Over the past few decades, medical imaging technologies have experienced dramatic growth and currently play a critical role in clinical oncology.[1, 2] However, the true potential of imaging in the clinical management of cancer patients has yet to be realized. With the advent of new and improved imaging techniques, such as molecular imaging, clinicians will not only be able to detect the location of a tumor in the body but also envisage biological processes and the expression and activity of specific molecules that influence the response to therapy and behavior of various tumors. Access to such knowledge is predicted to have a major impact on the detection and diagnosis of cancer, therapeutic development, and personalized treatment, in addition to dramatically improving researchers understanding of cancer.
One such area is the measurement of the extracellular pH (pHe) in solid tumors. For many years researchers have known the importance of the pHe in relation to cancer morbidity and mortality. A low pHe has been identified as an important factor in producing more aggressive cancer phenotypes and causing metathesis of the primary carcinoma, both of which are leading causes of cancer morbidity and mortality.[3, 4] As such, the ability to measure the pHe of solid tumors using non-invasive and accurate techniques that also provide high spatiotemporal resolution has become increasingly important and is of great interest to clinicians (Fig. 1).
While numerous diagnostic imaging modalities have been used in an attempt to measure tissue pH in vivo, including positron emission tomography (PET), optical imaging (OI), magnetic resonance spectroscopy (MRS), and magnetic resonance imaging (MRI),[5-8] there is currently no clinical method available for the in vivo determination of the pHe of tumors. Traditionally, pHe imaging has utilized low molecular weight molecules; however, these materials have limited application due to their fast elimination kinetics, non-specific nature, and the difficulty in determining their in vivo concentration. One of the most promising new approaches to imaging the pHe of tumors involves the use of pH-responsive polymers.
This perspective will highlight recent research in the application of pH-responsive organic polymers for the synthesis of the next generation of diagnostic imaging agents for detecting acidic biological environments in tumors.
The physiological microenvironment of solid tumors is normally characterized by poor perfusion and high metabolic rates.[3, 10] As a result, many regions within the tumor are chronically or transiently acidic and hypoxic (Fig. 1). An increase in the glycolytic rate within tumors leads to the formation of lactate (the conjugate base of lactic acid). In conjunction to this, carbonic acid (H2CO3) is produced by hydration of carbon dioxide (CO2), which is formed by oxidation in the tricarboxylic acid cycle. The excess lactate and carbonic acid produced is removed from the cells via several transport systems, resulting in an intracellular pH (pHi) that is maintained at a normal or alkaline pH when compared to normal cells. However, poor perfusion in tumors decreases the ability to remove the metabolic acids once they leave the cells and also results in regional hypoxia. The net result of these processes is a reduction of the pHe in solid tumors from 7.4 to approximately 6.7. Despite the acidic microenvironment in solid tumors, the tumor cells seem to be well suited to the environment and, in fact, in vitro studies have shown that tumor cells have maximum proliferation at a pHe of 6.8, in comparison to normal cells that require a pHe of 7.3. Ultimately, the tumor microenvironment leads to the development of tumor cells adapted to survive in an acidic microenvironment, where normal cells would die.
For many years researchers have known the importance of the pHe in relation to cancer morbidity and mortality. It has been shown that a low pHe is associated with tumorigenic transformation, chromosomal rearrangements, induction of growth factors and proteases, extracellular matrix breakdown, and increased migration and invasion. Consequently, there has been a significant focus on developing new non-invasive and accurate diagnostic techniques that provide the ability to measure the pHe of tumors in vivo.
Over the past few decades, pH-responsive polymer systems have received much attention for use in biomedical applications.[11-13] These polymer systems demonstrate solubility, volume, or chain conformation changes in aqueous solutions as a result of environmental changes in pH. The chemical structures of pH-responsive polymers usually include ionizable groups in their backbone, side group, or end group that demonstrate pH-dependent physico-chemical properties. These ionizable functional groups are capable of releasing or accepting protons upon environmental pH changes. Thus, electrostatic interactions between generated charged polymer chains repulse each other, which cause the polymer chains to extend or collapse and show an alternation between hydrophobic and hydrophilic behavior. Depending on their charge, these systems are usually classified as anionic or cationic pH-responsive polymers.
