Designing proteins with tunable activities from easily accessible external cues remains a biotechnological challenge. Here, we set out to create a small antibody-binding domain equipped with a molecular switch inspired by the allosteric response to calcium seen in naturally derived proteins like calmodulin. We have focused on one of the three domains of Protein G that show inherent affinity to antibodies. By combining a semi-rational protein design with directed evolution, we engineered novel variants containing a calcium-binding loop rendering the inherent antibody affinity calcium-dependent. The evolved variants resulted from a designed selection strategy subjecting them to negative and positive selection pressures focused on conditional antibody binding. Hence, these variants contains molecular “on/off” switches, controlling the target affinity towards antibody fragments simply by the presence or absence of calcium. From NMR spectroscopy we found that the molecular mechanism underlying the evolved switching behavior was a coupled calcium-binding and folding event where the target binding surface was intact and functional only in the presence of bound calcium. Notably, it was observed that the response to the employed selection pressures gave rise to the evolution of a cooperative folding mechanism. This observation illustrates why the cooperative folding reaction is an effective solution seen repeatedly in the natural evolution of fine-tuned macromolecular recognition. Engineering binding moieties to confer conditional target interaction has great potential due to the exquisite interaction control that is tunable to application requirements. Improved understanding of the molecular mechanisms behind regulated interactions is crucial to unlock how to engineer switchable proteins useful in a variety of biotechnological applications.

Conjugation of various reagents to antibodies has long been an elegant way to combine the superior binding features of the antibody with other desired but non-natural functions. Applications range from labels for detection in different analytical assays to the creation of new drugs by conjugation to molecules which improves the pharmaceutical effect. In many of these applications, it has been proven advantageous to control both the site and the stoichiometry of the conjugation to achieve a homogeneous product with predictable, and often also improved, characteristics. For this purpose, many research groups have, during the latest decade, reported novel methods and techniques, based on small molecules, peptides, and proteins with inherent affinity for the antibody, for site-specific conjugation of antibodies. This review provides a comprehensive overview of these methods and their applications and also describes a historical perspective of the field.

Radionuclide molecular imaging of human epidermal growth factor receptor type 2 (HER2) expression may help to stratify breast and gastroesophageal cancer patients for HER2-targeting therapies. Albumin-binding domain-derived affinity proteins (ADAPTs) are a new type of small (46-59 amino acids) protein useful as probes for molecular imaging. The aim of this first-in-humans study was to evaluate the biodistribution, dosimetry, and safety of the HER2-specific Tc-99(m)-ADAPT6. Methods: Twenty-nine patients with primary breast cancer were included. In 22 patients with HER2-positive (n = 11) or HER2-negative (n = 11) histopathology, an intravenous injection of 385 +/- 125 MBq of Tc-99(m)-ADAPT6 was performed, randomized to an injected protein mass of either 500 mu g (n = 11) or 1,000 mu g (n = 11). Planar scintigraphy followed by SPECT imaging was performed after 2, 4, 6, and 24 h. An additional cohort (n = 7) was injected with 165 +/- 29 MBq (injected protein mass, 250 mu g), and imaging was performed after 2 h only. Results: Injections of Tc-99(m)-ADAPT6 were well tolerated at all mass levels and not associated with adverse effects. Tc-99(m)-ADAPT6 cleared rapidly from the blood and most other tissues. The normal organs with the highest accumulation were the kidney, liver, and lung. Effective doses were 0.009 +/- 0.002 and 0.010 +/- 0.003 mSv/MBq for injected protein masses of 500 and 1,000 mu g, respectively. Injection of 500 mu g resulted in excellent discrimination between HER2-positive and HER2-negative tumors as early as 2 h after injection (tumor-to-contralateral breast ratio, 37 +/- 19 vs. 5 +/- 2; P < 0.01). The tumor-to-contralateral breast ratios for HER2-positive tumors were significantly (P < 0.05) higher for an injected mass of 500 mu g than for either 250 or 1,000 mu g. Conclusion: Injections of Tc-99(m)-ADAPT6 are safe and associated with low absorbed and effective doses. A protein dose of 500 mu g is preferable for discrimination between tumors with high and low expression of HER2. Further studies are justified to evaluate whether Tc-99(m)-ADAPT6 can be used as an imaging probe to stratify patients for HER2-targeting therapy in areas where PET imaging is not readily available.

