Positron Emission Tomography Imaging of Lesions pH Using 11 C-Labeled Bicarbonate

Abstract

Objectives:

As acid–base imbalance is involved in many pathological processes, the capability to image tissue pH alterations in the clinic could offer new ways to detect disease and respond to treatment. In this study, the authors show that tissue pH can be imaged in vivo with ¹¹C-labeled bicarbonate (H¹¹CO₃ ⁻) buffer and positron emission tomography (PET).

Methods:

H¹¹CO₃ ⁻ was produced by on-column NaOH adsorption. Biodistribution of H¹¹CO₃ ⁻ in normal mice was determined. In addition, uptake studies and inhibition experiments of H¹¹CO₃ ⁻ in the S180 fibrosarcoma-bearing mice and the inflammatory mice were investigated with PET imaging. The tumor and inflammatory interstitial pH was measured by a needle pH microelectrode.

Results:

PET imaging demonstrated the high uptake of H¹¹CO₃ ⁻ in mice tumor tissues and inflammatory tissues, which showed that the average tumor or inflammatory interstitial pH was significantly lower than the surrounding tissue. Administration of sodium bicarbonate in the drinking water increased the measured tumor pH, while the uptake of H¹¹CO₃ ⁻ in mice model tissues had no change. Similarly, administration with ammonium chloride (NH₄Cl) decreased the pH, whereas the unchanged uptake of H¹¹CO₃ ⁻ in mice model tissues was also found. However, after administration of acetazolamide, the low uptake of H¹¹CO₃ ⁻ in mice model tissues was observed.

Conclusions:

H¹¹CO₃ ⁻ solution is an endogenous bicarbonate buffer tracer that can be injected into patients without toxicity. H¹¹CO₃ ⁻ PET can be used clinically to image pathological processes that are associated with acid–base imbalance, such as cancer and inflammation.

Introduction

Maintenance of acid–base homeostasis is critical in mammalian tissues.¹ Intracellular pH (pHi) and extracellular pH (pHe) are regulated in a dynamic steady state driven by an open and dynamic bicarbonate (HCO₃ ⁻)–carbon dioxide (CO₂) system. Although this balance is quite robust, it can be altered in many pathological states, such as ischemia, renal failure, inflammation, cancer, and chronic obstructive pulmonary disease.^2,3 Studies over the last few decades have demonstrated that the pHi of malignant solid tumors is maintained within a range of 7.0–7.2,¹ and the pHe of normal tissues is significantly more alkaline in the range of 7.2–7.5,⁴ whereas the pHe of solid tumors is acidic within a range of 6.5–6.9⁴ and can be correlated with prognosis and response to treatment. In particular, an acidic tumor environment plays a significant role in tumor progression and is often associated with increased invasion and metastasis, as well as resistance to drug therapies.⁵ Therefore, it will become increasingly important to be able to measure acid–base imbalance in vivo with accuracy, precision, and high spatiotemporal resolution in experimental preclinical systems and in human beings.

Current methods to assess acid–base balance in patients are limited to systemic monitoring (e.g., blood gasses, urine pH, and so on) and cannot evaluate regiospecific imbalances that may occur. Delineation of pHi is impossible with electrode-based techniques due to the scale of the electrode compared with the tissue.⁵ In recent years, noninvasive measurement of tissue pH has been developed that can assess the pHi and/or pHe of tissues. The most recent advances in the in vivo measurement of tumor pH focus on molecular probes developed for magnetic resonance spectroscopy (MRS) and magnetic resonance imaging (MRI), nuclear medicine imaging, and optical imaging.⁶

Several magnetic resonance-based methods for the estimation of pH in vivo have been developed in recent years. These approaches include the use of T1-weighted MRI with extracellular molecular probes that exhibit pH-dependent ¹H and gadolinium (Gd³⁺) relaxation enhancement,^3,6,7
1H-MRS,^3,6,7,19F-MRS,^3,8,9 Y-MRS,^6,7 and ³¹P-MRS^6,7 pH- dependent chemical shift measurements of exogenous extracellular species, and ¹³C-MRS chemical shift imaging of hyperpolarized ¹³C-labeled bicarbonate⁴ and [1-¹³C]pyruvate.^8,9 Recently, ⁶⁴Cu-labeled (1,4,7,10-tetraazacyclododecane-1,4,7,10- tetraacetic acid) (DOTA) pH low insertion peptide (pHLIP) (⁶⁴Cu-DOTA-pHLIP)^5,6 and ¹⁸F (or ⁶⁸Ga)-labeled pHLIP¹⁰ positron emission tomography (PET) probes, fluorescent pHLIP probes,^10,11 and ^99mTc-labeled pHLIP (^99mTc-AH114567)¹² single-photon emission computed tomography (CT) probes have been developed for pH imaging in vivo. At present, however, these techniques cannot be readily translated to the clinical use. In addition, although ¹¹C-labeled probes such as ¹¹CO₂ and ¹¹C-labeled dimethyl oxazolidinedione based on the distribution of neutral and ionized species have been used for patient PET imaging,⁵ there are some limitations for hindering their further clinical use.²

