Coronavirus replication-transcription complex: the important and selective NMPylation of NiRAN-RdRp subunits to conserved sites in nsp9

Edited by Peter Sarnow, Stanford University School of Medicine, Stanford University, California, approved on December 25, 2020 (reviewed on October 25, 2020)

We report the interaction between subunits in the replication of coronaviruses-transcription complexes, which are essential for replication and evolutionary conservation. We provided evidence that the NiRAN domain associated with nsp12 has nucleoside monophosphate (NMP) transferase activity in trans, and identified nsp9 (an RNA binding protein) as its target. NiRAN catalyzes the covalent attachment of the NMP moiety to the conserved nsp9 amino terminus in a reaction that relies on Mn2+ ions and adjacent conserved Asn residues. It was found that NiRAN activity and nsp9 NMPylation are essential for coronavirus replication. The data allows us to link this activity of the nested virus enzyme marker to the previous observations in the hypothesis that the initiation of RNA synthesis in a class of RNA viruses is functionally and evolutionarily consistent.

The RNA-dependent RNA polymerase (RdRps) of Nidovirales (Coronaviridae, Arterioviridae, and 12 other families) is linked to the amino-terminal (N-terminal) domain in the non-structural protein (nsp) released from the polyprotein, called NiRAN 1ab is composed of viral main protease (Mpro). Previously, the arterial virus NiRAN-RdRp nsp’s own GMPylation/UMPylation activity was reported, and it was suggested to generate a transient for the transfer of nucleoside monophosphate (NMP) to the (currently unknown) virus and/or cell biopolymerization Things. Here, we show that the coronavirus (Human Coronavirus [HCoV]-229E and Severe Acute Respiratory Syndrome Coronavirus 2) nsp12 (NiRAN-RdRp) has Mn2+-dependent NMPylation activity, which is derived from nsp9 through the formation of Mpro-mediated nsp9 After the N-terminal flanking nsps is proteolytically released, the phosphoramidate is bonded to the primary amine (N3825) at the N-terminal of nsp9. Uridine triphosphate is the preferred nucleotide in this reaction, but adenosine triphosphate, guanosine triphosphate, and cytidine triphosphate are also suitable co-substrates. Mutation studies using recombinant coronavirus nsp9 and nsp12 proteins and genetically engineered HCoV-229E mutants determined the residues necessary for NiRAN-mediated nsp9 NMPylation and virus replication in cell culture. The data confirmed the prediction of NiRAN active site residues and determined the important role of nsp9 N3826 residues in nsp9 NMPylation and virus replication in vitro. This residue is part of the conserved N-terminal NNE tripeptide sequence and proved to be the only invariant residue of nsp9 and its homologs in the coronavirus family. This study provides a solid foundation for the functional study of NMPylation activity of other nested viruses and proposes possible targets for the development of antiviral drugs.

Nidovirales positive-stranded RNA virus infects a variety of vertebrates and invertebrates (1, 2). The order currently includes 14 families (3), of which the Coronavirus family has been extensively studied in the past 20 years. At that time, three zoonotic coronaviruses emerged from animal hosts and caused large-scale outbreaks of severe respiratory infections in humans. Including persistent pandemics caused by severe acute infectious diseases. Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) (4âââ7). Nidoviruses share a common genome organization, and the subunit of the membrane-bound replication-transcription complex (RTC) is encoded in the 5-?²-terminal two-thirds and the main structural subunit of the virus particle, as well as some accessories.  Protein, encoded in the 3??² end third of the genome (1). Except for one family of planarian viruses (Monoviridae) (8), all nested viruses encode RTC subunits in two large open reading frames (ORF) ORF1a and ORF1b, which are translated from genomic RNA of. ORF1a encodes polyprotein (pp) 1a, and ORF1a and ORF1b jointly encode pp1ab. With the general participation of the main protease (Mpro) encoded by ORF1a, both pp1a and pp1ab are proteolytically processed into a variety of non-structural proteins (nsps), also known as 3CLpro, because it has homology with the 3Cpro of the picornavirus (9). These nsps are thought to be assembled into a large dynamic RTC, catalyze the synthesis of genomic RNA (replication) and a set of subgenomic RNA (transcription), and are used to coordinate the expression of the ORF located downstream of ORF1b (10? ??12).

The core RTC includes RNA-dependent RNA polymerase (RdRp) (13), superfamily 1 helicase (HEL1) (14, 15) and several RNA processing enzymes, which are mainly encoded in ORF1b and in the coronavirus family It contains nsp12-nsp16 and nsp9-nsp12 in the Arterioviridae family (see reference 10ââ 12). RdRp and HEL1 represent two (one-fifth) conserved domains of the bird’s nest virus and have homology among other RNA viruses. Core replicase is believed to be assisted by other subunits, including several small nsps released from the carboxy-terminal (C-terminal) region of pp1a, downstream of Mpro (coronavirus nsp5 and arterial virus nsp4, respectively). They have limited family-specific protection and diverse activities (reviewed in reference 10ââ12).

Relatively recently, a domain with unique sequence motif characteristics was found at the amino terminus (N-terminus) adjacent to RdRp in all nested viruses, but no other RNA viruses (16). Based on its location and nucleotide transferase (nucleoside monophosphate [NMP] transferase) activity, this domain is named NiRAN (Nestvirus RdRp-related nucleotide transferase). The dual-domain combination of NiRAN-RdRp constitutes nsp12 in the Coronaviridae family and nsp9 in the Arterioviridae family, and in other nestoviridae, NiRAN-RdRp is expected to be released as an independent nsp from the viral polyprotein. In the coronavirus, the NiRAN domain contains ??1/450 residues and is connected to the C-terminal RdRp domain through the linker region (16?19). In Equine Arteritis Virus (EAV) (Arteriviridae), recombinant nsp9 shows Mn2+ ion-dependent (self) UMPylation and GMPylation activities, which depend on three conserved sequence bases in nestovirus, AN, BN and CN The residues in the sequence. Where N stands for NiRAN) (16). The N-terminal flanking of these motifs is a less conservative motif preAN. Some of these residues are also conserved in distantly related protein kinases, where they have been shown to be involved in nucleoside triphosphate (NTP) binding and catalytic activity (20, 21). Consistent with this observation, several key active site residues in the pseudokinase SelO from Pseudomonas syringae can be assembled with the recently published SARS-CoV-2 nsp7/8/12/13 supercomplex. Conserved Coronavirus NiRAN residues superimposed in the electron microstructure. Recombinant protein (17). It is speculated that the documented (self) U/GMPylation will produce a transient state to transfer NMP to the (currently unknown) substrate (16), and the structural similarity between NiRAN and protein kinase (17, 19) ) Is the hypothesis that NiRAN modifies other proteins.

