Nanobody: The “Magic Bullet” for Molecular Imaging?

Abstract

Molecular imaging involves the non-invasive investigation of biological processes in vivo at the cellular and molecular level, which can play diverse roles in better understanding and treatment of various diseases. Recently, single domain antigen-binding fragments known as 'nanobodies' were bioengineered and tested for molecular imaging applications. Small molecular size (~15 kDa) and suitable configuration of the complementarity determining regions (CDRs) of nanobodies offer many desirable features suitable for imaging applications, such as rapid targeting and fast blood clearance, high solubility, high stability, easy cloning, modular nature, and the capability of binding to cavities and difficult-to-access antigens. Using nanobody-based probes, several imaging techniques such as radionuclide-based, optical and ultrasound have been employed for visualization of target expression in various disease models. This review summarizes the recent developments in the use of nanobody-based probes for molecular imaging applications. The preclinical data reported to date are quite promising, and it is expected that nanobody-based molecular imaging agents will play an important role in the diagnosis and management of various diseases.

Introduction

As one of the most exciting and rapidly growing areas of science, molecular imaging has become an indispensable tool in cancer research, clinical trials and medical practice ^1-11. During the last several decades, there has been remarkable progresses in the number of imaging technologies and their applications, with the objective of developing a new generation of platforms with greater accuracy and sensitivity for in vivo characterization and quantification of various biological processes ¹². Using an array of imaging techniques such as positron emission tomography (PET), single photon emission computed tomography (SPECT), computed tomography (CT), magnetic resonance imaging (MRI), optical, ultrasound (US), photoacoustic imaging etc., researchers can now measure various dynamic biological processes in intact living subjects ^{1-3, 6-8, 12}. Molecular imaging provides the opportunity for understanding the underlying biology, early detection and characterization of various diseases, and evaluation of the therapeutic responses. However, no single imaging modality can provide information on all aspects of structure and function. Therefore, investigation of a subject using multiple imaging modalities is highly desirable and has gained increasing popularity ^13-17.

Regardless of the modality used, molecular imaging generally requires accumulation of the signal delivered by a probe at the target site. The major challenge associated with non-invasive in vivo imaging is to achieve a high contrast signal over nearby healthy tissues, in addition to the issues related to biocompatibility, toxicity and probe stability. In order to achieve high target to non-target ratio, the imaging tags are generally coupled with various targeting molecules such as antibodies ¹⁸, peptides ^{19, 20}, small molecule ligands ²¹, aptamers ^{22, 23}, etc. Among these, monoclonal antibodies (mAbs) have long been considered as attractive candidates for both targeted therapy as well as diagnostics due to their exquisite specificity towards cognate antigens. However, the utility of mAbs for imaging is limited by their large size (150 kDa), which leads to long circulation time in blood (e.g. a few days to weeks) and longer time to optimally accrete in the tumor tissue (typically several days). Advancement in antibody engineering has led to improvement in antibody pharmacokinetics without compromising its affinity and specificity ^{24, 25}. In this direction, numerous antibody fragments and variants such as Fab, F(ab´)₂, single chain Fv (scFv), diabodies and minibodies (molecular weight ranging from 25-100 kDa) were bioengineered ^24-26. In addition, the development of several non-traditional protein scaffolds such as domain antibodies, affibodies, nanobodies, and anticalins have been reported ^24-26.

The methods of obtaining engineered antibodies and recombinant antibody fragments, as well as their use as probes or vectors for non-invasive imaging and therapeutic applications, have been extensively reviewed ^{24, 25, 27, 28}. Recently, there has been significant interest in the utilization of nanobodies (derived from heavy chain-only antibodies occurring naturally in Camelidae) for molecular imaging investigations, using modalities such as radionuclide-based, optical and ultrasound imaging ^29-33. In this review, we aim to provide a timely and comprehensive overview of the progress in the use of nanobodies in molecular imaging studies to date.

Nanobodies

Nanobodies are recombinant, single-domain, variable fragments of camelid heavy chain-only antibodies (~95 kDa), which are able to bind selectively to a specific antigen ³⁴. Typically, nanobodies are the variable domain alone of heavy chain antibodies (i.e. V_HH) with approximate molecular weight of 12-15 kDa and are considered the smallest naturally derived antigen-binding fragment (Figure 1). The investigation of the crystal structures of V_HHs revealed a prolate (rugby ball) shape of approximately 2.5 nm in diameter and 4.2 nm in length ^{35, 36}. Because of their size in the nm range, the term 'nanobody' was coined by the Belgian company Ablynx®, which refers to the V_HH domain from Camelidae species ^{29, 35, 36}.

Figure 1.

A schematic representation of nanobody and antibody domains. Adapted from ²⁶.

