, 247–265. [DOI] [PMC free article] [PubMed] [Google Scholar]
(18).Ljubimova JY; Fujita M; Khazenzon NM; Lee B-S; Wachsmann-Hogiu S; Farkas DL; Black KL; Holler E Nanoconjugate Based on Polymalic Acid for Tumor Targeting. Chem.-Biol. Interact. 2008, 171, 195–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
(19).Georgieva JV; Hoekstra D; Zuhorn IS Smuggling Drugs into the Brain: an Overview of Ligands Targeting Transcytosis for Drug Delivery Across the Blood–Brain Barrier. Pharmaceutics 2014, 6, 557–583. [DOI] [PMC free article] [PubMed] [Google Scholar]
(20).Dehouck B; Fenart L; Dehouck M-P; Pierce A; Torpier G; Cecchelli R A New Function for the LDL Receptor: Transcytosis of LDL Across the Blood–Brain Barrier. J. Cell Biol. 1997, 138, 877–889. [DOI] [PMC free article] [PubMed] [Google Scholar]
(21).Demeule M; Regina A; Che C; Poirier J; Nguyen T; Gabathuler R; Castaigne J-P; Beliveau R Identification and Design of Peptides as a New Drug Delivery System for the Brain. J. Pharmacol. Exp. Ther. 2007, 324, 1064–1072. [DOI] [PubMed] [Google Scholar]
(22).Demeule M; Currie JC; Bertrand Y; Che C; Nguyen T; Regina A; Gabathuler R; Castaigne JP; Beliveau R Involvement of the Low-Density Lipoprotein Receptor-Related Protein in the Transcytosis of the Brain Delivery Vector Angiopep-2. J. Neurochem. 2008, 106, 1534–1544. [DOI] [PubMed] [Google Scholar]
(23).Tian X; Nyberg S; Sharp PS; Madsen J; Daneshpour N; Armes SP; Berwick J; Azzouz M; Shaw P; Abbott NJ; Battaglia G LRP-1-Mediated Intracellular Antibody Delivery to the Central Nervous System. Sci. Rep. 2015, 5, 11990. [DOI] [PMC free article] [PubMed] [Google Scholar]
(24).Johnsen KB; Burkhart A; Melander F; Kempen PJ; Vejlebo JB; Siupka P; Nielsen MS; Andresen TL; Moos T Targeting Transferrin Receptors at the Blood-Brain Barrier Improves the Uptake of Immunoliposomes and Subsequent Cargo Transport into the Brain Parenchyma. Sci. Rep. 2017, 7, 10396. [DOI] [PMC free article] [PubMed] [Google Scholar]
(25).Staquicini FI; Ozawa MG; Moya CA; Driessen WH; Barbu EM; Nishimori H; Soghomonyan S; Flores LG 2nd; Liang X; Paolillo V; et al. Systemic Combinatorial Peptide Selection Yields a Non-Canonical Iron-Mimicry Mechanism for Targeting Tumors in a Mouse Model of Human Glioblastoma. J. Clin. Invest. 2011, 121, 161–173. [DOI] [PMC free article] [PubMed] [Google Scholar]
(26).Xia H; Anderson B; Mao Q; Davidson BL Recombinant Human Adenovirus: Targeting to the Human Transferrin Receptor Improves Gene Transfer to Brain Microcapillary Endothelium. J. Virol. 2000, 74, 11359–11366. [DOI] [PMC free article] [PubMed] [Google Scholar]
(27).Yin T; Yang L; Liu Y; Zhou X; Sun J; Liu J Sialic Acid (SA)-Modified Selenium Nanoparticles Coated with a High Blood— Brain Barrier Permeability Peptide-B6 Peptide for Potential use in Alzheimer’s Disease. Acta Biomater. 2015, 25, 172–183. [DOI] [PubMed] [Google Scholar]
(28).Liu Z; Gao X; Kang T; Jiang M; Miao D; Gu G; Hu Q; Song Q; Yao L; Tu Y; et al. B6 Peptide-Modified PEG-PLA Nanoparticles for Enhanced Brain Delivery of Neuroprotective Peptide. Bioconjugate Chem. 2013, 24, 997–1007. [DOI] [PubMed] [Google Scholar]
(29).Oller-Salvia B; Sánchez-Navarro M; Ciudad S; Guiu M; Arranz-Gibert P; Garcia C; Gomis RR; Cecchelli R; García J; Giralt E; et al. MiniAp-4: a Venom-Inspired Peptidomimetic for Brain Delivery. Angew. Chem., Int. Ed. 2016, 55, 572–575. [DOI] [PMC free article] [PubMed] [Google Scholar]
