Nsp2 and nsp4 interactors are conserved across CoV homologs, including ER proteostasis factors.

A. Schematic of nsp2 & nsp4 FLAG-tagged construct. The placement of the N-terminal FLAG-tag for nsp2 and C-terminal FLAG-tag for nsp4 was based on previous construct designs(3). A corresponding 18 - 23 amino acid long leader sequence from the C-terminus of nsp3, including the PL2pro cleavage site, is added on the N-terminus of nsp4 constructs to improve protein biogenesis and insertion, as has been reported previously(13, 82).

B. Schematic of comparative affinity-purification mass spectrometry (AP-MS) workflow to identify nsp-host interactors.

C. Enrichment of select nsp2 and nsp4 interactors conserved across multiple CoV homologs versus negative control (Tdtomato) background. nsp2: n = 2 IP replicates/homolog in 1 MS run; nsp4: n = 2-4 IP replicates/homolog in 1 MS run. See Figure S1, S2, Tables S1, S2 for full dataset.

Conserved nonstructural protein (nsp) host interactors are pro-viral factors for CoV infection, including malectin (MLEC).

A. Schematic of RNAi screening of prioritized interactors using a replicase reporter virus (MHV-FFL2, luminescence) or plaque assay (MHV, titer).

B. Correlation plot of nsp2 interactor knockdown effect against replicase reporter virus (MHV-FFL2, luminescence) versus infectious titer (MHV, plaque). Scramble siRNA (blue) was used as basal control for comparison. siCeacam1 (MHV cellular receptor) was used as positive control (purple). Proteostasis factors are highlighted in green. Infections were performed at MOI 0.1 for 10 hpi, n=3-5. Dot represents mean, crosshatch indicates SEM. See Figures S3, S4, S5.

C. Same as in (B), but of nsp4 interactor knockdowns. See Figures S3, S4, S5.

D. MHV titers at MOI 0.1 or 1 in DBT cells treated with Scramble, siCeacam1, or siMlec siRNA (10 hpi), as measured by plaque assay. Paired Student’s t-test for significance, p<0.05 considered significant, n = 3 (MOI 0.1), n = 3-4 (MOI 1).

E. Viral replicase reporter levels (luminescence) in DBT cells treated with Scramble, siCeacam1, individual siMlec (#1-4), or pooled siMlec siRNA, then infected with MHV-FFL2 (MOI 1, 9 hpi). Student’s t-test used for significance, with p<0.05 considered significant. n = 4. See Figure S7.

Nsp2 maintains a genetic interaction with MLEC and protein-protein interactions with MLEC-associated protein complexes during infection.

A. Schematic of the role of MLEC in glycoprotein quality control.

B. Comparison of MHV-WT vs. MHV-del2 (nsp2 deleted) infectious titers in siMlec vs. Scramble-treated DBT cells, expressed as Log10 fold change. MHV-WT (MOI 0.1, 10 hpi) or MHV-del2 (MOI 0.5, 10 hpi). n = 3-5, with Mann-Whitney T-test for significance, p <0.05 considered significant. See Figure S8.

C. Volcano plot of MHV GFP-nsp2 interactome during infection (MOI 1, 9 hpi). Comparison to MHV-WT infected cells was used as a background control for the GFP-IP. Linear filters of Log2 fold change > 1 and adjusted p-value > 0.05 were applied. Viral proteins (purple), OST complex factors (green), and translocon factors (blue) are annotated. MHV-GFP2 (n = 8 IPs), MHV-WT (n = 7 IPs) combined in 1 MS run. One-way ANOVA with multiple testing corrections was used for significance. See Figure S9, Table S3.

D. Schematic of FT-MLEC construct design and IP-MS experimental design. FT-MLEC construct is composed of an N-terminal signal sequence (SS), followed by a FLAG-tag, and the MLEC luminal domain. There is a 10-residue-long transmembrane helix with a short 2 amino acid long cytosolic C-terminal tail. Stably expressing FT-MLEC or WT DBT cells were infected with MHV (MOI 2.5, 9 hpi) and subjected to DSP (0.5 mM) crosslinking, before lysis and FLAG IP. Elutions were reduced, alkylated, trypsin digested, and analyzed by LC-MS/MS. Parental DBT and MHV (n = 4 IPs), FT-MLEC DBT and mock (n = 6 IPs), FT-MLEC DBT and MHV (n = 6 IPs/condition), pooled into 1 MS run. See Figure S10.

E. Volcano plot of FT-MLEC interactors during MHV infection. Samples were normalized to total peptide amounts. Linear filters of Log2 fold change > 1 or < -1 and adjusted p-value > 0.05 were applied. Viral proteins and MLEC (purple), downregulated (red), and OST complex factors (green) are annotated. T-test with multiple testing corrections for significance. See Figure S11, Table S4.