Many features, including its unique and unique systematic association with nested viruses and genetic separation from RdRp, make NiRAN a reasonable key regulatory enzyme for nested viruses, which is critical to their emergence and identity. Previously, three possible functions involving NiRAN to regulate genome/subgenomic translation or replication/transcription were called. When considering the scarce and incomplete data available at the time, each function has its advantages and disadvantages (16). In this research, we aim to combine the biochemical and reverse genetic studies of the coronaviruses representing the two genera, and put our findings in the evolutionary background of the natural mutation of the coronavirus family, so as to gain insight into this Mysterious realm. We report major advances in the understanding of NiRAN through the identification of natural targets in RTC, which (among the three available hypotheses) contributes to the role of this domain in initiating the synthesis of nested virus RNA. This research also opens up possibilities for other roles of NiRAN on the virus host interface.

In order to characterize the enzymatic properties of the corona virus nsp12-related NiRAN domain, we produced a recombinant form of human coronavirus 229E (HCoV-229E) nsp12 in E. coli, with a His6 tag at the C-terminus, and combined the protein with [α32-P ] Incubate together with NTP in the presence of MnCl2 as described in Materials and Methods. The analysis of the reaction product indicated the presence of a radiolabeled protein co-migrating with nsp12 (106 kDa), indicating that coronavirus nsp12 catalyzes the formation of covalent protein-NMP adducts, preferentially formed with uridine monophosphate (UMP) (Figure 1A) And B). Quantitative analysis showed that compared with other nucleotides, the signal intensity of UMP incorporation increased by 2 to 3 times (Figure 1C). This data is consistent with the predicted NMP transferase activity of the NiRAN domain of the coronavirus (16), but indicates that the nucleotide preferences of the NiRAN domain of the coronavirus and the arterial virus are different.

Self-NMPylation activity of HCoV-229E nsp12. (A) HCoV-229E nsp12-His6 (106 kDa) was incubated with the designated [α-32P] NTP in the presence of 6 mM MnCl2 for 30 minutes (see Materials and Methods for details). The reaction products were separated by SDS-PAGE and stained with Coomassie brilliant blue. (B) The radiolabeled protein is visualized by phosphorous imaging. The positions of nsp12-His6 and protein molecular mass markers (in kilodaltons) are shown in A and B. (C) The intensity of the radioactive signal (mean ± SEM) was determined from three independent experiments. *P≤0.05. The signal strength (percentage) is related to UTP.

Although NiRAN-related enzyme activities have been shown to be essential for the replication of EAV and SARS-CoV in cell culture (16), the specific NiRAN function and potential targets have not yet been determined. The recently reported structural similarity between NiRAN and a family of proteins with protein kinase-like folds (17, 22) prompted us to test the hypothesis that NiRAN catalyzes the NMPylation of other proteins. We generated a set of potential homologous targets, including non-structural proteins encoded by HCoV-229E ORF1a (nsps 5, 7, 8, 9, 10), each containing a C-terminal His6 tag (SI appendix, Table S1) , And incubate these proteins with [α32-P] uridine triphosphate ([α32-P]UTP) in the presence or absence of nsp12. Bovine serum albumin and MBP-LacZα fusion protein produced in E. coli served as controls (Figure 2A, lanes 1 to 7). The radiolabeled protein was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and autoradiography, and it was found that there was a strong radioactive signal in the reaction containing nsp12 and nsp9. The position of the signal corresponds to the molecular mass of nsp9, indicating nsp12-mediated UMPylation of nsp9 (Figure 2B, track 7). No other test proteins were found to be UMPylated, which led us to conclude that nsp9 is a specific substrate of nsp12. Consistent with the self-NMPylation data shown in Figure 1, nsp12 is able to transfer all four NMPs to nsp9, although the efficiency is different, UMP> adenosine monophosphate (AMP)> guanosine monophosphate (GMP)> cytidine monophosphate (CMP) ) (Picture). 3 A and B). Under the conditions used in this assay (shorten the reaction and exposure time, reduce the concentration of nsp12; materials and methods), the self-NMPylation of nsp12 could not be detected (compare Figure 2B, lane 7, and Figure 1B), which proved an effective (And multiple rounds) UMP moved from nsp12 to nsp9. UMP transferase activity requires the presence of Mn2+ ions, as shown in Figure 3C, while only minimal UMP transferase activity was observed in the presence of Mg2+, and no activity in the presence of the other two divalent cations tested. Similar data was obtained in NMPylation assays containing cytidine triphosphate (CTP), guanosine triphosphate (GTP), and adenosine triphosphate (ATP) (SI appendix, Figure S1).

HCoV-229E nsp12-mediated UMPylation of nsp9. A series of protein substrates (including bovine serum albumin, MBP-lacZα, and a series of HCoV-229E nsps labeled with C-terminal His6 encoded by ORF1a) were used to evaluate the UMPylation activity of HCoV-229E nsp12-His6⁺-mediated protein. Incubate the protein with [α-32P] UTP for 10 minutes in the absence (A) or presence (B) of nsp12 as described in the materials and methods. At the top of A and B, SDS-polyacrylamide gel stained with Coomassie Brilliant Blue is shown, and at the bottom of A and B, the corresponding autoradiograms are shown. The position of the protein molecular mass marker (in kilodaltons) is given on the left. The position of nsp12-His6 (B, top) and the radioactive signal observed during the incubation of nsp12-His6 with nsp9-His6 (B, lane 7) are also indicated, which indicates that [α-32P]UMP to nsp9-His6 ( 12.9 kDa), which was not observed for other proteins tested.