(30).Ding H; Portilla-Arias J; Patil R; Black KL; Ljubimova JY; Holler E The Optimization of Polymalic Acid Peptide Copolymers for Endosomolytic Drug Delivery. Biomaterials 2011, 32, 5269–5278. [DOI] [PMC free article] [PubMed] [Google Scholar]
(31).Ding H; Inoue S; Ljubimov AV; Patil R; Portilla-Arias J; Hu J; Konda B; Wawrowsky KA; Fujita M; Karabalin N; et al. Inhibition of Brain Tumor Growth by Intravenous Poly (ß-L-malic acid) Nanobioconjugate with pH-dependent Drug Release. Proc. Natl. Acad. Sci. U. S.A. 2010, 107, 18143–18148. [DOI] [PMC free article] [PubMed] [Google Scholar]
(32).Patil R; Ljubimov AV; Gangalum PR; Ding H; Portilla-Arias J; Wagner S; Inoue S; Konda B; Rekechenetskiy A; Chesnokova A; Markman JL; Ljubimov VA; Li D; Prasad RS; Black KL; Holler E; Ljubimova JY MRI Virtual Biopsy and Treatment of Brain Metastatic Tumors with Targeted Nanobioconjugates: Nanoclinic in the Brain. ACS Nano 2015, 9, 5594–5608. [DOI] [PMC free article] [PubMed] [Google Scholar]
(33).Manitchotpisit P; Skory CD; Peterson SW; Price NPJ; Vermillion KE; Leathers TD Poly(β-L-malic acid) Production by Diverse Phylogenetic Clades of Aureobasidium Pullulans. J. Ind. Microbiol. Biotechnol. 2012, 39, 125–132. [DOI] [PubMed] [Google Scholar]
(34).Inoue S; Patil R; Portilla-Arias J; Ding H; Konda B; Espinoza A; Mongayt D; Markman JL; Elramsisy A; Phillips HW; Black KL; Holler E; Ljubimova JY Nanobiopolymer for Direct Targeting and Inhibition of EGFR Expression in Triple Negative Breast Cancer. PLoS One 2012, 7, e31070. [DOI] [PMC free article] [PubMed] [Google Scholar]
(35).Di Guardo G Lipofuscin, Lipofuscin-Like Pigments and Autofluorescence. Eur. J. Histochem. 2015, 59 (2485), 72–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
(36).Feldman TB; Yakovleva MA; Dontsov AE; Ostrovsky MA Fluorescence Emission and Excitation Spectra of Fluorophores of Lipofuscin Granules Isolated from Retinal Pigment Epithelium of Human Cadaver Eyes. Russ. Chem. Bull. 2010, 59, 276–283. [Google Scholar]
(37).Marmorstein AD; Marmorstein LY; Sakaguchi H; Hollyfield JG Spectral Profiling of Autofluorescence Associated with Lipofuscin, Bruch’s Membrane, and Sub-RPE Deposits in Normal and AMD Eyes. Invest. Ophthalmol. Visual Sci. 2002, 43, 2435–2441. [PubMed] [Google Scholar]
(38).Zheng X; Pang X; Yang P; Wan X; Wei Y; Guo Q; Zhang Q; Jiang X A Hybrid siRNA Delivery Complex for Enhanced Brain Penetration and Precise Amyloid Plaque Targeting in Alzheimer’s Disease Mice. Acta Biomater. 2017, 49, 388–401. [DOI] [PubMed] [Google Scholar]
(39).Robertson RT; Levine ST; Haynes SM; Gutierrez P; Baratta JL; Tan Z; Longmuir KJ Use of Labeled Tomato Lectin for Imaging Vasculature Structures. Histochem. Cell Biol. 2015, 143, 225–234. [DOI] [PubMed] [Google Scholar]
(40).Jahrling N; Becker K; Dodt HU 3D-reconstruction of Blood Vessels by Ultramicroscopy. Organogenesis 2009, 5, 227–230. [DOI] [PMC free article] [PubMed] [Google Scholar]
(41).Tsai PS; Kaufhold JP; Blinder P; Friedman B; Drew PJ; Karten HJ; Lyden PD; Kleinfeld D Correlations of Neuronal and Microvascular Densities in Murine Cortex Revealed by Direct Counting and Colocalization of Nuclei and Vessels. J. Neurosci. 2009, 29, 14553–14570. [DOI] [PMC free article] [PubMed] [Google Scholar]
(42).Zeidman P; Maguire EA Anterior Hippocampus: the Anatomy of Perception, Imagination and Episodic Memory. Nat. Rev. Neurosci. 2016, 17, 173–182. [DOI] [PMC free article] [PubMed] [Google Scholar]