Given that H¹¹CO₃ ⁻ solution is an endogenous HCO₃ ⁻ buffer, the authors propose that this PET tracer can image the lesion pH in vivo, which can be readily translated into clinical PET imaging. In this study, the authors describe H¹¹CO₃ ⁻ buffers as nontoxic PET probes that exploit the endogenous mammalian physiological buffering system for the in vivo imaging of acid–base imbalance, such as tumor and inflammatory lesions.

Materials and Methods

General

All chemicals obtained commercially were used without further purification unless otherwise indicated. Sep-Pak plus C18 cartridges were obtained from Waters Corporation (Milford, MA), which were preconditioned with ethanol (10 mL) and water (10 mL) before use. Radioactivity was determined using a calibrated ion chamber (Capintec CRC-15R; Capintec, Inc., Ramsey, NJ) or a γ counter (GC-1200; USTC Chuangxin Co. Ltd., Zonkia Branch, China). PET-CT images were performed using a GEMINI GXL scanner (Philips, Netherlands).

Cell culture and animal models

The S-180 fibrosarcoma cell line and the Kunming mice were obtained from the Laboratory Animal Center of Sun Yat-Sen University. Mice were housed 5 animals per cage, 5 week old and weighed 18–24 g, under standard laboratory conditions at 25°C and 50% humidity. S-180 fibrosarcoma cells (1.0–2.0 × 10⁷) were injected subcutaneously into the right shoulder blade and allowed to grow for 1–3 weeks till the tumor reached 5–10 mm (diameter). The mice were subcutaneously injected with 0.2 mL of turpentine oil (Sigma-Aldrich) into the shoulder area, and inflammation was allowed to develop for 3–4 d before the PET study. The Institutional Animal Care and Utilization Committee of the First Affiliated Hospital, Sun Yat-Sen University (approval no. 2012.001), approved the studies. All efforts were made to minimize animal suffering, to reduce the number of mice, and to utilize alternatives to in vivo techniques, if available.

Automated synthesis of ¹¹C-labeled bicarbonate

¹¹C-Carbon dioxide (¹¹CO₂) was prepared using an IBA Cyclone 10/5 cyclotron (IBA Technologies, Louvain-la-Neuve, Belgium) by the ¹⁴N(p,α)¹¹C nuclear reaction, delivered to the radiochemical laboratory, and trapped in a loop ring cooled at −160°C with liquid nitrogen. The synthetic route of ¹¹C-labeled bicarbonate was shown in Scheme 1. The automated synthesis was performed using PET-CS-II-IT-I ¹¹CH₃I synthesizer and PET-CS-I-IT-I ¹¹C-tracer synthesizer (Beijing PET Co., Beijing, China) (Fig. 1). In brief, after the cooling bath was removed and the trap was heated to ambient temperatures, the ¹¹CO₂ under a helium flow (20 mL/min) was bubbled into a Sep-Pak plus C18 cartridge loaded with 0.50 M NaOH solution. The cartridge was eluted with 0.025 M HCl or 0.3 M NaH₂PO₄ solution (pH 3–4) in bottle 1 through a 0.22 mm sterile filter, to give the desired form of ¹¹C-labeled bicarbonate solution. By varying pH of the ¹¹C-labeled bicarbonate solution, ¹¹C-labeled injection was either as ¹¹C-CO₂ at a pH below 5 or as H¹¹CO₃–¹¹CO₂ solution within a pH range of 7–8 or as H¹¹CO₃ ⁻ (or ¹¹CO₃ ²⁻) at a pH above 8.¹³ Radiochemical purity of ¹¹C-labeled bicarbonate was detected with silica plate-thin layer chromatography (TLC) eluted with methanol (¹¹C-carbonate or ¹¹C-bicarbonate, Rf = 0).

FIG. 1.

Biodistribution of H¹¹CO₃ ⁻ solution (pH 7–8) in normal mice.

Ex in vivo biodistribution

The 20 normal Kunming mice were divided into five groups (n = 4 per group), each mouse was injected with ∼3.7 MBq (∼100 μCi) of ¹¹C-labeled bicarbonate in 100–200 μL of 0.9% NaCl solution through the tail vein. At 5, 10, 30, 45, and 60 min postinjection of the radiotracer, the mice (n = 4 per each time point) were sacrificed by cervical fracture and dissected. Blood was obtained through the eyeball vein; tissue samples of interest (blood, brain, heart, lung, liver, kidney, pancreas, stomach, small intestine, and muscle in the right thigh) were rapidly dissected and weighed, and radioactivity was counted with a gamma (γ) counter. All measurements were background-subtracted and decay-corrected to the time of injection and then averaged. Data were expressed as a percentage of the injected dose per gram of tissue (% ID/g).