Molecular recognition in targeted therapeutics is typically based on immunoglobulins. Development of engi-neered scaffold proteins (ESPs) has provided additional opportunities for the development of targeted therapies. ESPs offer inexpensive production in prokaryotic hosts, high stability and convenient approaches to modify their biodistribution. In this study, we demonstrated successful modification of the biodistribution of an ESP known as ADAPT (Albumin-binding domain Derived Affinity ProTein). ADAPTs are selected from a library based on the scaffold of ABD (Albumin Binding Domain) of protein G. A particular ADAPT, the ADAPT6, binds to human epidermal growth factor receptor type 2 (HER2) with high affinity. Preclinical and early clinical studies have demonstrated that radiolabeled ADAPT6 can image HER2-expression in tumors with high contrast. However, its rapid glomerular filtration and high renal reabsorption have prevented its use in radionuclide therapy. To modify the biodistribution, ADAPT6 was genetically fused to an ABD. The non-covalent binding to the host's albumin resulted in a 14-fold reduction of renal uptake and appreciable increase of tumor uptake for the best variant, Lu-177-DOTA-ADAPT6-ABD035. Experimental therapy in mice bearing HER2-expressing xenografts demonstrated more than two-fold increase of median survival even after a single injection of 18 MBq Lu-177-DOTA-ADAPT6ABD035. Thus, a fusion with ABD and optimization of the molecular design provides ADAPT derivatives with attractive targeting properties for radionuclide therapy.

Albumin-binding fusion partners are frequently used as a means for the in vivo half-life extension of small therapeutic molecules that would normally be cleared very rapidly from circulation. However, in applications where small size is key, fusion to an additional molecule can be disadvantageous. Albumin-derived affinity proteins (ADAPTs) are a new type of scaffold proteins based on one of the albumin-binding domains of streptococcal protein G, with engineered binding specificities against numerous targets. Here, we engineered this scaffold further and showed that this domain, as small as 6 kDa, can harbor two distinct binding surfaces and utilize them to interact with two targets simultaneously. These novel ADAPTs were developed to possess affinity toward both serum albumin as well as another clinically relevant target, thus circumventing the need for an albumin-binding fusion partner. To accomplish this, we designed a phage display library and used it to successfully select for single-domain bispecific binders toward a panel of targets: TNFα, prostate-specific antigen (PSA), C-reactive protein (CRP), renin, angiogenin, myeloid-derived growth factor (MYDGF), and insulin. Apart from successfully identifying bispecific binders for all targets, we also demonstrated the formation of the ternary complex consisting of the ADAPT together with albumin and each of the five targets, TNFα, PSA, angiogenin, MYDGF, and insulin. This simultaneous binding of albumin and other targets presents an opportunity to combine the advantages of small molecules with those of larger ones allowing for lower cost of goods and noninvasive administration routes while still maintaining a sufficient in vivo half-life. 

Albumin binding domain derived affinity proteins (ADAPTs) are a class of small and folded engineered scaffold proteins that holds great promise for targeting cancer tumors. Here, we have extended the in vivo half-life of an ADAPT, targeting the human epidermal growth factor receptor 2 (HER2) by fusion with an albumin binding domain (ABD), and armed it with the highly cytotoxic payload mertansine (DM1) for an investigation of its properties in vitro and in vivo. The resulting drug conjugate, ADAPT6-ABD-mcDM1, retained binding to its intended targets, namely HER2 and serum albumins. Further, it was able to specifically bind to cells with high HER2 expression, get internalized, and showed potent toxicity, with IC50 values ranging from 5 to 80 nM. Conversely, no toxic effect was found for cells with low HER2 expression. In vivo, ADAPT6-ABD-mcDM1, radiolabeled with Tc-99m, was characterized by low uptake in most normal organs, and the main excretion route was shown to be through the kidneys. The tumor uptake was 5.5% ID/g after 24 h, which was higher than the uptake in all normal organs at this time point except for the kidneys. The uptake in the tumors was blockable by pre-injection of an excess of the monoclonal antibody trastuzumab (having an overlapping epitope on the HER2 receptor). In conclusion, half-life extended drug conjugates based on the ADAPT platform of affinity proteins holds promise for further development towards targeted cancer therapy.

Molecular recognition, or the specific interactions between a protein and its ligand, is central to biology and a key factor for many different clinical and technical applications. Despite antibodies being only one of many different affinity proteins, it has by far been the most successful. However, their large size and complex structure can be limiting in terms of cost and stability. Furthermore, their effector functions can sometimes be undesired or even detrimental. Over the past decades, many alternative affinity proteins have emerged to overcome some of these limitations.

The Albumin Binding Domain (ABD), originally present on the surface of certain bacterial cells, has previously been subjected to combinatorial protein engineering for the generation of ADAPTs (ABD Derived Affinity ProTeins) that bind to different targets. One of these, the ADAPT6, targets HER2 and has shown great promise as a tracer for radionuclide molecular imaging for diagnosis and stratification of HER2 positive patients. The work in this thesis has aimed to optimise the ADAPT6 tracer further and also describes the first-inhuman clinical trial for imaging of HER2-overexpressing breast cancer. The results establish that ADAPT6 is safe and well-tolerated by patients and able to detect primary tumours as well as metastases with very high contrast already 2 hours after injection. However, the high kidney uptake associated with its fast blood clearance prevents further use of ADAPT6 also in a therapeutic setting. By engineering the ADAPT6 to prolong its circulatory half-life and reduce the kidney uptake, this thesis has also aimed to explore the therapeutic potential of this molecule. As a first step towards this goal, the ADAPT6 was genetically fused to an ABD to allow for binding to a patient’s own serum albumin and hence avoid the same extent of renal filtration. Indeed, when evaluated in mice, fusion to ABD increased the retention in circulation by more than 200-fold and exhibited a dramatically decreased renal activity. Treatment of tumour-bearing mice with the ABD-fused ADAPT6 conjugated to a cytotoxic radionuclide significantly prolonged survival by more than two-fold and was not associated with any observable toxicity. Finally, this thesis also describes a novel combinatorial library from which several bispecific ADAPTs have been identified, binding to both albumin and other clinically relevant targets simultaneously. This miniature bispecific scaffold offers an opportunity to combine the benefits associated with small size such as good tissue extravasation and alternative administration routes while still maintaining a sufficient in vivo half-life.