Nanobodies are much smaller than common intact antibodies (~150 kDa), as well as their fragments such as Fab (~50 kDa) and scFv (~25 kDa) ^{29, 34-36}. The size reduction of an antibody into a nanobody (and the concomitant reduction in valency from bivalent to monovalent) can cause a dramatic change in biological activity, which provides many advantages over conventional antibodies and their recombinant fragments. Firstly, nanobodies are weakly immunogenic in humans because the genes encoding them share high degree of identity with the human type 3 VH domain (VH₃) ³⁴. To aid the clinical translation of nanobodies, further reduction in their immunogenic potential was achieved by Ablynx® via humanization of the nanobodies ^{37, 38}. Humanization of nanobodies has been described in detail by Vincke et al ³⁹. Secondly, nanobodies are easy to clone as they consist of only one domain and can therefore be expressed with high yield ⁴⁰. Thirdly, nanobodies rapidly and specifically bind tumor antigens, whereas the unbound ones are rapidly cleared from the blood mostly by renal elimination ^{34, 37, 41-44}. Therefore, high tumor-to-background ratios can be obtained as early as 1 h after tracer injection ^{29, 42, 45, 46}. Nanobodies also demonstrate high solubility and refolding capacity (stability) even on exposure to extreme conditions such as very low/high pH and temperature ⁴⁷. Lastly, the high variability of length and sequence of V_HHs and the small size allows nanobodies to efficiently enter into tissues and bind epitopes that typically cannot be reached by conventional intact antibodies ^{34, 37}. The CDRs of the nanobodies play a critical role in their stability and binding affinity ^{29, 48-51}. The V_HHs possess an additional disulfide bridge linking the CDRs (CDR1 and CDR3) which enables formation of a new kind of loop that helps in recognizing an increased variety of epitopes ^{29, 48, 50, 51}. In contrast, the binding interfaces of intact antibodies are more flat and less flexible ^{29, 52}, thereby limiting their interactions only on the surface of the antigens. A longer CDR3 in nanobodies also helps them to bind to unique epitopes.

However, the small size of nanobodies can sometimes be disadvantageous. The binding properties of nanobodies might be altered when conjugated with imaging labels such as fluorophores, although there are certain strategies available to address this issue ⁵³. Also, high uptake of radiolabeled nanobodies in kidneys and liver has been observed in certain cases which not only makes detection of lesions closer to these organs difficult but also results in unwanted dose to these organs ^{54, 55}. Rapid clearance of nanobodies may prevent optimum binding to all the epitopes at the disease site, resulting in reduced uptake compared to conventional antibodies. Such rapid clearance of nanobodies from the bloodstream through renal excretion may also hamper their potential for immunotherapy ²⁹. Many therapeutic applications require a low drug clearance rate to avoid high doses and frequent administration of the therapeutic moiety. The half-lives of nanobodies could be prolonged by fusion of a target specific nanobody unit to a unit binding to albumin ^{56, 57}. Even though significant increase in half-life could be achieved by this approach, the efficacy of nanobodies in immunotherapy is much less than that of conventional antibodies due to absence of effector domains in nanobodies ²⁹. Therefore, this strategy does not hold any significant advantage over conventional antibody-based immunotherapy. Despite these limitations, the nanobody technology seems very promising for the development of next generation molecular imaging agents, if not for therapeutic applications.

Generation of nanobodies

Heavy chain antibodies which are devoid of light chains can easily be produced in animals of the Camelidae family ^{34, 36, 40}, among which camels are the most appropriate choice since they exhibit better ratio of heavy chain antibodies to regular antibodies. In reality, llamas are also often used for this purpose as they are easier to raise and breed due to smaller size ³⁴. Generally, an animal is immunized with appropriate quantity of antigen once a week for several weeks to raise an H2-type antibody response. The V_HH gene fragments from lymphocytes are amplified by standard molecular biology techniques such as reverse transcription followed by polymerase chain reaction (RT-PCR), ligated in phage display vectors and transformed in bacteria to construct an immune V_HH library ^{26, 29, 34, 37, 58, 59}. Sometimes it is essential to generate V_HHs that are cross-reactive for both human and rodent analogs of the target antigen in order to facilitate pre-clinical testing prior to clinical translation ^{34, 40}. When a library of appropriate size is obtained, antigen-binding V_HH fragments are selected from the library. Selection of nanobodies involves enrichment of antigen-specific binders from immune or naive libraries by phage display, bacterial two-hybrid or bacterial surface display, and subsequent screening for individual colonies from enriched phage populations by enzyme-linked immune sorbent assay (ELISA) ³⁷. The detailed procedures for these processes have been well-documented in the literature and will not be discussed here ^{34, 37, 40}. In the following sections, we will discuss the innovative applications of nanobodies as potential next-generation molecular imaging agents.

Molecular imaging with nanobodies

Nanobodies fulfill most of the requisites of an ideal probe for successful molecular imaging. Owing to their small molecular size, they can easily reach their targets within a few hours post-injection and exhibit great potential in tumor detection, confirmation of target expression, and selection of patients who have the highest chance to benefit from targeted therapies. Several imaging techniques such as SPECT, PET, optical imaging and ultrasound have been successfully employed for molecular imaging using nanobodies (Table 1), with the majority using radionuclide-based techniques (i.e. SPECT and PET) since these techniques are sensitive, quantitative, and clinically relevant ^{60, 61}. SPECT and PET have the sensitivity needed to visualize most interactions between physiological targets and ligands, which can enable non-invasive detection down to the picomolar level. However, we would like to emphasize that the choice of a particular modality primarily depends on the specific query to be addressed through molecular imaging.

Table 1.