Most of the anionic pH-responsive polymers are polyacids, such as poly(acrylic acid) (PAA), poly(methacrylic acid) (PMAA), and sulfonamide-based polymers, which will release protons and become soluble when the environmental pH values is higher than their pKa values. In contrast, cationic pH-responsive polymers are usually polybases which contain amino or amine groups, such as poly((diisopropylamino)ethyl methacrylate) (PDPAEMA), poly(4-vinyl pyridine) (P4VP), poly(histidine), in which the functional group will accept a proton and become hydrophilic when the environmental pH values are lower than their pKa values. Since the pKa value of pH-responsive polymers is a function of their ionizable groups, adjusting the nature of the functional groups in the polymers can effectively modulate the pKa.
According to the pH profiles in body and the properties of pH-responsive polymers, pH-responsive drug delivery systems are usually designed to target or be triggered at the following two sites: the extracellular matrix of solid tumor tissues, which has a pH of 6.2–6.8, and the intracellular endosomes or lysosomes of other cells, which have a pH of 4.5–6.5. The majority of reported pH-responsive delivery systems have pKa values lower than 6.0, which are suitable only for intracellular targeting and delivery after cellular uptake. However, these intracellular pH-responsive systems are not specific to cancer cells because normal cells also have the same endosomal and lysosomal acid pH and, in addition, many cancer cells have pHi values that are neutral or slightly alkaline. Consequently, designing new pH-responsive polymer systems to specifically target the weakly acidic extracellular matrix microenvironment of solid tumors plays an important role in the development of pH-responsive imaging agents for the detection of tumors.
pH-RESPONSIVE POLYMERS FOR IMAGING pHe IN TUMORS
As mentioned previously, numerous biomedical imaging modalities have been investigated for pHe imaging in solid tumors. However, in the cases where pH-responsive polymers have been the critical component of the imaging agent, research has centered on either OI via fluorescent imaging or MRS/MRI.
By taking advantages of variations in the fluorescent properties of probes in response to the local pH, optical measurements can be converted to pH distribution data and used to measure the pH of tumors in vivo. Fluorescence-based pH sensing has been demonstrated using a variety of different materials, including quantum dots, fluorescent proteins, and other small molecule dyes, utilizing intramolecular or intermolecular energy transfer mechanisms. Recently there has been increasing interest in the use of polymeric fluorescent probes due to their increased retention times in vivo, optical signal amplification, and the ability to be functionalized with targeting moieties and biocompatible segments.[14, 16-18]
Gao and coworkers have recently demonstrated the ability to prepare tunable, pH-activatable micellar nanoparticles based on the supramolecular self-assembly of ionizable block copolymers [Fig. 2(a)]. The pH-responsive segment of the polymer utilized various amino groups to provide tunable hydrophobic groups in the physiological pH range of interest (5.0–7.0). In addition, a poly(ethylene oxide) (PEO) block was incorporated to stabilize the micelles and improve the biocompatibility of the nanoparticles. In order to introduce OI capabilities, the pH-insensitive dye tetramethyl rhodamine (TMR) was conjugated to the amino containing block. At higher pH values, the amino blocks are hydrophobic and self-assemble to form the hydrophobic cores of the micelle nanoparticles. This results in aggregation of the dye and quenching of the fluorescent signal via the Förster resonance energy transfer (FRET) and photoinduced electron transfer mechanisms. Conversely, at lower pH values, the amino blocks become protonated and hydrophilic, leading to the micelle disassembling and a large increase in the fluorescence emission (up to 55 times more intense). In essence, this produced an OI imaging agent based on a nonlinear optical responsive nanosystem that is controlled by changes in physiological pH.
In an alternative approach, that also utilized small molecule dyes, Wolfbeis and coworkers used both a pH-sensitive dye, fluorescein isothiocyanate (FITC), and a pH-insensitive dye, tetrakis(pentafluorophenyl) porphyrin (TFPP), in conjunction with the commercially available triblock copolymer Pluronic F-127. This copolymer is a nonionic surfactant composed of a central hydrophobic block of poly(propylene oxide) (PPO) and two hydrophilic outer blocks of PEO that forms micelle structures in aqueous solution. The FITC dye was attached to the end groups of the copolymer so that it was present on the surface of the micelle and, thus, exposed to the external environment. Whereas, the hydrophobic TFPP dye was located in the hydrophobic core of the micelle and used as a reference signal. These micelles demonstrated a large change in the green luminescence signal, increasing between pH 5 and 8, due to the pKa value of 6.4 for the FITC dye, while the red luminescence signal from the TFPP dye is not sensitive to pH. As a result, the system demonstrated a distinct color change from red to green as the pH was increased from 3.0 to 9.0.