HCoV-229E NiRAN-mediated biochemical and virological characterization of nsp9 NMPylation. (A and B) The role of the nucleotide co-substrate used in the reaction. Nsp12-His6 and nsp9-His6 are mixed and incubated in the presence of different [α-32P] NTPs in the standard NMPylation assay. (A, top) Coomassie-stained nsp9-His6 separated by SDS-PAGE. (A, bottom) Autoradiograph of the same area of ​​the gel. (B) The relative activity (mean ± SEM) in the presence of the designated nucleotide cofactor is determined from three independent experiments. *P≤0.05. (C) The role of metal ions. Shown is the standard NMPylation test in the presence of [α-32P] UTP and different metal ions, each with a concentration of 1 mM. In C, the top, Coomassie stained nsp9-His6 is shown, and in C, the bottom, the corresponding autoradiography is shown. The size of the labeled protein (in kilodaltons) is shown to the left of A and C. (D) The mutant form of HCoV-229E nsp12-His6 carrying the specified amino acid substitution is in [α-32P]UTP, as described in Materials and Methods. The radiolabeled nsp9-His6 produced in the NMPylation reaction is detected by phosphorylation imaging (D, top). The relative activity compared with the wild-type (wt) protein is shown in D, and the bottom is taken as the average (±SEM) from three Independent experiment. Asterisks indicate substitutions of non-conserved residues. (E) The virus titer in the culture supernatant of p1 cells obtained 24 hours after infection was determined by plaque assay. The codon substitutions in the NiRAN domain of the engineered HCoV-229E mutant are indicated (residue numbering is based on their position in pp1ab). The replication-deficient RdRp active site mutant nsp12_DD4823/4AA was used as a control.

In order to gain a deeper understanding of the active site of NiRAN and determine the residues related to the activity of nsp9-specific NMP transferase, we performed mutation analysis, in which we replaced the conservative residues in the NiRAN AN, BN and CN motifs (16) It is Ala (SI appendix, Figure S2). In addition, the impact of conservative Arg-to-Lys or Lys-to-Arg substitutions was evaluated in two cases. As a (negative) control, residues that are not or less conserved in the NiRAN domain of coronaviruses and other nested viruses are replaced with Ala. Replacing K4116A (in motif preAN), K4135A (AN), R4178A (BN), D4188A (motif BN) and D4280A (CN) significantly reduces or even eliminates nsp9 NMPylation through nsp12, while proteins with conservative substitutions (R4178K) , K4116R) retain 60% and 80% of their activity, which indicates that the relaxation of the restrictions on their respective side chains is physicochemically sensitive (Figure 3D). Replacing several other conserved residues E4145A, D4273A, F4281A and D4283A is much less harmful, and nsp9 UMPylation is only moderately reduced. Similar results were obtained in nsp9 NMPylation reactions involving other NTPs (Figure 3D and SI appendix, Figure S3), confirming that the observed effects on specific amino acid substitutions are independent of the type of nucleotide co-substrate used. Next, we tested the possible impact of these nsp12 substitutions on the replication of coronaviruses in cell culture. To this end, we used appropriate genetically engineered complementary DNA (cDNA) templates cloned in recombinant vaccinia virus (23, 24) to transcribe 5 -7 cells. Titration of infectious virus progeny produced in these cells showed that most HCoV-229E NiRAN mutants were not feasible (Figure 3E). A group of non-viable viral mutants includes alternatives that have been shown to eliminate or significantly reduce NMP transferase activity in vitro (K4116A, K4135A, R4178A, D4188A, D4280A, D4283A), but there are two other alternatives (K4116R, E4145A) 80% reserved? Their in vitro NMPylation activity suggests that additional restrictions are involved. Similarly, two other mutations (R4178K, F4281A) that caused a moderate decrease in NiRAN’s in vitro NMPylation activity produced live viruses, however, these viruses significantly reduced titers through replication. Consistent with the in vitro activity data shown in Figure 3D, replacing four other residues that are not conserved in coronavirus and/or other nested viruses (K4113A, D4180A, D4197A, D4273A) (8, 16) produced viable viruses The offspring, despite having a moderately reduced titer compared to the wild-type virus (Figure 3E).

In order to study whether NiRAN-mediated NMP transferase activity depends on the active RdRp domain, the two conserved Asp residues involved in the coordination of divalent metal ions (11) in RdRp motif C were replaced by Ala. The resulting protein nsp12_DD4823/4AA retains its nsp9 NMPylation activity, indicating that nsp12-mediated in vitro nsp9 NMPylation activity does not require polymerase activity (SI Appendix, Figure S4).

After establishing the nsp9-specific NMP transferase activity for nsp12, we tried to characterize the NMP-nsp9 adduct by mass spectrometry (MS). The complete protein mass spectrum of recombinant HCoV-229E nsp9 showed a peak at 12,045 Da (Figure 4A). The addition of nsp12 did not change the quality of nsp9, indicating that nsp12 and nsp9 would not form a stable complex under the conditions used (denaturation) (Figure 4A). In the presence of UTP and GTP, the mass measurement of the reaction containing nsp9 and nsp12 respectively showed that the protein mass of UTP moved 306 Da, and the protein mass of GTP moved 345 Da, indicating that each nsp9 molecule binds a UMP or GMP (Picture 4) C and D). It is speculated that the energy required for NiRAN-mediated nsp9 NMPylation comes from NTP hydrolysis and pyrophosphate release. Although a 10-fold molar excess of nsp9 (target) than nsp12 (enzyme) was used in this reaction, almost complete NMPylation of nsp9 was observed, indicating that the interaction between nsp12 and nsp9 is short-lived, and nsp12 can NMPylate more nsp9 in vitro molecule.

Single NMPylation of nsp9 in the presence of nsp12 and UTP or GTP. Shown is the deconvoluted complete protein mass spectrum of HCoV-229E nsp9 (SI appendix, Table S1) (AD). (A) nsp9 alone, (B) nsp9 + nsp12-His6, (C) nsp9 + nsp12-His6 in the presence of UTP, (D) nsp9 + nsp12-His6 in the presence of GTP.

To determine the nsp9 residues UMPylated by nsp12, nsp9-UMP was cleaved with trypsin. The resulting peptides were separated by nano-high performance liquid chromatography (HPLC) and analyzed by tandem mass spectrometry (MS/MS) online. Data analysis using the Byonic software package (Protein Metrics) showed UMPylation of the N-terminal amino acid. This is confirmed manually. The tandem mass spectrum of the precursor peptide [UMP]NNEIMPGK (SI appendix, Figure S5A) revealed a fragment at 421 m/z, indicating that UMP binds to residue 1 of nsp9.