Intraperitoneal administration of NaHCO₃/NH₄Cl

Two groups of S-180 fibrosarcoma-bearing mice or inflammatory mice (n = 4, each group) were used. One group was administered a single dose of either 0.7 mL 1 M NaHCO₃ or 0.7 mL 1 M NH₄Cl (intraperitoneal [i.p.] administration) and denied access to drinking water for the next 4 h, whereas the other group did not receive any treatment and had uninterrupted access to ad libitum water.¹⁴ At 5 min and 4 h after treatment with NaHCO₃ or NH₄Cl, PET imaging of tumor-bearing mice was performed with H¹¹CO₃ ⁻ buffer.

Intravenous administration of acetazolamide

Two groups of S-180 fibrosarcoma-bearing mice or inflammatory mice (n = 4, each group) were used. One group was administered a single dose of 50 mg/kg (intravenous administration) at a concentration of 50 mg/mL acetazolamide (ACT) (carbonic anhydrase [CA] inhibitor) and denied access to drinking water for the next experiment. At 45 min after treatment with ACT, PET imaging of tumor-bearing mice was performed with H¹¹CO₃ ⁻ buffer.

PET imaging

The mice were injected with 3.7 MBq (100 μCi) of H¹¹CO₃ ⁻ solution in 0.2 mL of 0.9% NaCl solution through the tail vein. They were anesthetized with 5% chloral hydrate solution (6 mL/kg) before imaging and remained in anesthesia state during the study. The CT scan was used for attenuation correction and localization of the lesion site. Mice were visually monitored for breathing and any other signs of distress throughout the entire imaging period. PET images were acquired at four-time points (10, 30, 45, and 60 min) postinjection. PET imaging was divided into six groups (three subgroups, and n = 4, each subgroup), including the tumor-bearing mice imaging (H¹¹CO₃ ⁻ solution pH <5, pH 7–8, and pH >8), the antitumor therapy with cyclophosphamide (H¹¹CO₃ ⁻ solution pH >8), the inflammatory imaging (H¹¹CO₃ ⁻ solution pH <5, pH 7–8, and pH >8), the NaHCO₃-treated inflammatory mice (H¹¹CO₃ ⁻ solution pH >8.0 and pH <5.0), the tumor-bearing mice treated with ACT (H¹¹CO₃ ⁻ solution pH <5, pH 7–8, and pH >8), and inflammatory mice treated with ACT (H¹¹CO₃ ⁻ solution pH <5, pH 7–8, and pH >8). The CT scan was used for attenuation correction and localization of the lesion site. A region of interest was placed on each tumor, inflammatory lesion, and a region of femoral muscle. After PET imaging, the animals were sacrificed, the tissues of tumors, inflammation region of the right thigh muscle, and the normal right thigh muscle were excised and weighted, and the radioactivity of these tissues was measured with the gamma counter.

Determination of pH in tumors, inflammatory lesions, and normal muscle

pH in tumor tissues, inflammatory lesion, and contralateral muscle of the model mice were measured using MI-419 needle (pin diameter is 0.8 mm) pH microelectrode and MI-402 reference microelectrode (Microelectrodes, Inc.). After peeling off the tumor, the inflammatory lesion, and the contralateral normal muscle of mice, the needle of pH microelectrode was inserted into the central portion of these tissues and the reference microelectrode placed in the subcutaneous area to measure tissue pH.

Statistical analysis

All data are expressed as mean ± standard deviation. Statistical analysis was performed with SPSS software, version 20.0 (SPSS, Inc.), for Windows (Microsoft). Comparisons between conditions were performed using the unpaired Student t test. A p-value of <0.05 was considered to indicate statistical significance.

Results

Radiosynthesis of H¹¹CO₃ ⁻ solution

Automated synthesis gave high radiochemical yield above 95% for H¹¹CO₃ ⁻–¹¹CO₃ ²⁻ solution (pH >8.3) and over 80% for H¹¹CO₃ ⁻–¹¹CO₂ solution (pH 7–8), but gave relative low yield of <50% for ¹¹CO₂-based solution (pH <6), with the total synthesis time of 8 min. Radiochemical purity of ¹¹C-labeled bicarbonate was detected with TLC eluted with methanol, with ¹¹C-carbonate or ¹¹C-bicarbonate Rf = 0. Three forms of ¹¹C-bicarbonate injected solution with pH <5, pH 7–8, and pH >8.3 were used for the next experiments.