(43).La Joie R; Landeau B; Perrotin A; Bejanin A; Egret S; Pelerin A; Mezenge F; Belliard S; De La Sayette V; Eustache F; et al. Intrinsic Connectivity Identifies the Hippocampus as a Main Crossroad Between Alzheimer’s and Semantic Dementia-Targeted Networks. Neuron 2014, 81, 1417–1428. [DOI] [PubMed] [Google Scholar]
(44).Xia M; Cheng X; Yi R; Gao D; Xiong J The Binding Receptors of Aß: an Alternative Therapeutic Target for Alzheimer’s Disease. Mol. Neurobiol 2016, 53, 455–471. [DOI] [PubMed] [Google Scholar]
(45).Ding H; Portilla-Arias J; Patil R; Black KL; Ljubimova JY; Holler E Distinct Mechanisms of Membrane Permeation Induced by Two Polymalic Acid Copolymers. Biomaterials 2013, 34, 217–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
(46).Bockenhoff A; Cramer S; Wolte P; Knieling S; Wohlenberg C; Gieselmann V; Galla HJ; Matzner U Comparison of Five Peptide Vectors for Improved Brain Delivery of the Lysosomal Enzyme Arylsulfatase A. J. Neurosci. 2014, 34, 3122–3129. [DOI] [PMC free article] [PubMed] [Google Scholar]
(47).Meng J; Agrahari V; Youm I Advances in Targeted Drug Delivery Approaches for the Central Nervous System Tumors: the Inspiration of Nanobiotechnology. J. Neuroimmune Pharmacol 2017, 12, 84–98. [DOI] [PubMed] [Google Scholar]
(48).Grimmer T; Goldhardt O; Guo L-H; Yousefi BH; Forster S; Drzezga A; Sorg C; Alexopoulos P; Forstl H; Kurz A; et al. LRP-1 Polymorphism is Associated with Global and Regional Amyloid Load in Alzheimer’s Disease in Humans in-vivo. Neuroimage Clin 2014, 4, 411–416. [DOI] [PMC free article] [PubMed] [Google Scholar]
(49).Patel P; Shah J Role of Vitamin D in Amyloid Clearance via LRP-1 Pregulation in Alzheimer’s Disease: A Potential Therapeutic Target? J. Chem. Neuroanat. 2017, 85, 36–42. [DOI] [PubMed] [Google Scholar]
(50).Sweeney MD; Sagare AP; Zlokovic BV Blood-Brain Barrier Breakdown in Alzheimer Disease and other Neurodegenerative Disorders. Nat. Rev. Neurol. 2018, 14, 133–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
(51).Zlokovic BV Neurovascular Pathways to Neurodegeneration in Alzheimer’s Disease and other Disorders. Nat. Rev. Neurosci. 2011, 12, 723–738. [DOI] [PMC free article] [PubMed] [Google Scholar]
(52).Wiesehan K; Willbold D Mirror-image Phage Display: Aiming at the Mirror. ChemBioChem 2003, 4, 811–815. [DOI] [PubMed] [Google Scholar]
(53).van Groen T; Kadish I; Wiesehan K; Funke SA; Willbold D In vitro and in vivo Staining Characteristics of Small, Fluorescent, Aß42-binding D-enantiomeric Peptides in Transgenic AD mouse models. ChemMedChem 2009, 4, 276–282. [DOI] [PubMed] [Google Scholar]
(54).Wiesehan K; Buder K; Linke RP; Patt S; Stoldt M; Unger E; Schmitt B; Bucci E; Willbold D Selection of D-amino-acid Peptides that Bind to Alzheimer’s Disease Amyloid Peptide Aß1–42 by Mirror Image Phage Display. ChemBioChem 2003, 4, 748–753. [DOI] [PubMed] [Google Scholar]
(55).Rozema DB; Ekena K; Lewis DL; Loomis AG; Wolff JA Endosomolysis by Masking of a Membrane-Active Agent (EMMA) for Cytoplasmic Release of Macromolecules. Bioconjugate Chem. 2003, 14, 51–57. [DOI] [PubMed] [Google Scholar]
(56).Schindelin J; Arganda-Carreras I; Frise E; Kaynig V; Longair M; Pietzsch T; Preibisch S; Rueden C; Saalfeld S; Schmid B; Schmid B; Tinevez JY; White DJ; Hartenstein V; Eliceiri K; Tomancak P; Cardona A Fiji: an Open-Source Platform for Biological-Image Analysis. Nat. Methods 2012, 9, 676–682. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ACS Nano