At the N-terminus of nsp9, Asn is conserved among the members of Orthocoronavirinae (SI appendix, Figure S6). Although we believe that the N-terminal primary amine nitrogen is the most likely acceptor for UMP, we decided to obtain additional evidence of NMP binding at the N-terminal. For this reason, the non-NMPylated and NMPylated N-terminal peptide nsp9 purified by HPLC was derived in the presence of acetone and sodium cyanoborohydride. Under these conditions, only free primary amines can be modified with propyl (25). The N-terminal nsp9-derived peptide with the sequence NNEIMPGK contains two primary amines, one at the N-terminus of Asn and the other at the side chain of Lys at the C-terminus. Therefore, propyl groups can be introduced at both ends. The extracted ion chromatograms of non-NMPylated peptides are shown in the SI appendix, Figure S5B. As expected, N-terminal and C-terminal (mono)propylated (SI appendix, Figure S5B, upper lane) and dipropylated peptides (SI appendix, Figure S5B, lower lane) can be identified. This pattern changes with the use of the NMPylated N-terminal peptide of nsp9. In this case, only C-terminal propylated peptides can be identified, but N-terminal propylated peptides and dipropylated peptides are not identified (SI Appendix, Figure S5C), indicating that UMP has been transferred to the N-terminal primary amine To prevent this group from making changes.

Next, we replace (with Ala or Ser) or delete the conserved residues at the N-terminus of nsp9 to define target-specific constraints. Based on our MS data showing that NiRAN forms an nsp9-NMP adduct with the primary amine of the N-terminal residue of nsp9, we hypothesized that nsp9 NMPylation requires the viral master protease (Mpro, nsp5) to release the nsp9 N-terminal from its polyprotein precursor. To test this hypothesis, we produced a precursor protein nsp7-11 containing nsp9 in E. coli and performed a standard NMPylation test in the presence of [α-32P] UTP (materials and methods). As shown in Figure 5A (lane 3), the uncut nsp7-11 precursor is not radiolabeled with nsp12. In contrast, if nsp7-11 is cleaved by recombinant nsp5 to release nsp9 (and other nsps) from the precursor, a radiolabeled protein that migrates with nsp9 is detected, confirming our conclusion that NiRAN and N- Selective formation of covalent nsp9-NMP adducts. The terminal primary amine of the N-terminal Asn (position 3825 in pp1a/pp1ab). This conclusion is also supported by experiments using the nsp9 construct, which contains one or two additional residues at the N-terminus. In both cases, NiRAN-mediated UMPylation of nsp9 was abolished (SI Appendix, Figure S7). Next, we produced a protein with one or two Asn residues deleted from the 3825-NNEIMPK-3832 peptide sequence at the N-terminal of nsp9. In both cases, nsp9 UMPylation was completely blocked (Figure 5B), providing additional evidence that the real nsp9 N-terminus acts as an NMP receptor.

The proteolytic processing of nsp9 and the role of N-terminal residues in nsp12-mediated UMPylation. (A) nsp9 UMPylation requires a free nsp9 N-terminal. Nsp7-11-His6 is pre-incubated at 30 °C in NMPylation detection buffer containing UTP in the presence or absence of recombinant Mpro (nsp5-His6). After 3 hours, start the NMPylation assay by adding nsp12-His6 as described in Materials and Methods. The reaction containing nsp5-His6 (lane 1) and nsp9-His6 (lane 2) was used as a control. After 10 minutes, the reaction was terminated and the reaction mixture was separated by SDS-PAGE. The protein was stained with Coomassie Brilliant Blue (A, top). The Nsp7-11-His6 precursor and the processed product resulting from nsp5-His6 mediated cleavage are shown on the right. Please note (due to their small size) that nsp7 and nsp11-His6 are not detectable in this gel, and the reaction is supplemented with nsp5-His6 (lanes 1 and 4; the position of nsp5-His6 is indicated by a solid circle) or nsp9-His6 (Lane 2) contains a small amount of MBP (indicated by open circles) as residual impurities because they are expressed as MBP fusion proteins (SI appendix, Table S1). (B) The Nsp9-His6 variant lacks one or two N-terminal Asn residues (residue numbering according to the position in pp1a/pp1ab) and is purified and incubated with nsp12-His6 and [α-32P] UTP. B, SDS-PAGE stained with Coomassie is shown at the top, B, the corresponding autoradiograph is shown at the bottom. The position of the molecular weight marker (in kilodaltons) is shown on the left. (C) HCoV-229E nsp9-His6 N-terminal conserved residues were replaced with Ala or Ser, and the same amount of protein was used in the nsp12-His6 mediated UMPylation reaction. The reaction products were separated by SDS-PAGE and stained with Coomassie Brilliant Blue (C, top), and radiolabeled nsp9-His6 was detected by phosphorescence imaging (C, middle). Using wild-type (wt) protein as a reference (set to 100%), the relative NMPylation activity (mean ± SEM) was calculated from three independent experiments. (D) Virus titers in the p1 cell culture supernatant of Huh-7 cells infected with HCoV-229E wild-type Huh-7 cells, and mutants carrying designated amino acid substitutions in nsp9 were determined by plaque assay. The replication-deficient RdRp motif C double mutant DD4823/4AA was used as a negative control.

The N-terminus of nsp9 (especially positions 1, 2, 3, and 6) is very conserved among members of the Orthocoronavirinae subfamily (SI appendix, Figure S6). In order to study the possible role of these residues in nsp12-mediated nsp9 NMPylation, two consecutive Asn residues at the N-terminus of nsp9 were replaced with Ala or Ser (alone or in combination). Compared with wild-type nsp9, replacing N3825 with Ala or Ser resulted in more than a two-fold reduction in nsp12-mediated UMPylation (Figure 5C). Consistent with our conclusion that NMPylation occurs at the N-terminal primary amine instead of the side chain of the N-terminal residue, we observed significant residual NMPylation with the replacement of N3825A and N3825S. Interestingly, if the second Asn is replaced by Ala or Ser, nsp9 UMPylation is reduced more strongly (more than 10 times), while the replacement of Ala at positions 3, 4, and 6 has only a moderate effect on nsp9 UMPylation (Figure 2) . 5C). Similar results were obtained using ATP, CTP or GTP (SI appendix, Figure S8). Collectively, these data indicate the key role of N2826 (position 2 in nsp9) in nsp9 NMPylation.