Biodistribution of H¹¹CO₃ ⁻

Three forms of ¹¹C-labeled bicarbonate had the similar biodistribution in mice (data not shown). A representative biodistribution of H¹¹CO₃ ⁻ solution (pH >8.0) is shown in Figure 1. Most tissues showed high uptake of H¹¹CO₃ ⁻ at 5 and 30 min after injection, and all the tissues demonstrated low uptake of H¹¹CO₃ ⁻ at 10, 45, and 60 min after injection. The pancreas had the highest uptake of H¹¹CO₃ ⁻ at 5, 30, and 60 min postinjection. The blood, kidney, and heart possessed the moderate uptake at 5 and 30 min postinjection. The lung showed the moderate uptake of H¹¹CO₃ ⁻ at 5 min postinjection and low uptake at other time point postinjection. Brain, liver, and muscle had low uptake of H¹¹CO₃ ⁻ within all observed time. In addition, H¹¹CO₃ ⁻ demonstrated slow clearance from most tissues.

Small-animal PET imaging

PET images of tumor-bearing mice, cyclophosphamide-treated tumor-bearing mice treated, and inflammatory mice are shown in Figures 2 and 3. PET imaging showed that tumor and inflammatory lesion had a high uptake of H¹¹CO₃ ⁻ in the S180 fibrosarcoma-bearing mice (Fig. 2) and the turpentine-induced inflammatory lesion-bearing mice (Fig. 3). The ratios of tumor to muscle were 2.4, 1.7, 1.3, and 1.0, and the ratios of inflammatory lesion to muscle were 1.5, 1.7, and 2.1. The three forms of H¹¹CO₃ ⁻ solution exhibited the similar results. However, the uptake of H¹¹CO₃ ⁻ (pH >8) in the tumor (pH 6.95 ± 0.05) was lower than that in untreated tumor (pH 7.00 ± 0.05) (Fig. 2).

FIG. 2.

PET images of the tumor-bearing mice with H¹¹CO₃ ⁻ solution [(A) pH <5, (B) pH 7–8, (C) pH >8.3] and the antitumor therapy mice with H¹¹CO₃ ⁻ solution [(D) pH >8.3]. (A–C) Most tumors had a high uptake of H¹¹CO₃ ⁻, (D) whereas the uptake of H¹¹CO₃ ⁻ in the tumor after antitumor therapy was lower compared with the untreated tumor. (E) At 15 min postinjection of H¹¹CO₃ ⁻ solution [(A) pH <5, (B) pH 7–8, (C) pH >8.3, (D) pH >8.3], the uptake value of tumor was higher compared with muscle, and the similar uptake of the treated tumor to muscle was observed, with tumor to muscle uptake ratios of 2.4, 1.7, 1.3, and 1.0, respectively. Circle: tumor. Trans, transect; Coron, coronal section, PET, positron emission tomography.

FIG. 3.

PET images of the inflammatory lesion-bearing mice with H¹¹CO₃ ⁻ solution [(A) pH <5, (B) pH 7–8, (C) pH >8.3]. (A–C) The inflammatory lesion had a high uptake of H¹¹CO₃ ⁻ at different time postinjection of H¹¹CO₃ ⁻ solution. (D) The uptake value of inflammatory lesion was higher compared with muscle at 15 min postinjection of H¹¹CO₃ ⁻ solution [(A) pH <5, (B) pH 7–8, (C) pH >8.3]; the uptake value of inflammatory lesion was higher compared with muscle, with inflammatory lesion to muscle uptake ratios of 1.5, 1.7, and 2.1, respectively. Circle: inflammatory lesion.

Uptake inhibition experiment

After i.p. injection of NaHCO₃ for 5 min and 4 h, PET imaging showed that the inflammatory lesion had high uptake of H¹¹CO₃ ⁻ (pH <5.0 and pH >8.3) after 15 min postinjection of the tracers (Fig. 4). The ratios of inflammatory lesion to muscle were 1.34, 1.40, 1.37, and 1.54. In addition, the group of the NaHCO₃ (or NH₄Cl)-treated tumor demonstrated the similar results as the NaHCO₃-treated inflammatory lesion (data not shown).

FIG. 4.

PET images of inflammatory lesion-bearing mice with H¹¹CO₃ ⁻ solution [(A, C) pH <5, (B, D) pH >8.3] after intraperitoneal injection of NaHCO₃ for 5 min (A, B) and 4 h (C, D). (A–D) The inflammatory lesion treated with NaHCO₃ had a high uptake at different times postinjection of the H¹¹CO₃ solution. (E) At 15 min postinjection of H¹¹CO₃ ⁻ solution, the uptake value of inflammatory lesion was higher compared with muscle, with the inflammatory lesion to muscle uptake ratios of 1.34, 1.40, 1.37, and 1.44, respectively. Circle: inflammatory lesion.