NIHMS1054603-supplement-ACS_Nano.pdf^{(1.2MB, pdf)}

[R1] (1).Dréan A; Goldwirt L; Verreault M; Canney M; Schmitt C; Guehennec J; Delattre J-Y; Carpentier A; Idbaih A Blood-Brain Barrier, Cytotoxic Chemotherapies and Glioblastoma. Expert Rev. Neurother. 2016, 16, 1285–1300. [DOI] [PubMed] [Google Scholar]

[R2] (2).Abbott NJ Blood–Brain Barrier Structure and Function and the Challenges for CNS Drug Delivery. J. Inherited Metab. Dis. 2013, 36, 437–449. [DOI] [PubMed] [Google Scholar]

[R3] (3).Alyautdin R; Khalin I; Nafeeza MI; Haron MH; Kuznetsov D Nanoscale Drug Delivery Systems and the Blood— Brain Barrier. Int. J. Nanomed. 2014, 9, 795–811. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] (4).Lopes S; Ribeiro F; Wojnarowicz J; Ƚojkowski W; Jurkschat K; Crossley A; Soares AM; Loureiro S Zinc Oxide Nanoparticles Toxicity to Daphnia Magna: Size-Dependent Effects and Dissolution. Environ. Toxicol. Chem. 2014, 33, 190–198. [DOI] [PubMed] [Google Scholar]

[R5] (5).Shang L; Nienhaus K; Nienhaus GU Engineered Nanoparticles Interacting with Cells: Size Matters. J. Nanobiotechnol. 2014, 12, 5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] (6).Elnegaard MP; List M; Christiansen H; Schmidt S; Mollenhauer J; Block I Protein-Based Nanotoxicology Assessment Strategy. Nanomedicine 2017, 13, 1229–1233. [DOI] [PubMed] [Google Scholar]

[R7] (7).Loureiro JA; Gomes B; Coelho MA; Carmo Pereira M. d.; Rocha S Targeting Nanoparticles Across the Blood–Brain Barrier with Monoclonal Antibodies. Nanomedicine 2014, 9, 709–722. [DOI] [PubMed] [Google Scholar]

[R8] (8).Baruch K; Deczkowska A; Rosenzweig N; Tsitsou-Kampeli A; Sharif AM; Matcovitch-Natan O; Kertser A; David E; Amit I; Schwartz M PD-1 Immune Checkpoint Blockade Reduces Pathology and Improves Memory in Mouse Models of Alzheimer’s Disease. Nat. Med. 2016, 22, 135–137. [DOI] [PubMed] [Google Scholar]

[R9] (9).Sevigny J; Chiao P; Williams L; Chen T; Ling Y; O’Gorman J; Hock C; Nitsch RM; Sandrock A Aducanumab (BIIB037), an Anti-Amyloid ß Monoclonal Antibody, in Patients with Prodromal or Mild Alzheimer’s Disease: Interim Results of a Randomized, Double-Blind, Placebo-Controlled, Phase 1b Study. Alzheimer’s Dementia 2015, 11, P277. [Google Scholar]

[R10] (10).Sevigny J; Chiao P; Bussiere T; Weinreb PH; Williams L; Maier M; Dunstan R; Salloway S; Chen T; Ling Y; et al. The Antibody Aducanumab Reduces Aß Plaques in Alzheimer’s Disease. Nature 2016, 537, 50–56. [DOI] [PubMed] [Google Scholar]