Overview of nanobodies based probes in pre-clinical molecular imaging

Target	Nanobody	Imaging modality (label)	Maximum %ID/g in the tumor	Disease model	References
EGFR	8B6	SPECT (99mTc)	~5	Human epidermoid carcinoma (A431), human prostate carcinoma (DU145)	42
	7C12	SPECT (99mTc)	~4.5	Human epidermoid carcinoma (A431)	43, 81, 82
	7D12	SPECT (99mTc)	~4.5	Human epidermoid carcinoma (A431)	44, 45, 81
		PET (68Ga)	~7
		Optical (IRDye800CW)	~17
HER-2	2Rs15d	SPECT (99mTc)	~4	Human colon carcinoma (LS174T), human breast cancer (SKBR3), human ovarian cancer (SKOV3)	59
	2Rs15d	PET (68Ga)	~4	Human ovarian cancer (SKOV3)	41
	11A4	Optical (IRDye800CW)	~2	Human breast cancer (SKBR3)	53
HGF	1E2 and 6E10	PET (89Zr)	~9	Human glioblastoma (U87 MG)	56
MMR	α-MMR	SPECT (99mTc)	~3	Mammary adenocarcinoma (TS/A), Lewis lung carcinoma (3LL-R)	54
		SPECT (99mTc)	~5	Rheumatoid arthritis	95
VCAM-1	cAbVCAM1-5	SPECT (99mTc)	~3	Atherosclerosis (ApoE-deficient mice)	101
		Ultrasound (microbubble)	NA	Murine adenocarcinoma (MC38)	77
CEA	CEA1	SPECT (99mTc)	~3	Human colon adenocarcinoma (LS174)	55

NA: Not applicable

For SPECT imaging, the nanobody needs to be labeled with a suitable γ-emitting radionuclide (e.g. ^99mTc or ¹¹¹In), having an ideal γ-energy of 100-250 keV ⁶⁰. These γ-rays are recorded by the detectors of a dedicated γ-camera or SPECT instrument, which can be converted into an image upon signal processing to pinpoint the localization of the radiolabeled nanobody. On the other hand, PET requires radiolabeling of the nanobody with a suitable positron-emitting radionuclide (e.g. ¹⁸F, ⁶⁴Cu, ⁶⁸Ga or ⁸⁹Zr) ⁶¹. Compared with SPECT, PET has greater advantages with respect to sensitivity and resolution hence has been gaining significantly more clinical popularity over the last decade ^61-63. However, most of the literature reports on radionuclide-based imaging with nanobodies have used ^99mTc for SPECT imaging (Table 1). The widespread interest in the use of ^99mTc is primarily due to its excellent nuclear decay characteristics, viable coordination chemistry for radiolabeling a wide variety of biomolecules and convenient availability from cost-effective ⁹⁹Mo/^99mTc generators ⁶⁴. Also, due to the presence of the hexahistidine tag on the nanobody, it can easily be radiolabeled with ^99mTc(CO)₃ without any chemical modification of the protein ⁴². However, the hexahistidine tag on the nanobody poses some concerns as it might reduce immune responses and thus, the necessity for its removal has also been suggested ⁴¹. The high operational cost of on-site cyclotrons needed for production of PET isotopes might be a deterrent for their use, which could be overcome with the use of ⁶⁸Ga derived from ⁶⁸Ge/⁶⁸Ga generators ^{41, 65}.

Optical imaging is an emerging technology that has also been employed for cancer imaging using nanobodies, after a suitable fluorophore is conjugated ⁶⁶. The recent development in near-infrared (NIR, 700-900 nm) fluorophores have made this imaging modality highly flexible, sensitive, fast and cost-effective ^{66, 67}. The major advantage of optical imaging is that it is simple to set up and does not require dedicated facilities. On the other hand, due to scattering and absorption of light in tissues, optical imaging phenomenon has a relatively small depth of penetration ranging from hundreds of microns to several centimeters, depending on the approach ^{68, 69}. Thus, while optical imaging is well suited to study surface lesions in pre-clinical studies, it cannot currently be used for whole-body imaging. The use of biomedical optoacoustics as a powerful tool for high-resolution visualization of optical contrast has recently been proposed in order to overcome the limitations imposed by light scattering in deep tissues ⁷⁰. However, this approach is still in its infancy and extensive research efforts would be required to determine its suitability for this purpose. In clinical context, application of optical imaging for image-guided surgery has a bright future and allows imaging of several kinds of tumors during surgery which are otherwise difficult to locate ^{71, 72}.

Ultrasound is a safe, fast, and inexpensive technique widely used in the clinic for medical diagnosis ⁷³. This imaging technique makes use of sound waves (typically with frequencies in the range of 1 to 40 MHz) which are reflected differently by different organs and tissues. When compared to other imaging modalities described above, the image quality in ultrasound is often inferior as blood is a poor scatterer of ultrasound waves at clinical diagnostic transmission frequencies ⁷³. Microbubbles (µB) are generally used to enhance contrast in ultrasound imaging due to their superior scattering properties and their dynamic response to the application of an ultrasonic field ⁷⁴. Sometimes, intraoperative ultrasound imaging is used during surgical procedures ^{73, 75, 76}. Recently, nanobody-based ultrasound imaging has been reported ⁷⁷. However, this approach is still in its infancy and extensive studies will be needed before it can be translated to clinical settings.

As an intriguing platform, nanobodies have been proposed as potential molecular imaging agents not only in oncology but also for monitoring and quantifying arthritis, atherosclerosis and other inflammatory diseases ^{29, 30, 37}. Various regulatory proteins and/or receptors are overexpressed in certain diseases, which represent the molecular basis for imaging using nanobody-based probes, most of which are summarized in the following text.