After administration of ACT, low uptake of H¹¹CO₃ ⁻ was found in tumor-bearing mice at 30 min postinjection of H¹¹CO₃ ⁻ solution (pH <5.0 and pH >8.3) (Fig. 5). However, PET images showed that the uptake of H¹¹CO₃ ⁻ solution (pH 6–7) in inflammatory lesion had no significant difference between before and after administration of ACT (Fig. 5) (p > 0.05).

FIG. 5.

PET images of inflammatory lesion-bearing mice in the control group and the acetazolamide-treated mice postinjection of H¹¹CO₃ ⁻ solution [(A) pH <5, (B) pH 7–8, (C) pH >8.3]. (A, C) After 45 min postinjection of H¹¹CO₃ ⁻ solution [(A) pH <5.0 and (C) pH >8.3], the uptake in the acetazolamide-treated inflammatory lesion was significantly lower than that in the untreated inflammatory lesion. (B) At different times postinjection of H¹¹CO₃ ⁻ solution [(B) pH 7–8], almost no uptake in untreated inflammatory lesion and treated inflammatory lesion was found. (D) At 15 min postinjection of H¹¹CO₃ ⁻ solution, the uptake value was no different between untreated inflammatory lesion and acetazolamide-treated inflammatory lesion (p > 0.05). Circle: inflammatory lesion.

In addition, the uptake of H¹¹CO₃ ⁻ (Fig. 6A, pH <5; Fig. 6B, pH >8.3) in tumor of the control group was low, but the uptake of H¹¹CO₃ ⁻ (pH >8.3) in ACT-treated tumor was significantly lower than that in the control group (Fig. 6) (p < 0.05).

FIG. 6.

PET images of the tumor-bearing mice in the control group and the acetazolamide-treated group postinjection of H¹¹CO₃ ⁻ solution [(A) pH <5, (B) pH >8.3]. (A) Low uptake of H¹¹CO₃ ⁻ [(A) pH <5] in the tumor and the acetazolamide-treated tumor was found. (B) The uptake of H¹¹CO₃ ⁻ [(B) pH >8.3] in the acetazolamide-treated tumor was significantly lower than that in the control group. ( C) At 45 min postinjection of H¹¹CO₃ ⁻ solution, (A) there is no significant difference between control group and acetazolamide-treated group (p > 0.05); however, (B) there is significant difference between them (p < 0.05). Circle: tumor.

Ninety minutes postinjection of NaHCO₃ and NH₄Cl in tumor-bearing mice and inflammatory lesion-bearing mice, PET images showed that the uptake of H¹¹CO₃ ⁻ in the NaHCO₃-treated group was low, whereas it was still high in the NH₄Cl-treated group.

PET images showed that tumor had low uptake of H¹¹CO₃ ⁻ after administration of ACT at 30 min postinjection of H¹¹CO₃ ⁻ solution (pH <5.0 and pH >8.3) (Fig. 6). However, PET images indicated that the uptake of H¹¹CO₃ ⁻ (pH 6–7) in inflammatory lesion had no significant difference between untreated group and ACT-treated group (Fig. 5). That is, the uptake of H¹¹CO₃ ⁻ solution (Fig. 6A, pH <5; Fig. 6B, pH >8.3) in tumor was high, but the uptake of H¹¹CO₃ ⁻ (pH >8.3) in ACT-treated tumor was significantly lower than that in the tumor group (Fig. 6) (p < 0.05).

Tumor and inflammatory tissue pH measurement

The average pH values in tumor, inflammatory lesion, and contralateral normal muscle of the model mice are shown in Table 1. Both tumor tissue and inflammation lesion were a bit acidic (pH 6.95) compared with contralateral normal muscle, which was a bit alkaline (pH 7.33). After anticancer therapy with cyclophosphamide, the pH value in tumor was nearly neutral (pH 7.00), while the contralateral normal muscle was also a bit alkaline (pH 7.27).

Table 1.

The Average pH Values in Tumor, Inflammatory Lesion, and Contralateral Normal Muscle of Mice

Tissue	Tumor-bearing mice		Antitumor therapy		Inflammatory mice
Tissue	Tumor tissue	Contralateral muscle	Tumor tissue	Contralateral muscle	Inflammatory tissue	Contralateral muscle
pH	6.95 ± 0.05	7.33 ± 0.03	7.00 ± 0.05	7.30 ± 0.03	6.95 ± 0.08	7.33 ± 0.02

n = 5, mean ± SD.

SD, standard deviation.