[R11] (11).Laubli H; Balmelli C; Bossard M; Pfister O; Glatz K; Zippelius A Acute Heart Failure due to Autoimmune Myocarditis under Pembrolizumab Treatment for Metastatic Melanoma. J. Immunother. Cancer. 2015, 3, 11. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R12] (12).Borlak J; Langer F; Spanel R; Schondorfer G; Dittrich C Immune-Mediated Liver Injury of the Cancer Therapeutic Antibody Catumaxomab Targeting EpCAM, CD3 and Fcγ Receptors. Oncotarget 2016, 7, 28059–28074. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] (13).Lee B-S; Fujita M; Khazenzon NM; Wawrowsky KA; Wachsmann-Hogiu S; Farkas DL; Black KL; Ljubimova JY; Holler E Polycefin, a New Prototype of a Multifunctional Nanoconjugate Based on Poly(ß-l-malic acid) for Drug Delivery. Bioconjugate Chem. 2006, 17, 317–326. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] (14).Ljubimova J; Fujita M; Lee B; Khazenzon N; Wachsmann-Hogiu S; Farkas D; Black K; Holler E In Nanoconjugates of Poly (Malic Acid) with Functional Modules for Drug Delivery; NSTI-Nanotech: Austin, TX, 2006; pp 354–357. [Google Scholar]

[R15] (15).Ljubimova JY; Ding H; Portilla-Arias J; Patil R; Gangalum PR; Chesnokova A; Inoue S; Rekechenetskiy A; Nassoura T; Black KL Polymalic Acid-Based Nano Biopolymers for Targeting of Multiple Tumor Markers: an Opportunity for Personalized Medicine? J. Visualized Exp. 2014, 88, e50668. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] (16).Ljubimova JY; Portilla-Arias J; Patil R; Ding H; Inoue S; Markman JL; Rekechenetskiy A; Konda B; Gangalum PR; Chesnokova A; Ljubimov AV; Black KL; Holler E Toxicity and Efficacy Evaluation of Multiple Targeted Polymalic Acid Conjugates for Triple-Negative Breast Cancer Treatment. J. Drug Targeting 2013, 21, 956–967. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] (17).Ljubimova JY; Fujita M; Ljubimov AV; Torchilin VP; Black KL; Holler E Poly(Malic Acid) Nanoconjugates Containing Various Antibodies and Oligonucleotides for Multitargeting Drug Delivery. Nanomedicine 2008, 3, 247–265. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] (18).Ljubimova JY; Fujita M; Khazenzon NM; Lee B-S; Wachsmann-Hogiu S; Farkas DL; Black KL; Holler E Nanoconjugate Based on Polymalic Acid for Tumor Targeting. Chem.-Biol. Interact. 2008, 171, 195–203. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] (19).Georgieva JV; Hoekstra D; Zuhorn IS Smuggling Drugs into the Brain: an Overview of Ligands Targeting Transcytosis for Drug Delivery Across the Blood–Brain Barrier. Pharmaceutics 2014, 6, 557–583. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] (20).Dehouck B; Fenart L; Dehouck M-P; Pierce A; Torpier G; Cecchelli R A New Function for the LDL Receptor: Transcytosis of LDL Across the Blood–Brain Barrier. J. Cell Biol. 1997, 138, 877–889. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] (21).Demeule M; Regina A; Che C; Poirier J; Nguyen T; Gabathuler R; Castaigne J-P; Beliveau R Identification and Design of Peptides as a New Drug Delivery System for the Brain. J. Pharmacol. Exp. Ther. 2007, 324, 1064–1072. [DOI] [PubMed] [Google Scholar]

[R22] (22).Demeule M; Currie JC; Bertrand Y; Che C; Nguyen T; Regina A; Gabathuler R; Castaigne JP; Beliveau R Involvement of the Low-Density Lipoprotein Receptor-Related Protein in the Transcytosis of the Brain Delivery Vector Angiopep-2. J. Neurochem. 2008, 106, 1534–1544. [DOI] [PubMed] [Google Scholar]

[R23] (23).Tian X; Nyberg S; Sharp PS; Madsen J; Daneshpour N; Armes SP; Berwick J; Azzouz M; Shaw P; Abbott NJ; Battaglia G LRP-1-Mediated Intracellular Antibody Delivery to the Central Nervous System. Sci. Rep. 2015, 5, 11990. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] (24).Johnsen KB; Burkhart A; Melander F; Kempen PJ; Vejlebo JB; Siupka P; Nielsen MS; Andresen TL; Moos T Targeting Transferrin Receptors at the Blood-Brain Barrier Improves the Uptake of Immunoliposomes and Subsequent Cargo Transport into the Brain Parenchyma. Sci. Rep. 2017, 7, 10396. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] (25).Staquicini FI; Ozawa MG; Moya CA; Driessen WH; Barbu EM; Nishimori H; Soghomonyan S; Flores LG 2nd; Liang X; Paolillo V; et al. Systemic Combinatorial Peptide Selection Yields a Non-Canonical Iron-Mimicry Mechanism for Targeting Tumors in a Mouse Model of Human Glioblastoma. J. Clin. Invest. 2011, 121, 161–173. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R26] (26).Xia H; Anderson B; Mao Q; Davidson BL Recombinant Human Adenovirus: Targeting to the Human Transferrin Receptor Improves Gene Transfer to Brain Microcapillary Endothelium. J. Virol. 2000, 74, 11359–11366. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R27] (27).Yin T; Yang L; Liu Y; Zhou X; Sun J; Liu J Sialic Acid (SA)-Modified Selenium Nanoparticles Coated with a High Blood— Brain Barrier Permeability Peptide-B6 Peptide for Potential use in Alzheimer’s Disease. Acta Biomater. 2015, 25, 172–183. [DOI] [PubMed] [Google Scholar]