In another study, Wang and coworkers took advantage of the pH dependent change in redox state of dopamine to prepare a pH-responsive conjugated polymer. In aqueous media, dopamine has been shown to reversibly convert between hydroquinone (reduced state) and quinone (oxidized state) with changes in pH. By attaching dopamine to a polyfluorene derivative, the fluorescence of the conjugated polymer backbone can be activated by changes in pH. In acidic environments, the dopamine primarily exists as the hydroquinone form, which lacks the ability to quench the polymer fluorescence, due to the absence of electron transfer from the polymer main chain to the hydroquinone. Under basic conditions, however, the dopamine is predominantly in the quinone form, allowing for efficient electron transfer and quenching of the fluorescence from the polymer main chain. Using this system, Wang and coworkers demonstrated a linear correlation between the maximum fluorescent intensity of the polymer and the solution pH over the pH range of 5.0–9.0.
Wolfbies and coworkers used a different method to prepare a pH-responsive fluorescent imaging agent. A ratiometric fluorescent nanogel capable of sensing pH in the range of 6.0–8.0 was synthesized using a polyurethane hydrogel made pH-sensitive by incorporating the pH-indicator bromothymol blue (BTB). The fluorescent properties were introduced by the addition of two standard fluorophores, coumarin 6 and Nile Red, that undergo efficient FRET inside the nanogel. The sensing capabilities of this system relied upon both efficient FRET between the two fluorophores and spectral overlap of the absorption of BTB with the two fluorophores. This resulted in the red fluorescence of Nile Red at a pH of 6.0 and the green fluorescence of coumarin 6 at a pH of 8.0. By using a ratio of the absorption maximums of both Nile Red and coumarin 6, pH values between 6.0 and 8.0 can be determined.
Magnetic Resonance Imaging/Spectroscopy
The most studied techniques for the non-invasive measurement of pHe and pHi are based on magnetic resonance (MR). Both endogenous and exogenous low molecular weight MR active compounds have been used to measure pH in vivo. MRS methods are typically based on the difference in chemical shift between pH dependent and independent resonances. The use of MRS for pH measurement is centered on the concept that protonation reactions are in fast exchange when they are in same compartment, allowing for chemical shifts to be used to predict pH.
Alternatively, an MR image is generated from the nuclear magnetic resonance (NMR) of water protons. The observed contrast in MRI essentially depends on factors such as the water proton density, the longitudinal relaxation time (T1), and the transverse relaxation time (T2) of these protons. Typically, contrast agents (CAs) are used in MRI to aid in diagnostic imaging by increasing the contrast between the particular organ or tissue of interest and the surrounding tissues in the body. For these CAs to be effective, they must have a strong local effect on T1 and T2. Over the past decade, considerable efforts have been made in researching the development of low molecular weight MRI CAs that are pH-responsive.[6, 22, 23] These MRI CAs are significantly different than conventional MRI CAs in that, while high relaxivity values are desirable, the critical property of a pH-responsive MRI CA is the relative change in relaxivity on exposure to changes in the pH of the surrounding environment. Typically pH-responsive MRI CAs function as a result of either water accessibility and correlation times that are pH dependent or via pH-sensitive chemical exchange saturation transfer (CEST).[23, 24]
Lee and coworkers were one of the first groups to prepare a pH-responsive MRI CA based on a pH-responsive polymer. In this case, a diblock copolymer consisting of PEO as the hydrophilic, biocompatible segment, and a poly(β-amino ester) (PAE) as the pH-sensitive segment, was used to form micelle nanoparticles [Fig. 2(b)]. To enable their use as an MRI CA, iron oxide nanoparticles were encapsulated in the hydrophobic core of the micelle. The pH-sensitive nature of the MRI contrast was demonstrated by monitoring the T2 while changing the pH from 7.6 to 6.2. The results demonstrated that the transverse relaxivity increased when the pH was lower than 6.8 due to the release of the iron oxide nanoparticles from the core of the micelles. This release was driven by the change of the PAE from hydrophobic to hydrophilic, and thus dissolution of the micelles, at a pH of 6.8. In addition to the in vitro studies, this is one of the few systems to have been tested in an animal model. Using a mouse model with a subcutaneous tumor, the authors demonstrated that the T2-weighted MRI images of the pH-responsive systems show an increased signal in the tumor over a 24-h period while a pH-insensitive system demonstrated no change in signal.