Administration of NaHCO₃ in the drinking water increased the measured tumor pH to 7.21 ± 0.03 (Table 2) and the inflammatory tissue pH to 7.23 ± 0.04 (Table 3), while gavage with NH₄Cl over 90 min decreased the tumor pH to 6.51 ± 0.04 (Table 2) and the inflammatory lesion pH to 6.48 ± 0.03 (Table 3). Similarly, administration of ACT decreased the tumor pH to 6.62 ± 0.02 (Table 2) and the inflammatory lesion pH to 6.65 ± 0.05 (Table 3).

Table 2.

pH Values in the Tumor at Forty-Five Minutes After Administration of NaHCO ₃, NH ₄ Cl, and Acetazolamide

	NaHCO₃		NH₄Cl		ACT
Tumor	Before treatment	After treatment	Before treatment	After treatment	Before treatment	After treatment
pH	6.83 ± 0.04	7.21 ± 0.03	6.98 ± 0.05	6.51 ± 0.04	6.82 ± 0.03	6.62 ± 0.02

Mean ± SD.

ACT, acetazolamide.

Table 3.

pH Values in the Inflammatory Lesion at Forty-Five Minutes After Administration of NaHCO ₃, NH ₄ Cl, and Acetazolamide

	NaHCO₃		NH₄Cl		ACT
Lesion	Before treatment	After treatment	Before treatment	After treatment	Before treatment	After treatment
pH	6.85 ± 0.05	7.23 ± 0.04	6.90 ± 0.04	6.48 ± 0.03	6.87 ± 0.02	6.65 ± 0.05

Mean ± SD.

Discussion

¹¹C-bicarbonate buffer was prepared from the on-column reaction of ¹¹CO₂ with 0.50 M NaOH solution loaded in a Sep Pak plus C18 cartridge, followed by neutralization with dilute hydrochloric acid or NaH₂PO₄ solution (pH 3–4). There existed three forms of ¹¹C-bicarbonate buffer, as ¹¹C-CO₂ at a pH below 5 or as H¹¹CO₃ ⁻–¹¹CO₂ buffer within a pH range of 7–8 or as H¹¹CO₃ ⁻–¹¹CO₃ ²⁻ at a pH above 8.3.^{12
11}CO₂-based bicarbonate solution (pH <5.0) gave low radiochemical yield (<50%) because ¹¹CO₂ is easy to escape from the solution. H¹¹CO₃ ⁻–¹¹CO₂ buffer (pH 7–8) contained small amounts of ¹¹CO₂, resulting in high radiochemical yield (>80%). H¹¹CO₃ ⁻–¹¹CO₃ ²⁻ solution gave the highest radiochemical yield of over 95%, due to almost no ¹¹CO₂ produced from the ¹¹C-bicarbonate solution (pH >8.3). Compared with ¹¹CO₂-based solution (pH <5.0), the radiochemical yield of H¹¹CO₃ ⁻–¹¹CO₂ solution (pH 7–8) and H¹¹CO₃ ⁻–¹¹CO₃ ²⁻ solution (pH >8.3) is very high. Thus, ¹¹C-bicarbonate solutions (pH 7.0–8.0 and pH >8.3), as a nontoxic PET tracer, are easy to perform automated radiosynthesis for translation into clinical use.

Biodistribution profiles of three forms of ¹¹C-bicarbonate solution (pH <5, pH 7–8, and pH >8.3) in mice were similar. The pancreas had the highest uptake, the brain had low uptake, and rapid clearance from all organs or tissues was observed after 30 min postinjection of the ¹¹C-bicarbonate solution. Similar biodistribution profiles of three ¹¹C-bicarbonate tracers could be contributed to interconversion between bicarbonate ion and CO₂ in the blood to form the similar H¹¹CO₃ ⁻–¹¹CO₂ buffer postinjection of the tracers. All organs or tissues appeared to be high uptake of ¹¹C-bicarbonate in mice at 5 and 30 min postinjection, respectively. The possible reason was that the red cell membrane and the capillary endothelium were much more permeable to neutral molecules (CO₂) than HCO₃ ⁻.^{15
11}CO₂ gave rise to the first high uptake at 5 min after injection, while H¹¹CO₃ ⁻ gave rise to the second high uptake at 30 min after injection.