[R28] (28).Liu Z; Gao X; Kang T; Jiang M; Miao D; Gu G; Hu Q; Song Q; Yao L; Tu Y; et al. B6 Peptide-Modified PEG-PLA Nanoparticles for Enhanced Brain Delivery of Neuroprotective Peptide. Bioconjugate Chem. 2013, 24, 997–1007. [DOI] [PubMed] [Google Scholar]

[R29] (29).Oller-Salvia B; Sánchez-Navarro M; Ciudad S; Guiu M; Arranz-Gibert P; Garcia C; Gomis RR; Cecchelli R; García J; Giralt E; et al. MiniAp-4: a Venom-Inspired Peptidomimetic for Brain Delivery. Angew. Chem., Int. Ed. 2016, 55, 572–575. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] (30).Ding H; Portilla-Arias J; Patil R; Black KL; Ljubimova JY; Holler E The Optimization of Polymalic Acid Peptide Copolymers for Endosomolytic Drug Delivery. Biomaterials 2011, 32, 5269–5278. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] (31).Ding H; Inoue S; Ljubimov AV; Patil R; Portilla-Arias J; Hu J; Konda B; Wawrowsky KA; Fujita M; Karabalin N; et al. Inhibition of Brain Tumor Growth by Intravenous Poly (ß-L-malic acid) Nanobioconjugate with pH-dependent Drug Release. Proc. Natl. Acad. Sci. U. S.A. 2010, 107, 18143–18148. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] (32).Patil R; Ljubimov AV; Gangalum PR; Ding H; Portilla-Arias J; Wagner S; Inoue S; Konda B; Rekechenetskiy A; Chesnokova A; Markman JL; Ljubimov VA; Li D; Prasad RS; Black KL; Holler E; Ljubimova JY MRI Virtual Biopsy and Treatment of Brain Metastatic Tumors with Targeted Nanobioconjugates: Nanoclinic in the Brain. ACS Nano 2015, 9, 5594–5608. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] (33).Manitchotpisit P; Skory CD; Peterson SW; Price NPJ; Vermillion KE; Leathers TD Poly(β-L-malic acid) Production by Diverse Phylogenetic Clades of Aureobasidium Pullulans. J. Ind. Microbiol. Biotechnol. 2012, 39, 125–132. [DOI] [PubMed] [Google Scholar]

[R34] (34).Inoue S; Patil R; Portilla-Arias J; Ding H; Konda B; Espinoza A; Mongayt D; Markman JL; Elramsisy A; Phillips HW; Black KL; Holler E; Ljubimova JY Nanobiopolymer for Direct Targeting and Inhibition of EGFR Expression in Triple Negative Breast Cancer. PLoS One 2012, 7, e31070. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R35] (35).Di Guardo G Lipofuscin, Lipofuscin-Like Pigments and Autofluorescence. Eur. J. Histochem. 2015, 59 (2485), 72–73. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R36] (36).Feldman TB; Yakovleva MA; Dontsov AE; Ostrovsky MA Fluorescence Emission and Excitation Spectra of Fluorophores of Lipofuscin Granules Isolated from Retinal Pigment Epithelium of Human Cadaver Eyes. Russ. Chem. Bull. 2010, 59, 276–283. [Google Scholar]

[R37] (37).Marmorstein AD; Marmorstein LY; Sakaguchi H; Hollyfield JG Spectral Profiling of Autofluorescence Associated with Lipofuscin, Bruch’s Membrane, and Sub-RPE Deposits in Normal and AMD Eyes. Invest. Ophthalmol. Visual Sci. 2002, 43, 2435–2441. [PubMed] [Google Scholar]