In order to obtain additional evidence of the functional correlation between the N-terminus of nsp9 and NMPylation, we performed a multiple sequence alignment (MSA) of the nsp9 sequence of the Coronavirus family (varying between 104 and 113 residues) (SI Appendix, Figure S6 ). In total, in 47 (known and putative) species of 5 genera of the Orthocoronavirinae subfamily that infect different mammals, birds, and reptile hosts, only 8 residues in total were found to be invariant. The most extensive changes, including deletions and insertions, were observed in the cycles between the secondary structure elements of nsp9, as determined by previous structural studies (26 ??28). Five invariant residues were found in the β strand and α helix of the C-terminal part of nsp9. Three invariant residues constitute the NNE motif of the N terminus of nsp9. It is revealed that the second Asn of this motif is the only invariant residue, which is also shared by the hypothetical nsp9 of the distantly related frog coronavirus, and represents the Microhyla letovirus 1 species in the subfamily Letovirinae of Alphaletovirus. The conservation of residues in nsp9 secondary structure elements can be rationalized by structural considerations to maintain folding or known RNA binding properties. However, this reasoning does not seem to apply to the conservation of NNE, and prior to this study, the nature of the constraints that limit the variation of the tripeptide sequence was completely obscured.

In order to determine the importance of nsp9-NMPylation and NNE conservation in coronavirus replication, we produced HCoV-229E mutants, which carry single or double substitutions of nsp9 N-terminal residues, indicating that nsp9 NMPylation is harmful in vitro. Before we start, we try to answer the question whether these substitutions (near the nsp8|9 cleavage site) affect the proteolytic processing of the C-terminal pp1a region. A set of nsp7-11 polyprotein constructs containing corresponding substitutions at the N-terminus of nsp9 were produced in E. coli and cut with recombinant Mpro. The proteolytic cleavage of the four sites (including the nsp9 flanking site) is not significantly affected by any introduced substitutions (SI appendix, Figure S9), excluding structural changes in these proteins that interfere with Mpro-mediated nsp8|9 cleavage ( Or other) website.

Huh-7 cells were transfected with genome-length HCoV-229E RNA, encoding Ala or Ser substitutions in the conserved NNE tripeptides (N3825, N3826, and E3827) at the nsp9 N terminus, showing that most of the mutations are fatal. We were able to rescue the virus by replacing the Ser or Ala of the N-terminal Asn (N2835A or N2835S), but failed to recover the virus with other single and double mutations in the NNE sequence (N3826A, N3826S, NN3825/6AA, NN3825/6SS) , E3827A) (Figure 5D).

These results indicate that the replication of coronaviruses in tissue culture is restricted (same or similar), limiting the natural mutation of nsp9 NMPylation sites in the body, and supporting the key role of this response in the life cycle of coronaviruses.

In the last set of experiments, we produced C-terminal His6 labeled SARS-CoV-2 nsp12 and nsp9, and two mutant forms of nsp12 in E. coli. The active site residues in the NiRAN and RdRp domains were respectively Use Ala instead (Figure 6A and SI appendix, Table S2). K4465 in SARS-CoV-2 nsp12 corresponds to K4135 in HCoV-229E (SI Appendix, Figure S2), which proved to be required for NiRAN activity and HCoV-229E replication (Figure 3D and E). This residue also corresponds to the arterial virus EAV nsp9 K94 residue, which was previously shown to be necessary for NiRAN self-UMPylation/self-GMPylation (16). As shown in Figure 6B, SARS-CoV-2 nsp12 has UMP transferase activity using nsp9 as a substrate, while the nsp12_K4465A active site mutant is inactive. The double substitution in the SDD characteristic sequence of RdRp motif C does not affect UMP transferase activity (Figure 6B), indicating that RdRp activity has no direct effect in nsp9 UMPylation. Similar data was obtained using CTP, GTP and ATP (SI appendix, Figure S10). In summary, these data indicate that NiRAN-mediated nsp9 NMPylation has a conservative activity in coronaviruses representing different genera of the orthocoronavirus subfamily.

SARS-CoV-2 nsp12-mediated NMPylation of nsp9. (A) Coomassie stained SDS-polyacrylamide gel showing the recombinant protein used in the NMPylation test. As a control, a mutant protein with active site substitution in the NiRAN domain (K4465A) and RdRp domain (DD5152/3AA) of SARS-CoV-2 nsp12 was used. Residue numbering is based on position in pp1ab. (B) Autoradiograph of UMPylation detection using nsp9-His6 and [α-32P]UTP as the substrate of nsp12-His6 (wild type [wt] and mutant). The molecular mass (in kilodaltons) of the labeled protein is shown on the left.

NiRAN domains are generally conserved in Nidovirales (16), indicating that they catalyze enzymatic reactions essential for Nidovirus replication. In this study, we were able to prove that the NiRAN domain of the coronavirus transfers NMP (generated from NTP) to nsp9, a mysterious RNA binding protein involved in virus replication (26  ?? 29 ), to determine it as a natural target and partner of coronavirus RTC.

The NiRAN domain shares three sequence motifs (AN, BN, and CN), which contain a very small number of residues that are conserved in all families in the monophyletic but highly differentiated Nidovirales order (8, 16) . Recent studies have shown that they are structurally related to a largely uncharacterized family of protein kinase-like proteins, which were originally called the SelO family (17, 19, 22, 30, 31). SelO-related proteins have kinase folds, but lack several conserved active site residues in classical kinases (22, 32). Based on the reverse orientation of ATP molecules bound to the active site and stabilized by specific interactions, SelO was hypothesized and subsequently confirmed to transfer AMP (instead of phosphate) to the protein substrate (22), while another bacterial SelO-like protein YdiU has recently been shown to catalyze the covalent attachment of UMP to Tyr and His residues of different protein substrates (33).

In order to confirm and expand the prediction of the putative active site residues of the coronavirus NiRAN domain, we used biochemical and reverse genetics methods to perform mutation analysis on the coronavirus nsp12 (Figure 3D and E and SI appendix, Figure S3 and table) S1â S4). The data shows that the replacement of HCoV-229E K4135, R4178 and D4280 with Ala eliminates in vitro NMP transferase activity and virus replication in cell culture (Figure 3D and E and SI appendices, Figure S3), supporting their presence in NTP γ-phosphate ( K4135, R4178) and the coordination of active site metal ions (D4280). The E4145A substitution of the conserved Glu in the range of the bird’s nest virus predicted to stabilize the K4135 (17) position was shown to eliminate viral replication, but surprisingly, the activity was retained in the in vitro NMPylation assay (Figure 3D and E and SI appendix, Figure S3 And Tables S1–S4). A similar observation was made when the corresponding substitution was introduced in the YdiU homolog of Salmonella typhimurium (E130A) (33). Taken together, these data support the regulatory function of this conserved residue rather than the catalytic function.