F. Waterfall plot of FT-MLEC interactome comparing MHV versus mock infection. Samples were normalized to FT-MLEC bait levels. Components of the OST complex (green), the glycoprotein processing pathway (orange), and the five most downregulated interactors (red) are annotated. See Figure S11.

MLEC promotes early CoV replication events and viral protein biogenesis.

A. Schematic of main events during the CoV replication cycle. Viral RNA (vRNA), double membrane vesicle (DMV), ER-Golgi Intermediate Compartment (ERGIC).

B. Abundance of positive-sense MHV genomic RNA ((+)gRNA) as measured by RT-qPCR (MHV WT, MOI 1, 2 hpi) and normalized to Rpl13a transcript levels. Paired Student’s T-test for significance, p<0.05 considered significant, n = 2-3.

C. MHV-FFL2 replicase reporter time course during infection from 2 – 10 hpi (MOI 1). Mock infected cells used as background luminescence control. Two-way ANOVA, Benjamini, Krieger, Yekutieli multiple comparisons, n = 3, * p<0.05, ** p<0.01, ***p<0.005, ****p<0.0001, color coding represents results of siMlec (pink) or siCeacam1 (teal) compared to Scramble control (black).

D. Abundance of negative-sense MHV genomic RNA ((-)gRNA) as measured by RT-qPCR (MHV WT, MOI 1, 10 hpi). Paired Student’s T-test, p<0.05 considered significant, n = 2-3. See Figure S12.

E. Relative abundance of intracellular viral proteins during MHV infection in DBT cells treated with Scramble or siMlec siRNA (MOI 1, 10 hpi), measured by quantitative proteomics. Purple protein names denote viral glycoproteins. Spike protein (S), membrane protein (M), nucleocapsid protein (N). One-way ANOVA with multiple testing correction. * p<0.05, ** p<0.01, ***p<0.005, ****p<0.0001, n = 6-7 biological replicates/condition across 2 MS runs. See Figure S13, Table S5.

Primers used in this study.

MLEC mediates replication through the glycoprotein quality control pathway.

A. Schematic of glycoprotein biogenesis and quality control pathway. NGI-1 is an OST inhibitor that blocks N-linked glycosylation(71).

B. Representative Western blot of DBT lysates after OST inhibition and MLEC KD during MHV-FFL2 infection (MOI 1, 10 hpi). Blotting for glycosylated and non-glycosylated ERDJ3 served as a marker of N-linked glycosylation inhibition. Lysates were matched with the corresponding luminescence reading. See Figure S14.

C. Levels of MHV-FFL2 replicase reporter with NGI-1 or DMSO treatment at the time of infection in cells with Scramble, CEACAM1, or MLEC KD in the same lysates analyzed in (B). One replicate of siCeacam1 fell below the limit of detection (dotted line). Mock-infected cells were used as a control for background luminescence. Student’s T-test for significance, with p<0.05 considered significant, n = 6. See Figure S14.

D. Abundance of N-linked glycosylated Spike peptides in siMlec versus Scramble-treated DBT cells. Cells were treated with siRNA and infected with mock or MHV (MOI 1, 10 hpi). Lysates were trypsinized, TMTpro labeled, pooled, then treated with PNGase F to cleave oligosaccharides, resulting in a deamidation mass shift (N to D) detectable by LC-MS/MS. Peptide sequence is listed, with “N*” denotating a deamidation modified Asn in a canonical glycosylation sequon (N-X-S/T, where X is not Proline). Peptides where the modified Asn could not be resolved lack the “N*” annotation. Peptide position within the protein is in brackets. One-way ANOVA, adjusted p-value. * p<0.05, ** p<0.01, ***p<0.005, ****p<0.0001, n = 4/condition, 1 MS run. See Figure S15, Table S6.

MLEC knockdown suppresses SARS-CoV-2 replication.

A. Luminescence levels of a secreted SARS-CoV-2 replicon nano-Glo luciferase (sec-nLuc) reporter(72) for Delta or Omicron in HEK293T cells treated with siMLEC or Scramble. GFP was used as a negative control. Omicron data is displayed in subset box. Delta, n = 4; Omicron, n = 5; Student’s t-test for significance, with p < 0.05 considered significant. Bars represent mean values, ± SEM. See Figure S16.

B. Data from (A) presented as a fold change comparison of siMLEC vs. Scramble treatment with respective SARS-CoV-2 replicon. Mean fold change value is annotated above the graph (0.43 and 0.41 respectively). Bars represent mean values, ± SEM.