Imaging of the epidermal growth factor receptor

Epidermal growth factor receptor (EGFR; also called HER-1) is a member of the HER-kinase family, together with HER-2, HER-3, and HER-4 ^{78, 79}. EGFR is overexpressed in a variety of human tumors including non-small cell lung cancer, breast, head and neck, gastric, colorectal, esophageal, prostate, bladder, renal, pancreatic, and ovarian cancers ⁷⁸. Activation of EGFR causes signaling that may lead to cell division, increased motility, angiogenesis and decreased apoptosis ⁸⁰. These effects are mediated by a complex series of signaling mechanisms, such as engagement of the mitogen-activated protein kinase (MAPK) and phosphatidylinositol-3 kinase (PI3K) pathways ^{78, 80}. Conventionally, immunohistochemical staining of tumor biopsies is used for assessing EGFR overexpression. However, it is more advantageous to measure EGFR expression using molecular imaging approach since it is non-invasive and can assess heterogeneity of both the primary tumor and the metastases. In addition, non-invasive imaging of EGFR expression can be repeated multiple times during the course of the disease and/or therapeutic intervention to provide biological insights and improve patient management.

SPECT imaging of EGFR expression using an anti-EGFR nanobody as the targeting agent was first reported by Huang et al. ⁴². The nanobody (8B6) was radiolabeled through its C-terminal hexahistidine-tag with ^99mTc-tricarbonyl intermediate [^99mTc(H₂O)₃(CO)₃]. The hexahistidine tag on the C-terminal of the nanobodyenables easy radiolabeleing with ^99mTc(H₂O)₃(CO)₃ without any additional chemical modification. Therefore, this facile approach based on ^99mTc(I)-tricarbonyl chemistry is the preferred method for radiolabeling nanobodies with ^99mTc. The radiolabeled nanobody demonstrated high specificity and selectivity towards EGFR-expressing cells. Tracer clearance was mainly via renal excretion and it was relatively fast (with a half-life of 1.5 h) ⁴². SPECT imaging studies showed that it was possible to differentiate between tumors with high and moderate EGFR expression. In another study, the in vivo tumor uptake and biodistribution of two ^99mTc-labeled anti-EGFR nanobodies (7C12 and 7D12) was compared using pinhole SPECT/CT (Figure 2) ⁸¹. It was found that binding of both the tracers was EGFR specific. In addition, high tumor uptake with low liver uptake, and rapid blood clearance were observed in both cases. However, in addition to rapid renal clearance, retention of the radiolabeled nanobodies in the kidneys was also observed which might be a cause of concern as it may lead to unwanted renal toxicity. In order to understand this phenomenon of renal toxicity, the role of megalin (an endocytic receptor which is highly expressed in renal proximal tubule) on the renal uptake of anti-EGFR ^99mTc-7C12 nanobody was evaluated in a related study ⁴³. It was concluded that megalin contributed to the renal accumulation of ^99mTc-7C12. This was confirmed by the ability of gelofusine and/or lysine, known inhibitors of megalin, to reduce renal uptake of ^99mTc-7C12. Co-injection of gelofusine and/or lysine with the tracer lowered the renal uptake and increased the tumor uptake. This procedure serves not only to allay the concerns regarding renal toxicity associated with nanobodies, but also sets an example for optimization of nuclear imaging protocols using nanobodies for further clinical studies. In another report, it was further demonstrated that tumor uptake of ^99mTc-7C12 was correlated with tumor burden ⁸². This was confirmed by reduction of ^99mTc-7C12 uptake upon Erlotinib (EGFR inhibitor) therapy. The authors suggested that quantification of ^99mTc-7C12 uptake would be helpful in monitoring the therapy response of EGFR expressing tumors.

Figure 2.

Nanobody-based imaging of EGFR expression. Transverse, coronal, and sagittal views of SPECT/CT images of mice bearing A431 tumor injected with ^99mTc-7C12 (A) or ^99mTc-7D12 (B). Images were acquired at 1 h after injection. Adapted from ⁸¹.

Vosjan et al. reported PET imaging of EGFR expression using 7D12 nanobody ⁴⁴. The nanobody was conjugated with p-isothiocyanatobenzyl-desferrioxamine (Df-Bz-NCS) and radiolabeled with ⁶⁸Ga. The short half-life of ⁶⁸Ga (t_½ = 68 min) is apt for radiolabeling nanobodies which demonstrate a fast in vivo pharmacokinetics. The biodistribution studies revealed high tumor uptake in EGFR-positive tumors and high tumor-to-blood ratio within 1 h post-injection. Tumors could be clearly visualized by PET imaging, and the results were comparable with that of SPECT using the same nanobody ⁸¹. In another study by Oliveira et al, optical imaging of EGFR using 7D12 nanobody conjugated with the NIR dye IRDye800CW was reported ⁴⁵. Again, rapid visualization of the tumor (within 30 min post-injection) was observed. Also, the tumor uptake observed here (~17 %ID/g) was much higher than what was observed with ^99mTc-7D12 (~4.5 %ID/g) or ⁶⁸Ga-7D12 (~7 %ID/g) (Table 1). The significant difference in tumor uptake might be attributed to the differences in the conjugation efficiency and the nature of the fluorophore. Owing to its rapid accumulation into the tumor and the high tumor to background ratios that could be achieved, this optical probe holds promise for use in image-guided surgery.