[R38] (38).Zheng X; Pang X; Yang P; Wan X; Wei Y; Guo Q; Zhang Q; Jiang X A Hybrid siRNA Delivery Complex for Enhanced Brain Penetration and Precise Amyloid Plaque Targeting in Alzheimer’s Disease Mice. Acta Biomater. 2017, 49, 388–401. [DOI] [PubMed] [Google Scholar]

[R39] (39).Robertson RT; Levine ST; Haynes SM; Gutierrez P; Baratta JL; Tan Z; Longmuir KJ Use of Labeled Tomato Lectin for Imaging Vasculature Structures. Histochem. Cell Biol. 2015, 143, 225–234. [DOI] [PubMed] [Google Scholar]

[R40] (40).Jahrling N; Becker K; Dodt HU 3D-reconstruction of Blood Vessels by Ultramicroscopy. Organogenesis 2009, 5, 227–230. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R41] (41).Tsai PS; Kaufhold JP; Blinder P; Friedman B; Drew PJ; Karten HJ; Lyden PD; Kleinfeld D Correlations of Neuronal and Microvascular Densities in Murine Cortex Revealed by Direct Counting and Colocalization of Nuclei and Vessels. J. Neurosci. 2009, 29, 14553–14570. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R42] (42).Zeidman P; Maguire EA Anterior Hippocampus: the Anatomy of Perception, Imagination and Episodic Memory. Nat. Rev. Neurosci. 2016, 17, 173–182. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R43] (43).La Joie R; Landeau B; Perrotin A; Bejanin A; Egret S; Pelerin A; Mezenge F; Belliard S; De La Sayette V; Eustache F; et al. Intrinsic Connectivity Identifies the Hippocampus as a Main Crossroad Between Alzheimer’s and Semantic Dementia-Targeted Networks. Neuron 2014, 81, 1417–1428. [DOI] [PubMed] [Google Scholar]

[R44] (44).Xia M; Cheng X; Yi R; Gao D; Xiong J The Binding Receptors of Aß: an Alternative Therapeutic Target for Alzheimer’s Disease. Mol. Neurobiol 2016, 53, 455–471. [DOI] [PubMed] [Google Scholar]

[R45] (45).Ding H; Portilla-Arias J; Patil R; Black KL; Ljubimova JY; Holler E Distinct Mechanisms of Membrane Permeation Induced by Two Polymalic Acid Copolymers. Biomaterials 2013, 34, 217–225. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R46] (46).Bockenhoff A; Cramer S; Wolte P; Knieling S; Wohlenberg C; Gieselmann V; Galla HJ; Matzner U Comparison of Five Peptide Vectors for Improved Brain Delivery of the Lysosomal Enzyme Arylsulfatase A. J. Neurosci. 2014, 34, 3122–3129. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R47] (47).Meng J; Agrahari V; Youm I Advances in Targeted Drug Delivery Approaches for the Central Nervous System Tumors: the Inspiration of Nanobiotechnology. J. Neuroimmune Pharmacol 2017, 12, 84–98. [DOI] [PubMed] [Google Scholar]

[R48] (48).Grimmer T; Goldhardt O; Guo L-H; Yousefi BH; Forster S; Drzezga A; Sorg C; Alexopoulos P; Forstl H; Kurz A; et al. LRP-1 Polymorphism is Associated with Global and Regional Amyloid Load in Alzheimer’s Disease in Humans in-vivo. Neuroimage Clin 2014, 4, 411–416. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R49] (49).Patel P; Shah J Role of Vitamin D in Amyloid Clearance via LRP-1 Pregulation in Alzheimer’s Disease: A Potential Therapeutic Target? J. Chem. Neuroanat. 2017, 85, 36–42. [DOI] [PubMed] [Google Scholar]

[R50] (50).Sweeney MD; Sagare AP; Zlokovic BV Blood-Brain Barrier Breakdown in Alzheimer Disease and other Neurodegenerative Disorders. Nat. Rev. Neurol. 2018, 14, 133–150. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R51] (51).Zlokovic BV Neurovascular Pathways to Neurodegeneration in Alzheimer’s Disease and other Disorders. Nat. Rev. Neurosci. 2011, 12, 723–738. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] (52).Wiesehan K; Willbold D Mirror-image Phage Display: Aiming at the Mirror. ChemBioChem 2003, 4, 811–815. [DOI] [PubMed] [Google Scholar]