Molekylär igenkänning, eller de specifika interaktionerna mellan ett protein och dess ligand, är ett centralt koncept inom biologin och en nyckelfaktor för många olika kliniska och tekniska tillämpningar. Trots att antikroppar bara är ett av många olika affinitetsproteiner har det varit det tveklöst mest framgångsrika. De har dock vissa begränsningar som beror av deras storlek samt deras komplexa struktur. Dessutom kan deras effektorfunktioner ibland vara oönskade eller till och med medföra negativa effekter. Under de senaste decennierna har därför många alternativa affinitetsproteiner utvecklats.

Proteinet ABD (Albumin Binding Domain), som ursprungligen återfinns på ytan på vissa bakterieceller, har tidigare utsatts för kombinatorisk proteinteknik för att generera ADAPT:er (ABD Derived Affinity ProTeins) som binder till olika målproteiner. En av dessa, ADAPT6, riktar sig mot HER2 och har uppvisat lovande resultat inom molekylär avbildning med hjälp av radionuklider, för diagnos och stratifiering av HER2-positiva patienter. Arbetet i denna avhandling har syftat till att optimera ADAPT6-molekylen ytterligare och beskriver också en första klinisk prövning i människa för avbildning av HER2-överuttryckande bröstcancer. Resultaten visar att ADAPT6 är säker och väl tolererad av patienter och att den kan visualisera primära tumörer såväl som metastaser med mycket hög kontrast redan 2 timmar efter injektion. Dess snabba eliminering från blodet är dock associerad med ett högt upptag i njurarna vilket förhindrar ytterligare användning av ADAPT6 som behandling. Genom att konstruera en ADAPT6-molekyl med förlängd halveringstid har denna avhandling också syftat till att utforska den terapeutiska potentialen för denna molekyl. Som ett första steg mot detta mål fuserades ADAPT6 genetiskt till en ABD-molekyl för att möjliggöra bindning till patientens eget serumalbumin och därmed undvika samma omfattning av filtrering i njurarna. Studier i möss visade att fusion till ABD ökade retentionen i blodet mer än 200 gånger och uppvisade samtidigt en dramatiskt minskad njuraktivitet. Behandling av tumörbärande möss med ABD-fuserad ADAPT6 konjugerad till en cytotoxisk radionuklid förlängde överlevnaden signifikant och var inte associerad med någon toxicitet. Slutligen beskriver denna avhandling också ett nytt kombinatoriskt bibliotek från vilket flera bispecifika ADAPT:er har identifierats, vilka binder till både albumin och andra kliniskt relevanta målprotein samtidigt. Detta minimerade bispecifika protein har potential att åtnjuta de fördelar som är associerade med en liten storlek, såsom ökad vävnadspenetration samt möjligheten till alternativa administrationsvägar, medan det samtidigt kan upprätthålla en tillräcklig halveringstid i blodet.

ADAPTs are small engineered non-immunoglobulin scaffold proteins, which have demonstrated very promising features as vectors for radionuclide tumour targeting. Radionuclide imaging of human epidermal growth factor 2 (HER2) expression in vivo might be used for stratification of patients for HER2-targeting therapies. ADAPT6, which specifically binds to HER2, has earlier been shown to have very promising features for in vivo targeting of HER2 expressing tumours. In this study we tested the hypothesis that dimerization of ADAPT6 would increase the apparent affinity to HER2 and accordingly improve tumour targeting. To find an optimal molecular design of dimers, a series of ADAPT dimers with different linkers, -SSSG- (DiADAPT6L1), -(SSSG)(2)- (DiADAPT6L2), and -(SSSG)(3)- (DiADAPT6L3) was evaluated. Dimers in combination with optimal linker lengths demonstrated increased apparent affinity to HER2. The best variants, DiADAPT6L2 and DiADAPT6L3 were site-specifically labelled with In-111 and I-125, and compared with a monomeric ADAPT6 in mice bearing HER2-expressing tumours. Despite higher affinity, both dimers had lower tumour uptake and lower tumour-to-organ ratios compared to the monomer. We conclude that improved affinity of a dimeric form of ADAPT does not compensate the disadvantage of increased size. Therefore, increase of affinity should be obtained by affinity maturation and not by dimerization.