Imaging of the human epidermal growth factor receptor 2

The human epidermal growth factor receptor type 2 (HER-2 or CD340) is another member of the HER-kinase family that is overexpressed in various tumor types such as breast, ovarian, prostatic, and colorectal cancer ^{79, 83, 84}. HER-2 is a transmembrane glycoprotein consisting of three domains: extracellular (N-terminal), transmembrane (single alpha-helix), and intracellular (tyrosine kinase).

The generation of a large number of anti-HER-2 nanobodies for non-invasive imaging of HER-2 expression was reported by Vaneycken et al. ⁵⁹. The authors could identify 38 different nanobodies that could target HER-2 expression, out of which the nanobody 2Rs15dHis6 was selected as lead compound for further development as a clinical molecular imaging agent. The nanobody was radiolabeled with ^99mTc adopting the procedure described earlier in the review. SPECT imaging and biodistribution studies showed high uptake of ^99mTc-labeled 2Rs15dHis6 in HER-2 positive tumors, fast blood clearance, low accumulation in non-target organs (except kidneys which are responsible for tracer clearance), and high concomitant tumor-to-blood and tumor-to-muscle ratios at 1 h post-injection (Figure 3A). Recently, PET imaging of HER-2 expression using 2Rs15d nanobody was reported by the same group ⁴¹. The hexahistidine tag from the lead compound reported earlier ⁵⁹ was removed in order to reduce the risk of immunogenicity and the non-tagged format was produced. The nanobody could be radiolabeled with ⁶⁸Ga after conjugation with the bifunctional chelator, p-isothiocyanatobenzyl-1,4,7-triazacyclononane-1,4,7-triacetic acid (p-SCN-Bz-NOTA). Similar as that observed in the previous study ⁵⁹, biodistribution and PET/CT studies revealed high tumor uptake, high tumor-to-blood and tumor-to-muscle ratios as early as 1 h post-injection, resulting in high-contrast PET/CT images (Figure 3B). On the basis of mouse toxicity and dosimetry studies, the radiotracer was found to be safe and the authors claimed that ⁶⁸Ga-NOTA-2Rs15d was ready for first-in-human clinical trials.

Figure 3.

Nanobody-based imaging of HER-2 expression. (A) Transverse and coronal views of SPECT/CT images of HER-2 positive SKOV3 tumor-bearing mice at 1 h post-injection of ^99mTc-2Rs15d. (B) PET/CT images of rats bearing SKOV3 (left) or HER-2 negative MDA-MB-435D (right) tumor xenografts at 1 h post-injection of ⁶⁸Ga-2Rs15d. (C) In vivo optical imaging of SKBR3 tumor-bearing mice at 4 h post-injection of HER-2 specific (11A4, 18C3 or 22G12) or negative control (R2) nanobodies conjugated to IRDye800CW (abbreviated as IR). Tumors are indicated with red arrow and kidneys with green arrow. Adapted from ⁴¹, ⁵³, ⁵⁹.

Pruszynski et al. reported the radiolabeling of another anti-HER-2 nanobody (5F7GGC) with ^125/131I using the residualizing agent, N^ɛ-(3-[¹³¹I]iodobenzoyl)-Lys⁵-N^α-maleimido-Gly¹-GEEEK ⁸⁵. The residualizing agent contains multiple negatively charged D-amino acids and is effective in enhancing the retention of radioactivity in tumor cells after receptor mediated internalization of the radiolabeled nanobody. The potential of this radiotracer for targeting breast cancer was demonstrated, which could be extended for future SPECT and PET imaging of HER-2 expression in cancer patients using ¹²³I and ¹²⁴I, respectively. In another study, rapid optical imaging of human breast cancer xenografts with high tumor-to-background ratios (Figure 3C) was reported using anti-HER-2 nanobody (11A4) conjugated to IRDye 800CW ⁵³. Under the guidance of the real-time fluorescent images obtained from the camera using 11A4-IR probe, surgical resection of a HER-2-positive xenograft from a mouse could be demonstrated. Both tumor and kidney were clearly visible through the skin during the operation. This was the first study where image-guided surgery was successfully carried out using nanobody-based optical imaging probe. Further development of this method would help in precise and rapid identification of HER-2 positive tumors, allow assessment of response to HER-2 therapies and might assist in image-guided surgeries if required. The first clinical trial with a nanobody-based molecular imaging tracer targeting HER-2 is currently ongoing (EudraCT 2012-001135-31) and the results are yet to be reported ³³.

Imaging of the hepatocyte growth factor

Hepatocyte growth factor (HGF), also known as scatter factor, is a paracrine cellular growth, motility and morphogenic factor ⁸⁶. Secreted by mesenchymal cells as a single-chain inactive polypeptide, HGF acts as a multifunctional cytokine on cells of mainly epithelial and endothelial origin, as well as on hematopoietic progenitor cells. HGF is the only known ligand for c-Met, a membrane receptor that is essential for embryonic development and is expressed on the epithelial cells of many organs such as liver, prostate, pancreas, muscle, kidney, and bone marrow ⁸⁷. In tumor cells, c-Met activation triggers diverse series of signaling cascades resulting in cell growth, proliferation, invasion, metastasis, and escape from apoptosis ^86-88. Elevated levels of HGF and c-Met have been observed in most solid tumors and are associated with increased aggressiveness of tumors and poor prognostic outcome of cancer patients ^86-88.

