The detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-specific antibodies in the serum of an individual indicates prior infection or vaccination. However, it provides limited insight into the protective nature of this immune response. Neutralizing antibodies recognizing the viral spike protein are more revealing, yet their measurement traditionally requires virus- and cell-based systems that are costly, time-consuming, inflexible, and potentially biohazardous. Here, we present a cell-free quantitative neutralization assay based on the competitive inhibition of trimeric SARS-CoV-2 spike protein binding to the angiotensin converting enzyme 2 (ACE2) receptor. This high-throughput method matches the performance of the gold standard live virus infection assay, as verified with a panel of 206 seropositive donors with varying degrees of infection severity and virus-specific IgG titers, achieving 96.7% sensitivity and 100% specificity. Furthermore, it allows for the parallel assessment of neutralizing activities against multiple SARS-CoV-2 spike protein variants of concern. We used our assay to profile serum samples from 59 patients hospitalized with coronavirus disease 2019 (COVID-19). We found that, although most sera had high activity against the 2019-nCoV parental spike protein and, to a lesser extent, the α (B.1.1.7) variant, only 58% of serum samples could efficiently neutralize a spike protein derivative containing mutations present in the β (B.1.351) variant. Thus, we have developed an assay that can evaluate effective neutralizing antibody responses to SARS-CoV-2 spike protein variants of concern after natural infection and that can be applied to characterize vaccine-induced antibody responses or to assess the potency of monoclonal antibodies.

The clinical outcome of SARS-CoV-2 infections, which can range from asymptomatic to lethal, is crucially shaped by the concentration of antiviral antibodies and by their affinity to their targets. However, the affinity of polyclonal antibody responses in plasma is difficult to measure. Here we used microfluidic antibody affinity profiling (MAAP) to determine the aggregate affinities and concentrations of anti–SARS-CoV-2 antibodies in plasma samples of 42 seropositive individuals, 19 of which were healthy donors, 20 displayed mild symptoms, and 3 were critically ill. We found that dissociation constants, Kd, of anti–receptor-binding domain antibodies spanned 2.5 orders of magnitude from sub-nanomolar to 43 nM. Using MAAP we found that antibodies of seropositive individuals induced the dissociation of pre-formed spike-ACE2 receptor complexes, which indicates that MAAP can be adapted as a complementary receptor competition assay. By comparison with cytopathic effect–based neutralisation assays, we show that MAAP can reliably predict the cellular neutralisation ability of sera, which may be an important consideration when selecting the most effective samples for therapeutic plasmapheresis and tracking the success of vaccinations.

In the first days of embryogenesis, transposable element–embedded regulatory sequences (TEeRS) are silenced by Kruppel-associated box (KRAB) zinc finger proteins (KZFPs). Many TEeRS are subsequently co-opted in transcription networks, but how KZFPs influence this process is largely unknown. We identify ZNF417 and ZNF587 as primate-specific KZFPs repressing HERVK (human endogenous retrovirus K) and SVA (SINE-VNTR-Alu) integrants in human embryonic stem cells (ESCs). Expressed in specific regions of the human developing and adult brain, ZNF417/587 keep controlling TEeRS in ESC-derived neurons and brain organoids, secondarily influencing the differentiation and neurotransmission profile of neurons and preventing the induction of neurotoxic retroviral proteins and an interferon-like response. Thus, evolutionarily recent KZFPs and their TE targets partner up to influence human neuronal differentiation and physiology.

Gene expression aberration is a hallmark of cancers, but the mechanisms underlying such aberrations remain unclear. Human endogenous retroviruses (HERVs) are genomic repetitive elements that potentially function as enhancers. Since numerous HERVs are epigenetically activated in tumors, their activation could cause global gene expression aberrations in tumors. Here, we show that HERV activation in tumors leads to the up-regulation of hundreds of transcriptional suppressors, namely, Krüppel-associated box domain–containing zinc-finger family proteins (KZFPs). KZFP genes are preferentially encoded nearby the activated HERVs in tumors and transcriptionally regulated by these adjacent HERVs. Increased HERV and KZFP expression in tumors was associated with better disease conditions. Increased KZFP expression in cancer cells altered the expression of genes related to the cell cycle and cell-matrix adhesion and suppressed cellular growth, migration, and invasion abilities. Our data suggest that HERV activation in tumors drives the synchronized elevation of KZFP expression, presumably leading to tumor suppression.

Physical interactions between proteins are essential for most biological processes governing life1. However, the molecular determinants of such interactions have been challenging to understand, even as genomic, proteomic and structural data increase. This knowledge gap has been a major obstacle for the comprehensive understanding of cellular protein–protein interaction networks and for the de novo design of protein binders that are crucial for synthetic biology and translational applications2–9. Here we use a geometric deep-learning framework operating on protein surfaces that generates fingerprints to describe geometric and chemical features that are critical to drive protein–protein interactions10. We hypothesized that these fingerprints capture the key aspects of molecular recognition that represent a new paradigm in the computational design of novel protein interactions. As a proof of principle, we computationally designed several de novo protein binders to engage four protein targets: SARS-CoV-2 spike, PD-1, PD-L1 and CTLA-4. Several designs were experimentally optimized, whereas others were generated purely in silico, reaching nanomolar affinity with structural and mutational characterization showing highly accurate predictions. Overall, our surface-centric approach captures the physical and chemical determinants of molecular recognition, enabling an approach for the de novo design of protein interactions and, more broadly, of artificial proteins with function.