The uptake mechanism of three ¹¹C-bicarbonate tracers in an acidic environment possibly involves the following aspects. As the pHe of solid tumor is acidic,^1

–5 the main form of ¹¹C-bicarbonate in an acidic environment is ¹¹CO₂. CO₂ is a weak acid with a pKa of 6.12. As ¹¹CO₂, but not ¹¹C-bicarbonate ion, readily diffuses across intact blood–brain barrier (BBB) or cell membrane, regional cerebral ¹¹C activity following ¹¹CO₂ inhalation reflects the regional cerebral pH (rpH) of the combined intra- and extracellular cerebral compartments; the rpH being weighted toward pHi. Tumors with an intact or a disrupted BBB are also expected to have a raised ¹¹CO₂ solubility compared with normal tissue. Thus, PET with ¹¹CO₂, like ¹¹C-bicarbonate solution (pH <5), can be used for measurement of tumor rpH. A theoretical disadvantage of the ¹¹CO₂ approach is that fixation of ¹¹C activity in the form of nonvolatile substrates can occur.¹⁴ In addition, ¹¹C-bicarbonate anion can easily diffuse across a disrupted blood–tumor barrier into cancer cells, besides that they enter tumor mainly existing as ¹¹CO₂ in an acidic environment in the presence of the enzyme CA.^13,14 Therefore,¹¹C-bicarbonate (pH 7–8 or pH >8.3) is a potential endogenous imaging agent for PET imaging of tumor.

PET images showed that the tumor had high uptake of three ¹¹C-bicarbonate tracers at 15 and 30 min after administration, while tumor showed a low uptake after 45 min postinjection. The inflammatory lesion had high uptake of H¹¹CO₃ ⁻ in PET images within all the observed time. The uptake of H¹¹CO₃ ⁻ (pH >8.3) in the tumor after antitumor therapy was lower than that in the untreated tumor, but a little high uptake of H¹¹CO₃ ⁻ (pH >8.3) in the treated tumor was also found at 60 min postinjection, which possibly was due to the balance of H¹¹CO₃ ⁻–¹¹CO₂ in vivo after 60 min postinjection. The pH measurements showed that the tumor and the inflammatory lesion possessed are a bit acidic (6.95), while the contralateral normal thigh muscle is a bit alkaline (7.33). Compared with the untreated tumor (pH 7.00), pH in the cyclophosphamide-treated tumor was a bit alkaline (pH 7.30), which was closed to the pHe of normal tissues.⁴ However, the limitations of pH measurement in this study are as follows. There must be sampling errors during the measurement of needle use and reproducibility of pH meters, and a lot more research is needed.

After i.p. injection of NaHCO₃ in inflammatory mice for 5 min and 4 h, PET imaging showed that the uptake of ¹¹C-bicarbonate (pH <5 and pH 8.3) in inflammatory lesion was still high. In addition, the similar results were observed in the NaHCO₃ (or NH₄Cl)-treated tumor and in the NH₄Cl-treated inflammatory lesion. The pH measurement demonstrated that administration of NaHCO₃ over 90 min increased tumor pH and inflammatory tissue pH to >7.2, while gavage with NH₄Cl over 90 min decreased tumor pH and inflammatory tissue pH to <6.6. These results indicated that NaHCO₃ could not make tumor pH and inflammatory tissue pH increase by >7.0, and NH₄Cl could not make tumor pH and inflammatory tissue pH decrease to <7.0 for a short time of 5 min. In addition, after treatment with NaHCO₃ or NH₄Cl for a long time of 4 h, NaHCO₃ or NH₄Cl could fully clear from the mice body and result in renewing tumor pH and inflammatory tissue pH to balance (pH <7.0). Therefore, high uptake of ¹¹C-bicarbonate (acid injectate and alkaline injectate) in tumor and inflammatory lesion was possibly closely associated with acidic environment.

PET images also showed that the uptake of ¹¹C-bicarbonate solution (pH >8.3) in tumor and inflammatory lesion after administration of ACT was significantly higher than that in untreated tumor and untreated inflammatory lesion, especially at 45 min postinjection of the tracers. Compared with ¹¹C-bicarbonate (pH >8.3), the uptake of ¹¹C-bicarbonate (pH <5) in tumor and inflammatory lesion had no obvious change before and after treatment with ACT. Mammalian animals have robust pH buffer systems that maintain pH within a narrow optimal pH range (7.35–7.45), and the HCO₃ ⁻–CO₂ system is the principal system. Bicarbonate (HCO₃ ⁻) is rapidly exchanged reverse reaction with CO₂ that is catalyzed by the enzyme CA. Absence of CA or CA inhibition (such as ACT) slows down the interconversion between bicarbonate and CO₂. Based on the reason that administration of ACT could obviously decrease tumor pH and the inflammatory tissue pH, low uptake of ¹¹C-bicarbonate (pH >8.3) in the ACT-treated tumor and ACT-treated inflammatory lesion was found. Therefore, the authors could deduce that uptake of ¹¹C-bicarbonate (alkaline injectate) in the tumor and inflammatory lesion was closely related to acidic environment in the presence of CA.