The potential of nanobodies for PET imaging and therapy of HGF-expressing tumors was reported by Vosjan et al. ⁵⁶. Two nanobodies (1E2 and 6E10) were each molecularly fused to albumin-binding units (Alb8) to obtain serum half-life extension, conjugated with Df-Bz-NCS chelator and radiolabeled with ⁸⁹Zr, with the goals of preventing the binding of HGF to c-Met receptor and assess the use of these novel tracers in cancer diagnosis and therapy ^{56, 89}. Although biodistribution studies in nude mice showed high uptake in xenograft U87 MG glioblastoma tumors (> 7 %ID/g), blood clearance of the tracers was very slow due to fusion of nanobodies with albumin units and no PET imaging data was reported.

In this study, the therapeutic potential of these nanobodies was also evaluated in the same animal model ⁵⁶. All the mice treated with the nanobodies exhibited delayed tumor growth when compared with the control group that was injected with saline, and no toxicity to normal organs was observed. Taken together, although the therapeutic efficacy of albumin fused anti-HGF nanobodies was demonstrated, the nanobodies without fusion to albumin seem to be the better choice for molecular imaging applications because of their rapid blood clearance and fast tumor accumulation.

Imaging of the macrophage mannose receptor

The macrophage mannose receptor (MMR) is a transmembrane glycoprotein that is expressed on macrophages ⁹⁰, which is primarily responsible for endocytic clearance of certain glycoproteins and phagocytosis of unopsonized microorganisms. It has been reported that certain components of the tumor stroma, especially in the hypoxic areas, that are highly pro-angiogenic overexpress MMR, which plays an important role during tumor growth and related biological processes such as angiogenesis, metastasis, and immune suppression ⁹¹. Therefore, MMR is an attractive marker for non-invasive molecular imaging of tumor-associated macrophages.

Anti-MMR nanobodies were developed and successfully used for SPECT/CT imaging of tumor-infiltrating macrophages ⁵⁴. The nanobodies were radiolabeled with ^99mTc and injected in mammary adenocarcinoma and Lewis lung carcinoma (TS/A and 3LL-R) tumor-bearing mice. Biodistribution studies and pinhole SPECT/CT imaging showed rapid targeting of MMR expressing cells in tissues, with high tumor-to-background ratio (Figure 4A). Tumor retention of the radiolabeled nanobody was target-specific and absent in MMR knockout mice. However, in both the TS/A and 3LL-R models, ^99mTc-labeled anti-MMR nanobody accumulated to a higher extent in the liver and spleen (due to the presence of MMR expression in these organs) than the tumor, which could be minimized by co-injection of excess unlabeled, bivalent anti-MMR nanobody without compromising tumor uptake ⁵⁴ . The authors suggested that this strategy of co-injecting bivalent cold anti-MMR nanobody to reduce extratumoral tracer uptake would help in clinical translation of this radiotracer. In such a strategy, bivalent cold nanobodies would be efficient at binding non tumor sites and thus reducing non specific signals. The large size of the bivalent nanobody prevents its penetration into the tumor leaving the tumor targeting sites free for the labeled monovalent nanobodies.

Figure 4.

Nanobody-based imaging of MMR expression. (A) Coronal and transverse views of SPECT/CT images of 3LL tumor-bearing wide-type (WT) or MMR knockout (MMR^-/-) mice at 3 h post-injection of ^99mTc-labeled BCII10 control nanobody (Nb) or anti-MMR (α-MMR) nanobody. (B) SPECT/CT images of mice showing signs of arthritis in both hind limbs at 3 h post-injection of ^99mTc-labeled α-MMR nanobody or BCII10 control nanobody. Tracer accumulation of ^99mTc-labeled α-MMR nanobody was evident in knees, ankles, and metatarsal joints (indicated by arrows), but not the control nanobody. Adapted from ⁵⁴, ⁹⁵.

Interestingly, MMR was also found to be involved in the formation of multinucleated osteoclasts, by binding to terminal high-mannose-type oligosaccharides expressed on osteoclast precursor cells ^{92, 93}. Osteoclasts, derived from monocyte or macrophage lineage, are a type of bone cells involved in bone resorption, which degrade the bone tissue by removing its mineralized matrix and breaking up the organic bone. Therefore, they are also responsible for the progressive destruction of bone tissue in rheumatoid arthritis ⁹⁴. SPECT/CT imaging with ^99mTc-labeled anti-MMR nanobodies was demonstrated as an effective tool for monitoring joint inflammation in collagen-induced arthritis in a mouse model for rheumatoid arthritis ⁹⁵. Using flow cytometry and quantitative PCR with anti-MMR nanobodies, it was shown that MMR was highly expressed on macrophages and osteoclasts derived from the bone marrow. In vivo SPECT/CT imaging showed a highly specific uptake of the radiolabeled agent in the inflamed joints of arthritic mice (Figure 4B). With further research efforts, this approach might prove to be a useful tool for monitoring and quantifying joint inflammation in rheumatoid arthritis patients.