Replacing the conserved Phe residue (F4281A) within the range of nestovirus in the HCoV-229E NiRAN domain (8) resulted in a decrease in NMPylation activity in vitro and a significant decrease in virus replication in cell culture (Figure 3D, E and SI) appendix, Figure S3). The data is consistent with the important regulatory function of this residue, such as the homologous DFG motif Phe residue shown previously. In classical protein kinases, it is part of the Mg2+ binding loop and helps to assemble and regulate the spine? ? ? Required for effective catalytic activity (32, 34). Substituting Ala and Arg for K4116 residues (in the preAN motif), respectively, eliminated viral replication and, as expected, had different effects on NMP transferase activity in vitro, depending on the amino acid side chain introduced (Figure 3D And E and SI appendices, Figure S3). The functional data is consistent with the structural information, indicating that this residue has established an interaction with ATP phosphate (17). In the NiRAN domain of other nested virus families, the position of HCoV-229E pp1a/pp1ab K4116 is occupied by Lys, Arg or His (8), indicating that the functional restriction of this specific residue has been relaxed. The substitution of D4188A and D4283A eliminates or strongly reduces enzyme activity and eliminates virus replication (Figure 3). These two residues are conserved in most (but not all) nested viruses (8), indicating an important family-specific but possibly non-catalytic function. Ala substitutions of several other Lys and Asp residues (K4113A, D4180A, D4197A and D4273A) that are not conserved in the Coronaviridae or other Nestioviridae families (8) were used as controls. As expected, these substitutions are largely tolerable, with a slight decrease in enzyme activity and viral replication in some cases (Figure 3 and SI appendix, Figure S3). Overall, the coronavirus mutagenesis data is very consistent with the self-GMP and reverse genetics data of EAV NiRAN-RdRp (16), in which EAV nsp9 (coronavirus nsp12 ortholog) residue K94 (corresponding to HCoV- 229E K4135) important functions), R124 (corresponding to R4178), D132 (corresponding to D4188), D165 (corresponding to D4280), F166 (corresponding to F4281). In addition, the HCoV-229E mutagenesis data is consistent with and expanded from the previously reported SARS-CoV reverse genetics data (16), just as very similar to those observed for the corresponding CN motif Phe-to-Ala mutant SARS-CoV_nsp12 The phenotype described -F219A and HCoV-229E_F4281A (Figure 3 D and E and SI appendix, Figure S3 and Table S1-S4).

Compared with EAV orthologs (16), which have a clear preference for UTP and GTP (in the self-NMPylation reaction), our study shows that the coronavirus NiRAN domain (represented by HCoV-229E and SARS-CoV-2) can be effectively transferred All four NMPs, although there is a slight preference for UMP (Figures 1 and 3). The relatively low specificity of the specific NTP co-substrate is consistent with the recently reported SARS-CoV-2 nsp7/8/12/13 supercomposite structure, in which ADP-Mg2+ binds to the active site of NiRAN, but not with adenine Part of the formation of specific interactions (17). In our study, the type of nucleotide used in the NMPylation reaction has no differential effect on the activity of the mutant protein (SI Appendix, Figure S3), indicating that none of these residues are closely related to the binding of a specific nucleobase. The structural basis and potential biological significance of the different NTP co-substrate preferences observed in the NiRAN domains of coronaviruses and arterial viruses remain to be studied; they may be true or may be due to the limitations of their respective studies. At present, it cannot be ruled out that the potential NMPylator activity of the arterial virus NiRAN domain (compared to the previously characterized self-NMPylation activity) has a different co-substrate preference, taking into account that the similarity between the arterial and the coronavirus NiRAN domain is at its limit. Sequence-based Compare (16). Compared with the pseudokinase SelO, which uses Mg2+ as a cofactor, the activity of coronavirus and arterial virus NiRAN is dependent on Mn2+ (16) (Figure 3C and SI appendix, Figure S1). The Mn2+ dependence and obvious preference for UTP is an unusual feature of protein NMPylators, and has only recently been confirmed in the YdiU protein of Salmonella typhimurium, which catalyzes the strict Mn2+-dependent protein chaperone UMPylation to protect cells from stress induction Cell ATP pool (33).

The recently described structural similarity between the coronavirus NiRAN domain and cellular protein kinases (17, 19) provides additional support for NiRAN’s ability to covalently link NMP to other proteins we have reported in this study. We focused our search for possible NiRAN targets on the proteins encoded by HCoV-229E ORF1a, which are known to directly or indirectly assist RTC’s ORF1b-encoded replicase (12, 35). Our experiments provide conclusive evidence for the effective and specific NMPylation of nsp9 (Figure 2). If the target protein is used in a molar excess that is 8 to 10 times higher than that of the enzyme (nsp12), it is confirmed that nsp9 is completely (mono)NMPized (Figure 4). We concluded that the interaction between nsp12 and nsp9 is short-lived and will not form a stable complex with nsp9 (in the absence of other RTC subunits). This conclusion is supported by protein interaction studies on the SARS-CoV proteome (35). MS analysis identified the primary amine of the N-terminal residue of nsp9 as the NMPylation site (SI appendix, Figure S5). The formation of the phosphoramidate bond and the N-terminal amino group distinguishes the NiRAN-mediated NMPylation activity from the Pseudomonas syringae SelO-mediated AMPylation reaction, which catalyzes the formation of O-linked AMP at Ser, Thr, or Tyr residues Peptide adduct (22), and S. typhimurium YdiU forms O-linked (with Tyr) and N-linked (with His) peptide-UMP adducts. The limited information available on the SelO family of proteins indicates that members of this large protein family differ greatly in the formation of peptide-NMP adducts. This is an interesting observation that deserves further study.

