Summary The complex architecture of transmembrane proteins requires quality control (QC) of folding, membrane positioning, and trafficking as prerequisites for cellular homeostasis and intercellular communication. However, it has remained unclear whether transmembrane protein-specific QC hubs exist. Here we identify cereblon (CRBN), the target of immunomodulatory drugs (IMiDs), as a co-chaperone that specifically determines chaperone activity of HSP90 toward transmembrane proteins by means of counteracting AHA1. This function is abrogated by IMiDs, which disrupt the interaction of CRBN with HSP90. Among the multiple transmembrane protein clients of CRBN-AHA1-HSP90 revealed by cell surface proteomics, we identify the amino acid transporter LAT1/CD98hc as a determinant of IMiD activity in multiple myeloma (MM) and present an Anticalin-based CD98hc radiopharmaceutical for MM radio-theranostics. These data establish the CRBN-AHA1-HSP90 axis in the biogenesis of transmembrane proteins, link IMiD activity to tumor metabolism, and nominate CD98hc and LAT1 as attractive diagnostic and therapeutic targets in MM.

No abstract

No abstract

The CD98 heavy chain (CD98hc) constitutes both a promising cell surface target for the treatment of cancers and a transcytosis receptor potentially useful for the brain delivery of therapeutics. However, pharmacokinetic studies and safety assessment of cognate antibodies or nonimmunoglobulin binding proteins in rodents is hampered by cross‐species variability of both amino acid sequence and glycosylation pattern. Here, we report the crystal structure of the murine CD98hc extracellular domain and a comprehensive comparison with its human ortholog, revealing only one conserved surface patch that is neither shielded by glycosylation nor by the cell membrane with an accessible surface area typical for an antibody epitope. Our results imply the necessity of a surrogate approach for CD98hc‐specific binding proteins with predictive power for clinical investigations.

Anticalin proteins directed against the prostate-specific membrane antigen (PSMA), optionally having tailored plasma half-life using PASylation technology, show promise as radioligands for PET-imaging of xenograft tumors in mice. To investigate their suitability, the short-circulating unmodified Anticalin was labeled with 68Ga (τ1/2 = 68 min), using the NODAGA chelator, whereas the half-life extended PASylated Anticalin was labeled with 89Zr (τ1/2 = 78 h), using either the linear chelator deferoxamine (Dfo) or a cyclic derivative, fusarinine C (FsC). Different PSMA targeting Anticalin versions (optionally carrying the PASylation sequence) were produced carrying a single exposed N- or C-terminal Cys residue and site-specifically conjugated with the different radiochelators via maleimide chemistry. These protein conjugates were labeled with radioisotopes having distinct physical half-lives and, subsequently, applied for PET-imaging of subcutaneous LNCaP xenograft tumors in CB17 SCID mice. Uptake of the protein tracers into tumor versus healthy tissues was assessed by segmentation of PET data as well as biodistribution analyses. PET-imaging with both the 68Ga-labeled plain Anticalin and the 89Zr-labeled PASylated Anticalin allowed clear delineation of the xenograft tumor. The radioligand A3A5.1-PAS(200)-FsC·89Zr, having an extended plasma half-life, led to a higher tumor uptake 24 h p.i. compared to the 68Ga·NODAGA-Anticalin imaged 60 min p.i. (2.5% ID/g vs 1.2% ID/g). Pronounced demetallation was observed for the 89Zr·Dfo-labeled PASylated Anticalin, which was ∼50% lower in the case of the cyclic radiochelator FsC (p < 0.0001). Adjusting the plasma half-life of Anticalin radioligands using PASylation technology is a viable approach to increase radioisotope accumulation within the tumor. Furthermore, 89Zr-ImmunoPET-imaging using the FsC radiochelator is superior to that using Dfo. Our strategy for the half-life adjustment of a tumor-targeting Anticalin to match the physical half-life of the applied radioisotope illustrates the potential of small binding proteins as an alternative to antibodies for PET-imaging.