Imaging of the vascular cell adhesion molecule-1

Vascular cell adhesion molecule-1 (VCAM-1), also known as cluster of differentiation 106 (CD106), is a protein that is encoded by the VCAM-1 gene in humans ⁹⁶. The VCAM-1 protein mediates the adhesion of lymphocytes, monocytes, eosinophils, and basophils to vascular endothelium and also functions in leukocyte-endothelial cell signal transduction. VCAM-1 is expressed at low levels in non-atherosclerotic arteries. However, hypercholesterolemia rapidly induces VCAM-1 expression by endothelial cells in regions prone to atheroma formation ^{97, 98}. It has been reported that arterial muscle cells and even macrophages can express VCAM-1 in hypercholesterolemic animals ⁹⁹. In vitro, pro-inflammatory cytokines readily augment the expression of VCAM-1 in endothelial cells and other cell types relevant to atherosclerosis ^{99, 100}. VCAM-1 binds to very late antigen-4 (VLA4) present on the surface of leukocytes. Generally, active inflammation characterized by leukocyte infiltration is recognized as a major criterion for defining vulnerable plaques ⁹⁹. Therefore, VCAM-1 is a highly relevant molecular target for non-invasive detection of such lesions.

Preclinical imaging of atherosclerotic plaques using anti-VCAM-1 nanobodies was reported by Broisat et al. ¹⁰¹. The nanobodies were radiolabeled with ^99mTc for SPECT imaging of atherosclerotic lesions in ApoE-deficient mice ^{101, 102}. Among the various nanobodies studied, cAbVCAM1-5 was cross-reactive for human VCAM-1 and exhibited high lesion-to-control, lesion-to-heart, and lesion-to-blood ratios. Uptake of ^99mTc-cAbVCAM1-5 in atherosclerotic lesions could be readily visualized in ApoE-deficient mice, while no tracer uptake was observed in the control animals or with the non-targeting control nanobody (cAbBcII10). Co-injection of a 100-fold excess of unlabeled competitor cAbVCAM1-1 (a nanobody recognizing the same VCAM-1-epitope as cAbVCAM1-5) with ^99mTc-cAbVCAM1-5 resulted in significant decrease in ^99mTc-cAbVCAM1-5 uptake in the liver, lymphoid tissues, and atherosclerotic lesions, thereby demonstrating specificity of the radiolabeled nanobody (Figure 5A). The uptake of ^99mTc-cAbVCAM1-5 in VCAM-1-positive lesions was further confirmed by autoradiography and immunohistochemistry. Though this study successfully demonstrated the prospective of radiolabeled nanobodies as tracers in nuclear cardiology for the first time, several pertinent factors such as immunogenicity, toxicity and safety need to be evaluated before such tracers can actually be translated to clinical settings.

Figure 5.

Nanobody-based imaging of VCAM-1 expression. (A) SPECT/CT images of atherosclerotic lesions in ApoE-deficient mice injected with ^99mTc-cAbVCAM1-5 alone (no competition) or with a 100-fold excess of unlabeled cAbVCAM1-1 (competition). Competition resulted in significant decrease of tracer uptake in the liver, lymphoid tissues, and atherosclerotic lesions, thereby demonstrating VCAM-1 specificity. (B) Transverse B-mode ultrasound images of MC38 tumors overlaid with contrast-enhanced signal at 10 min post-injection of either μB-cAbVCAM1-5 or a non-targeting control μB-cAbGFP4. Adapted from ⁷⁷, ¹⁰¹.

Recently, the same group investigated the suitability of anti-VCAM-1 nanobodies for ultrasound imaging of VCAM-1 expression in MC38 murine tumor model ⁷⁷. The nanobodies were conjugated to the surface of microbubbles through biotin-streptavidin interaction. Contrast-enhanced ultrasound imaging using μB-cAbVCAM1-5 showed enhanced echo signal intensity in the tumor compared to control microbubbles, within 10 min after injection (Figure 5B). After bursting of the microbubbles, the signal almost completely disappeared, indicating that the majority of unbound circulating microbubbles had cleared from the bloodstream. Overall, this preliminary study showed that nanobody-based molecular ultrasound holds the potential for rapid and cost-effective in vivo detection of VCAM-1 expression in the tumor vasculature.

Imaging of the carcinoembryonic antigen

Carcinoembryonic antigen (CEA) is a complex and highly glycosylated macromolecule which is normally produced in gastrointestinal tissue during fetal development ¹⁰³. Although usually present at very low levels in healthy adults, CEA is highly expressed in many cancer types such as colorectal carcinoma and adenocarcinomas of the lung, breast, other gastrointestinal organs and the ovaries. Therefore, CEA has been an intensively studied target for cancer diagnosis and therapy over the last several decades ¹⁰³.

The potential of nanobodies for SPECT imaging of tumor CEA expression was investigated by Cortez-Retamozo et al. ⁵⁵. The CEA specific nanobody (CEA1) was radiolabeled with ^99mTc. Biodistribution and SPECT imaging studies in mice bearing CEA-positive human colon adenocarcinoma (LS174T) tumor showed both hepatic and renal clearance of the radiolabeled nanobody, resulting in low blood radioactivity levels at 3 h post-injection with average tumor uptake of > 3 %ID/g, which showed good contrast between the CEA-positive tumor and the contralateral muscle region. However, a major limitation of this radiotracer is the high liver and kidney uptake, which might make detection of lesions closer to or inside these organs difficult. The liver and kidney uptake can possibly be reduced without reducing the tumor uptake by co-injection of an optimal dose of cold nanobody ⁵⁴ and excess amount of amino acid such as lysine ⁴³ along with ^99mTc-CEA1, as reported for other receptor imaging using nanobodies.