The data obtained in this study led us to hypothesize that the NMPylation of nsp9 requires a free N-terminus. In the context of viral replication, this will be provided by the proteolytic cleavage of the nsp8|nsp9 processing site in the replicase polyprotein pp1a mediated by Mpro and pp1ab. In most coronaviruses, the difference between this specific site (VKLQ|NNEI in HCoV-229E) and all other coronavirus Mpro cleavage sites is Asn (rather than another small residue, such as Ala, Ser Or Gly) occupy P1â? ? ? Location (36). The peptide cleavage data obtained in the early studies showed that the cleavage efficiency of the nsp8|nsp9 site was lower than that of other sites, indicating that 1) this specific site may have a regulatory role in the timely coordinated processing of the C-terminal pp1a region, or 2) a The role of the special conserved nsp9 N-terminus in virus replication (37). Our data (Figure 5A) showed that the recombinant form of nsp9 carrying the real N-terminal sequence was effectively NMPized by nsp12. The N-terminal flanking sequence was removed by factor Xa (nsp9-His6; SI appendix, Table S1) or Mpro-mediated cleavage (nsp7-11-His6; Figure 5A and SI appendix, Table S1). Importantly, the uncut nsp9-containing precursor nsp7-11-His6 showed resistance to NMPylation of nsp12, which is consistent with our data, indicating that the nsp9-NMP adduct is formed through the N-terminal primary amine ( SI appendix, Figure S5). To gain a deeper understanding of NiRAN substrate specificity, we then focused on the adjacent N-terminal residues of nsp9. In the absence of other proteins, they are structurally flexible, preventing them from being detected in the unlabeled form of nsp9 (26 28, 38), indicating their limited natural variation This is due to the important sequence-specific (not secondary structure related) function of the nsp9 N-terminal fragment. Ala substitutions of conserved residues in this region (Figures 5C and D and SI appendix, Figure S8) reveal that N3826 is essential for nsp9 NMPylation in vitro, while N3825A and E3827A substitutions lead to a decrease in NMPylation, while M3829A and P3830A substitutions do not. Obviously affect nsp9 NMPylation. Although the substitution of N-terminal Asn (N3825A, N3825S) has only a moderate effect on nsp9 NMPylation and virus replication in cell culture (Figure 5C and D), the deletion of an Asn residue sequence from the N-terminal 3825-NN dipeptide proved to be It is lethal to viruses, indicating that one Asn residue is required before another residue at the N-terminus, preferably Asn, although it seems that substitution of similar residues can be partially tolerated (Figure 5B, C, and D). We conclude that the 3825-NN dipeptide, especially the conserved and essential N3826 residue within the coronavirus range (SI appendix, Figure S6), ensures the correct binding and orientation of the nsp9 N-terminus in the active site of NiRAN.

Substituting Ala (E3827A) for the conserved Glu of all subfamilies retains nsp9 NMPylation in vitro but is lethal to viruses in cell culture (Figure 5C and D), indicating the additional function of this residue, for example, in key interactions (NMPylated or unmodified) nsp9 N-terminus and other factors involved in virus replication. Nsp9 mutations did not affect the proteolytic process of nsp9 or any adjacent nsps (39) (SI Appendix, Figure S9), indicating that the lethal phenotypes of several nsp9 mutations observed were not caused by the dysregulation of the C proteolytic process-terminal pp1a area.

The above data provides evidence that after Mpro-mediated treatment of the nsp8|9 cleavage site in pp1a/pp1ab, the N-terminus of nsp9 can be UMPylated (or partially modified with another NMP). In addition, the excellent conservation of the N-terminus of nsp9 (including the singular and invariant Asn residues in the coronavirus family) and the reverse genetics data obtained in this study (Figures 3E and 5D) led us to conclude that the described nsp9 NMPylation is biologically related and essential for coronavirus replication. The functional consequences of this modification remain to be studied, for example, regarding the previously described (non-specific) nsp9 (unmodified form) RNA binding activity (2628). N-terminal NMPylation may also affect the interaction of nsp9 with protein or RNA substrates or the formation of different four-level assemblies. These have been observed in structural studies and have been confirmed to be functionally related to coronavirus replication, although especially in the absence of In the case of this modification (26- ââ29, 40).

Although the target specificity of the coronavirus NiRAN domain still needs to be characterized in more detail, our data shows that the protein target specificity of the coronavirus NiRAN domain is very narrow. Although the conservation of key active site residues (8, 16) in the NiRAN domain of all nidovirus families strongly supports the activity of the conserved NMPylator these proteins, the identity of the substrate binding pocket residues of this domain Conservation and conservation remain to be characterized, and may differ between different families of Nidovirales purposes. Similarly, the relevant targets of other nested viruses have yet to be determined. They may be remote orthologs of nsp9 or other proteins, because the sequences outside the five replicase domains that are generally conserved in nested viruses are less conserved (8), including the genome array between Mpro and NiRAN, Among them, nsp9 is located in the coronavirus.

In addition, we cannot currently rule out the possibility that the NiRAN domain has additional (including cellular) targets. In this case, it is worth mentioning that the bacterial homologues in this emerging protein NMPylators (NMPylators) (30, 31) seem to have “master regulators”? NMP modulates a variety of cellular proteins to regulate or eliminate their downstream activities, thereby playing a role in a variety of biological processes, such as cellular stress response and redox homeostasis (22, 33).

In this study (Figures 2 and 4 and SI Appendix, Figures S3 and S5), we were able to prove that nsp12 transferred the UMP (NMP) part to a single (conserved) position in nsp9, while other proteins were not modified in the used Under the conditions, well-defined (rather than loose) substrate specificity is supported. Consistent with this, compared with N-terminal nsp9 NMPylation, nsp12′s own NMPylation activity is very low, its detection requires longer autoradiography exposure time, and a 10-fold increase in nsp12 concentration is used. In addition, our MS analysis failed to provide evidence for NMPylation of nsp12, which suggests that NiRAN domain self-NMPylation is (at best) a secondary activity. However, it should be noted that other studies have provided preliminary evidence that the self-AMPylation status of bacterial NMPylator may control their NMPylation activity on other protein substrates (22, 33). Therefore, more research is needed to investigate the possible functional effects of self-NMPylation activities reported for EAV nsp9 (16) and coronavirus nsp12 (this study), including the proposed chaperone-like effect on the folding of the C-terminal RdRp domain (16) ).

Previously, several hypotheses regarding the possible downstream functions of the nidoviral NiRAN domain have been considered, including RNA ligase, RNA -capped guanylate transferase and protein priming activity (16), but none of them are compatible with the available downstream functions. The information obtained in the following positions is exactly the same time without making additional assumptions. The data obtained in this study is most consistent with (but cannot prove) that the NiRAN domain is involved in the initiation of protein-induced RNA synthesis. It was previously believed that the function of the NiRAN domain in 5?? ²-RNA capping or RNA ligation reactions is not affected by these and Support of other data. Therefore, for example, the active site of NiRAN is considered to involve the conserved Asp as a general base (D252 in Pseudomonas syringae SelO; D4271 in HCoV-229E pp1ab; D208 in SARS-CoV-2 nsp12) (SI Appendix ,figure 2). S2) (17, 22, 33), while the catalysis in the ATP-dependent RNA ligase and RNA capping enzyme is carried out by the covalent enzyme-(lysyl-N)-NMP intermediate, which involves a non- Changed Lys residue (41). In addition, the remarkable sequence-based specificity of the coronavirus NiRAN for conserved protein targets and the relaxed specificity for NTP co-substrates (prefers UTP) opposes NiRAN-mediated capping enzyme or RNA ligase-like functions.