Conclusions

In conclusion, PET with H¹¹CO₃ ⁻ can be used to image the lesion pH in vivo. ¹¹C-bicarbonate solution as a noninvasive and nontoxic endogenous PET tracer can be easy to perform automated radiosynthesis for translation into clinical use. Thus, H¹¹CO₃ ⁻ could be used to image the lesion pH alteration and the response to therapy in clinical setting.

Footnotes

Acknowledgments

This work was supported, in part, by the National Natural Science Foundation of China (No. 81571704, No. 81371584, No. 81671719, No. 81701723), the Science and Technology Foundation of Guangdong Province (No. 2016B090920087, 2014A020210008, No. 2013B021800264, 2013B010404018), and the Science and Technology Planning Project Foundation of Guangzhou (No. 201504301345364, No. 201510010145, No. 201604020169).

Disclosure Statement

No competing financial interests exist.

References

Zhang

, Lin

, Gillies

. Tumor pH and its measurement. J Nucl Med, 2010; 51:1167.

Gallagher

, Kettunen

, Brindle

. Imaging pH with hyperpolarized 13C. NMR Biomed, 2011; 24:1006.

Gallagher

, Kettunen

, Day

, et al. Magnetic resonance imaging of pH in vivo using hyperpolarized 13C-labelled bicarbonate. Nature, 2008; 453:940.

Robey

, Baggett

, Kirkpatrick

, et al. Bicarbonate increases tumor pH and inhibits spontaneous metastases. Cancer Res, 2009; 69:2260.

Vavere

, Biddlecombe

, Spees

, et al. A novel technology for the imaging of acidic prostate tumors by positron emission tomography. Cancer Res, 2009; 69:4510.

Chen

, Pagel

. Evaluating pH in the extracellular tumor microenvironment using CEST MRI and other imaging methods. Adv Radiol, 2015; 2015:206405.

Hashim

, Zhang

, Wojtkowiak

, et al. Imaging pH and metastasis. NMR Biomed, 2011; 24:582.

Golman

, Zandt

, Lerche

, et al. Metabolic imaging by hyperpolarized 13C magnetic resonance imaging for in vivo tumor diagnosis. Cancer Res, 2006; 66:10855.

Day

, Kettunen

, Gallagher

, et al. Detecting tumor response to treatment using hyperpolarized 13C magnetic resonance imaging and spectroscopy. Nat Med, 2007; 13:1382.

10.

Andreev

, Dupuy

, Segala

, et al. Mechanism and uses of a membrane peptide that targets tumors and other acidic tissues in vivo. Proc Natl Acad Sci U S A, 2007; 104:7893.

11.

Reshetnyak

, Yao

, Zheng

, et al. Measuring tumor aggressiveness and targeting metastatic lesions with fluorescent pHLIP. Mol Imaging Biol, 2011; 13:1146.

12.

Macholl

, Morrison

, Iveson

, et al. In vivo pH imaging with (99m)Tc-pHLIP. Mol Imaging Biol, 2012; 14:725.

13.

Johnson

, Hoop

, Kazemi

. Movement of CO₂ and HCO₃ from blood to brain in dogs. J Appl Physiol Respir Environ Exerc Physiol, 1983; 54:989.

14.

Brooks

, Beaney

, Thomas

, et al. Studies on regional cerebral pH in patients with cerebral tumours using continuous inhalation of 11CO₂ and positron emission tomography. J Cereb Blood Flow Metab, 1986; 6:529.

15.

Endeward

, Cartron

, Ripoche

, et al. Red cell membrane CO₂ permeability in normal human blood and in blood deficient in various blood groups, and effect of DIDS. Transfus Clin Biol, 2006; 13:123.

Positron Emission Tomography Imaging of Lesions pH Using 11 C-Labeled Bicarbonate

Abstract

Objectives:

Methods:

Results:

Conclusions:

Introduction

Materials and Methods

General

Cell culture and animal models

Automated synthesis of 11 C-labeled bicarbonate

Ex in vivo biodistribution

Intraperitoneal administration of NaHCO3/NH4Cl

Intravenous administration of acetazolamide

PET imaging

Determination of pH in tumors, inflammatory lesions, and normal muscle

Statistical analysis

Results

Radiosynthesis of H 11 CO3 − solution

Biodistribution of H 11 CO3 −

Small-animal PET imaging

Uptake inhibition experiment

Tumor and inflammatory tissue pH measurement

Discussion

Conclusions

Footnotes

Acknowledgments

Disclosure Statement

References

Automated synthesis of ¹¹C-labeled bicarbonate

Intraperitoneal administration of NaHCO₃/NH₄Cl

Radiosynthesis of H¹¹CO₃ ⁻ solution

Biodistribution of H¹¹CO₃ ⁻