Conclusion and future perspectives

Molecular imaging is the foundation for accurate diagnosis and pre-treatment staging of various diseases, monitoring the response to therapy and providing surveillance after therapeutic intervention. Recent advances in molecular imaging using nanobody-based probes have enhanced the possibility of high-contrast imaging of different types of tumors within much shorter time period after tracer injection, when compared to the conventional intact antibody based approaches. Nanobodies are smaller than intact antibodies and can penetrate into solid tumor tissue more efficiently, thereby representing a promising class of targeting ligands for non-invasive molecular imaging of specific targets of interest. The maturity in nanobody technology would improve our understanding of the molecular alterations in various types, subtypes and stages of cancers and other diseases, which can help the choice of suitable targeted therapeutics for each individual patient. However, although it has been more than 2 decades since nanobodies were first described, the translation of these promising imaging agents and technologies to the clinic has been very slow. Meanwhile, the development of a new diagnostic probe is an iterative process that requires substantial effort and sometimes luck. The reasons for slow clinical translation is complex and may include considerable regulatory hurdles, limited potential market, lobbying by the manufacturers of other molecular imaging probes, lack of reimbursement strategies for novel imaging agents, etc. Despite these hurdles, the exciting results obtained to date with nanobodies for molecular imaging clearly indicated that these tracers will have multifaceted applications in future clinical practices.

In this review we have described the recent advances in the field of molecular imaging using nanobody-based probes, which provided compelling evidence that these new molecules might be considered as “magic bullets” of molecular imaging in the foreseeable future. Although we have restricted our discussion to molecular imaging applications only, nanobodies may also be the molecules of choice in biotechnology, development of biological sensors and other clinical applications. Primarily, nanobodies have been proposed as efficient probes for oncologic imaging. With further improvement in engineering of various nanobody formats and refinement in conjugation chemistry, they may also be useful for monitoring various other diseases such as amyloidoses, viral infections, cardiac problems, etc. The image-guided surgery approach using nanobody-based optical imaging probes, if successfully translated to clinical settings would assist the surgeons in performing more radical tumor resections with improved cosmesis, thereby improving the management and welfare of cancer patients.

Utilization of nanobodies for targeted therapy has been proposed. However, we believe that the use of nanobodies themselves as drugs may not be the best option, due to their rapid clearance from blood, and opine that the conventional antibodies will continue to play a dominant role in immunotherapy. However, nanobodies could play an important role as diagnostic tools in tumor detection, evaluation of target expression before initiating a targeted therapy, prediction or monitoring response to a certain targeted therapy, as well as detection of possible reoccurrence of the disease. Enabling same day imaging of various lesions using nanobody-based probes would accelerate evaluation of patient care and improve patient management. Additionally, various types of nanocarriers, such as iron oxide nanoparticles, silica nanoparticles, copper sulfide nanoparticles, gold nanostructures, carbon nanomaterials, and upconversion nanophosphors can be conjugated with nanobodies for non-invasive visualization of tumors ^104-113. This approach can not only find application in molecular imaging of various types of receptor expressions but also holds promise for simultaneous tumor targeted delivery of anticancer drugs. The high surface area to volume ratio of the nanocarriers enables rich surface chemistry for attachment of nanobodies on their surface while retaining high loading capacity for detection elements and therapeutic payload ¹¹⁴. The small size of nanobodies can enable the attachment of a large number of nanobodies per nanocarrier, which may lead to significantly enhanced tumor targeting efficiency and specificity. However, nanoparticle usage might be limited by their pharmacokinetics, high liver and spleen uptake that might make visualization of deep seated tumors in proximity to these organs difficult ¹¹⁵. Also, there might be issues related to toxicity of nanoparticles in certain cases ^{116, 117}.

Most of the applications reported to date using nanobodies have employed SPECT or PET as the imaging modality, though efforts towards the use of other techniques such as optical and ultrasound imaging are also emerging. No single modality is perfect and sufficient to gain all the necessary information and therefore combination of multiple molecular imaging techniques can offer synergistic advantages over any single modality. It should be possible to develop nanobody-based multimodal contrast agents or imaging probes to address this issue in molecular imaging. Such probes would provide much more information than single modality nanobody-based probes that exist in the literature ¹³. For example, the synergistic combination of PET and magnetic resonance imaging (MRI) would merge the advantage of high soft tissue contrast of MRI with the functional information of PET, thereby leading to improved diagnostic accuracy and radiotherapy planning. The dual modality molecular imaging agents using optical and MR based probes might be helpful in surgical procedures where optical imaging can be used to guide the scalpel and MR to ensure that all cancerous material has been completely removed. However, direct attachment of multimodal probes with nanobodies might be challenging due to the limited number of conjugation sites available and the potential interference with their receptor binding ability. The use of nanocarriers would be suitable for this purpose as they have large surface areas where multiple functional moieties can be incorporated for multi-modality molecular imaging ¹⁰⁶. With further impetus on development of such advanced probes, nanobody-based multimodality imaging approaches hold the potential to carve a niche for themselves in the clinic.

Overall, the recent preclinical success of nanobodies clearly encourages further research and development efforts incorporating broader application of molecular imaging, eventually in clinical settings. Interdisciplinary and concerted efforts of synthetic chemists, molecular biologists, biomedical and imaging scientists, clinicians, and the regulatory agencies are needed to achieve the ultimate goal of personalized medicine.