A recent report from Kikuchi and coworkers used a different approach in the development of a pH-responsive MRI CA. In this study, the authors took advantage of the morphological changes in both a pH-responsive polymer and a pH-responsive crosslinked polymer nanoparticle to create a pH-responsive MRI CA (Fig. 3). These materials were synthesized using methacrylic acid as the monomer and N,N′-methylenebisacrylamide as the crosslinker. An amine-modified gadolinium (Gd) chelate was then attached via the carboxylic acid functionality to produce the pH-responsive MRI CA. In each case, MRI studies indicated that the longitudinal relaxivities of the CAs increased with a decreasing pH. This change can be attributed to a decrease in the molecular tumbling of the water molecules as a result of the conformation change in the polymer with pH. The results also demonstrated that the crosslinked systems provided the largest changes in relaxivity with pH and, thus, had the most potential for application as a pH-responsive MRI CA.
Bae and coworkers have also recently reported a pH-responsive MRS probe. Once again; this system was based upon the pH-dependent formation and dissolution of polymeric micelles. However, in this case the pH-induced transformation of the micelles allowed for pH imaging by a pH-induced “on/off” sensing of NMR and MRS. The pH-responsive block copolymers in this study contained a hydrophilic and biocompatible PEO block and either a poly(l-histidine) (PHis) or poly(l-lactic acid) (PLLA) block [Fig. 2(c)]. PEO-b-PHis has been shown to form pH-sensitive micelles in basic solutions but is physically dissociated in acidic conditions. The authors demonstrated that incorporating PEO-b-PLLA into this system, to produced a mixed micelle structure, provided some degree of control over the pH at which the micelles dissociate. 1H MRS was then used to show a pH-dependence on the chemical shift of protons on the PHis block of the copolymer.
FUTURE DIRECTIONS AND CHALLENGES
Despite the remarkable research that has already been conducted in the development of pH-responsive imaging agents for the next generation of cancer diagnostics, there is much work to be performed before these systems reach the clinic. The vast majority of current reports on these imaging agents as OI or MRS/MRI CAs have focused on in vitro studies. The translation of these materials from the lab to the clinic will require comprehensive in vivo studies investigating the pharmacological and toxicological properties of the imaging agents. In addition to translation of pH-responsive imaging agents from the lab to the clinic, researchers are constantly expanding applications and functionality of these systems. For example, the use of hydrophobic to hydrophilic transitions in pH-responsive polymers for use in drug delivery has been investigated extensively. By combining both of these areas, theranostic nanomedicines (i.e., nanomedicines containing both diagnostic and therapeutic components) can be prepared.
To summarize, there is a clinical need for the development of new techniques that enable the non-invasive in vivo determination of tumor pHe due to the correlation between low tumor pHe and increased cancer morbidity and mortality. While there has been a significant amount of research into the measurement of pHe in vivo using low molecular weight materials, to date all of the techniques developed suffer from significant limitations that have precluded their use clinically. The use of pH-responsive polymers for the preparation of imaging agents offers tremendous potential in the area of pHe imaging. Recent research has started to realize this potential in the areas of OI and MRS/MRI. However, further work is required, particularly in the area of in vivo animal studies, before these new imaging agents can reach the clinic.
The authors would like to thank the Colorado Office of Economic Development for their financial support of Liping Zhu.
Liping Zhu received her B.Sc. in Polymer Materials and Engineering in 2003 and M.Sc. in Materials Processing and Engineering in 2009 from Donghua University in Shanghai. She joined Dr. Stephen G. Boyes' research group in 2009 as a Ph.D. student. She is currently a fourth year graduate student and her research interests focus on design and synthesis of novel biocompatible polymer systems and functionalization of nanoparticles for cancer diagnostics and therapy.
Patrizia P. Smith received a B.A., cum laude, in Chemistry from the University at Buffalo (Buffalo, NY) and is currently working toward a Ph.D. in Applied Chemistry under the advisement of Dr. Stephen G. Boyes. Her current research interests include the development of tailored biopolymers as functional scaffolds for bone tissue engineering as well as the synthesis of multifunctional nanomedicines for targeted diagnostics and therapy.
Stephen G. Boyes Received his Ph.D. in Chemical Engineering from the University of New South Wales and was a postdoc at the University of Akron. Currently, he is an Associate Professor in the Department of Chemistry and Geochemistry at the Colorado School of Mines. His research interests focus on polymers for surface modification, polymer brushes, living radical polymerizations, nanomaterials for diagnostics and therapy, and hybrid organic–inorganic solar cells.