[R53] (53).van Groen T; Kadish I; Wiesehan K; Funke SA; Willbold D In vitro and in vivo Staining Characteristics of Small, Fluorescent, Aß42-binding D-enantiomeric Peptides in Transgenic AD mouse models. ChemMedChem 2009, 4, 276–282. [DOI] [PubMed] [Google Scholar]

[R54] (54).Wiesehan K; Buder K; Linke RP; Patt S; Stoldt M; Unger E; Schmitt B; Bucci E; Willbold D Selection of D-amino-acid Peptides that Bind to Alzheimer’s Disease Amyloid Peptide Aß1–42 by Mirror Image Phage Display. ChemBioChem 2003, 4, 748–753. [DOI] [PubMed] [Google Scholar]

[R55] (55).Rozema DB; Ekena K; Lewis DL; Loomis AG; Wolff JA Endosomolysis by Masking of a Membrane-Active Agent (EMMA) for Cytoplasmic Release of Macromolecules. Bioconjugate Chem. 2003, 14, 51–57. [DOI] [PubMed] [Google Scholar]

[R56] (56).Schindelin J; Arganda-Carreras I; Frise E; Kaynig V; Longair M; Pietzsch T; Preibisch S; Rueden C; Saalfeld S; Schmid B; Schmid B; Tinevez JY; White DJ; Hartenstein V; Eliceiri K; Tomancak P; Cardona A Fiji: an Open-Source Platform for Biological-Image Analysis. Nat. Methods 2012, 9, 676–682. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

A Combination of Tri-Leucine and Angiopep-2 Drives a Polyanionic Polymalic Acid Nanodrug Platform Across the Blood–Brain Barrier

Liron L Israel

Oliver Braubach

Anna Galstyan

Antonella Chiechi

Ekaterina S Shatalova

Zachary Grodzinski

Hui Ding

Keith L Black

Julia Y Ljubimova

Eggehard Holler

Abstract

Graphcial Abstract

RESULTS AND DISCUSSION

Synthesis and Characterization of Nanoconjugates.

Table 1.

Figure 1.

Table 2.

Grouping of Conjugates with Different ξ-Potentials.

Composition.

Biodistribution and Drug Toxicity Studies.

Characterization of Nanoconjugate Fluorescence in Brain Parenchyma and Vascular Labeling.

Figure 2.

Concentration-dependent BBB Penetration of P/LLL/AP2/rh.

Figure 3.

BBB Penetration Depends on Nanoconjugate Composition.

Figure 4.

Screening BBB-Penetrating Peptide Moieties.

Figure 5.

Nanoconjugate Pharmacokinetics in Blood and Brain.

Figure 6.

Estimating Drug Concentration in the Brain Parenchyma Based on Optical Ratio Measurements.

Figure 7.

BBB Penetration of a D1 Carrying Version of P/LLL/ AP2/rh.

Figure 8.

P/LLL/AP2/rh Is Not Internalized by Neurons or Glia.

CONCLUSION

METHODS

Materials.

Calculation of Molecular Masses.

Synthesis of PMLA/LLL(40%)/MEA(10%) (LLL Containing Pre-conjugate).

Synthesis of PMLA/MEA(10%) (Preconjugate Not Containing LLL).

General Procedure for Thiol Quantification using Ellman’s Reagent.

Synthesis of Angiopep-2-PEG3400-maleimide.

Synthesis of Peptide-PEG2000-maleimide.

Synthesis of PMLA/LLL(40%)/peptide(2%)/rh(1%).

Synthesis of PMLA/LLL/AP2(2%)/M4(2%)/rh(1%).

Preparation of PMLA/LLL/Angiopep-2(2%)/D1 (2%)/rh(1%).

Calibration Curve Preparation for Peptide-PEG-maleimide.

Synthesis of PMLA/Peptide(2%)/rh(1%) Conjugate.

SEC-HPLC Analysis.

Rhodamine Quantification of the Final Nanoconjugates.

ζ Potential.

Dynamic Light Scattering.

Chemical Characterization.

FTIR Measurements.

1H NMR Measurements.

Energy Calculations and 3D Images.

Animal Drug Injections.

Retroorbital Blood Collection and Tissue Collection.

Pharmacokinetic (PK) Measurements Using Serum.

Biodistribution and Organ Imaging Experiment.

Tissue Processing and Staining.

Neuronal and Glial Counterstaining.

Image Acquisition and Optical Analysis.

Analysis of Optical Data.

Supplementary Material

ACKNOWLEDGMENTS

Footnotes

REFERENCES

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

¹H NMR Measurements.