Obviously, a lot of extra work is needed to verify and, if proven, elaborate on the possible role of nsp9-UMP (nsp9-NMP) in protein-induced RNA synthesis, which will connect several interesting but (so far) reports previously reported. Isolated observations. For example, it has been determined that the  end of the negative-strand RNA of coronavirus starts with an oligo(U) strand (42, 43). This observation is consistent with the idea that the synthesis of negative-strand RNA is initiated by binding the UMPylated form of nsp9 to the poly(A) tail ( triggers), which may be promoted by its RNA binding The activity and/or interaction with another RTC protein. The UMP portion provided by nsp9 can then be used as a “primer” for nsp7/8/nsp12-mediated oligouridylation, using the 3??²-poly(A) tail in the genomic RNA or Another oligo (A)-containing sequence serves as a template, similar to the mechanism established for the picornavirus VPg protein (44). What if the proposal is “non-normative”? ? ? ? The initiation of the (protein-induced) negative-strand RNA synthesis provides a link to the observations, indicating that the coronavirus negative-strand RNA has UMP (instead of UTP) at its  end (42), which is considered to indicate that the nucleic acid Dicer cleaves the  end phosphorylated by an unknown uridine-specific endonuclease. If confirmed, this nucleic acid hydrolytic activity can help release the oligomeric UMPylated form of nsp9 from the 5 ² end of the nascent negative strand. The possible role of nsp9 in protein priming is also consistent with previous reverse genetics studies, which have shown that nsp9 (and nsp8) interact critically and specifically with the conserved cis-acting RNA element near the 3 end of the coronavirus genome. 45). According to this report, these previous observations can now be re-examined and expanded through further research.

In summary, our data determined the specific activity of a proprietary nested virus enzyme tag linked to RdRp at the N-terminus. In coronavirus, this newly discovered NiRAN-mediated UMPylator/NMPylator activity is used to rely on Mn2+ and adjacent Asn residues and cause the formation of (low-energy) phosphoramidate bonds with the N-terminal primary amine. Through Mpro-mediated cleavage at the nsp8|9 cleavage site, the nsp9 target can be used for NMPylation, indicating the functional coupling between the protease and the NiRAN domain, which extends to RdRp. The conservation of key residues in the nsp12 NiRAN active site and the nsp9 target, combined with data obtained from two coronaviruses including SARS-CoV-2, provides strong evidence that nsp9 NMPylation is a coronavirus Conservative features are also a key step in virus replication. The available data lead us to conclude that the specific role of NMPylated form of nsp9 in protein-induced RNA synthesis is a reasonable scenario for coronavirus and other nested viruses, and NiRAN may also target other unidentified proteins. Regulate the virus.  Host interaction. If confirmed, the involvement of protein primers in viral RNA synthesis will increase the sequence affinity of the Mpro/3CLpro and RdRp domains between the previously detected coronavirus and picornavirus-like supergroup (9), which have now been unified in the recently established Pisonivirites (46) in the category.

Our data also shows that the basic, selective and conservative enzyme activities identified in this study can be used as targets for antiviral drugs. Compounds that interfere with the binding (and subsequent modification) of the conserved nsp9 N-terminus in the active site of NiRAN can be developed into effective and versatile antiviral drugs, suitable for the treatment of animal and human coronaviruses from different (sub)genus Infections, including SARS-CoV-2 and Middle East Respiratory Syndrome Coronavirus.

The coding sequence of the coronavirus protein produced in this study was amplified by RT-PCR using RNA isolated from Huh-7 infected with HCoV-229E or Vero E6 infected with SARS-CoV-2, and inserted using standard cloning procedures. pMAL-c2 (New England Biological Laboratory) or pASK3-Ub-CHis6 (47) expression vector (SI Appendix, Tables S1 and S2). Single codon substitutions were introduced by PCR-based site-directed mutagenesis (48). To produce the MBP fusion protein, E. coli TB1 cells were transformed with the appropriate pMAL-c2 plasmid construct (SI appendix, Table S1). The fusion protein was purified by amylose affinity chromatography and cleaved with factor Xa. Subsequently, the C-terminal His6-tagged protein was purified by Ni-immobilized metal affinity chromatography (Ni-IMAC) as previously described (49). To produce the ubiquitin fusion protein, E. coli TB1 cells used the appropriate pASK3-Ub-CHis6 plasmid construct (SI Appendix, Tables S1 and S2) and pCGI plasmid DNA encoding ubiquitin-specific C-terminal hydrolase 1 (Ubp1). Transformation (47). The C-terminal His6-tagged coronavirus protein was purified as previously described (50).

The self-NMPylation test of HCoV-229E nsp12-His6 was performed as described in EAV nsp9 (16). In short, nsp12-His6 (0.5 µM) contains 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)-KOH, pH 8.0, 5 mM dithiothreitol ( DTT), 6 mM MnCl2, 25 µM buffer, incubate the specified NTP and 0.17 µM matched [α32-P]NTP (3,000 Ci/mmol; Hartmann Analytic) at 30 °C for 30 minutes. In all other (standard) NMPylation assays of nsp12-mediated nsp9 NMPylation, the reaction conditions are adjusted as follows: nsp12-His6 (0.05 µM) and nsp9-His6 (4 µM) in the presence of 50 mM HEPES-KOH (pH 8.0), 5 mM DTT, 1 mM MnCl2, 25 µM indicated NTP, and 0.17 µM matched [α32-P]NTP. After incubating for 10 minutes at 30°C, the reaction sample was mixed with SDS-PAGE sample buffer: 62.5 mM tris(hydroxymethyl)aminomethane HCl (pH 6.8), 100 mM DTT, 2.5% SDS, 10% glycerol and 0.005% bromophenol blue. The protein was denatured by heating at 90 °C for 5 minutes and separated by 12% SDS-PAGE. The gel is fixed and stained with Coomassie Brilliant Blue solution (40% methanol, 10% acetic acid, 0.05% Coomassie Brilliant Blue R-250), decolorized, and exposed to a phosphorescent imaging screen for 20 hours (to detect nsp12 from NMPylation) or (maximum) 2 Hours (to assess nsp9 NMPylation). A Typhoon 9200 imager (GE Healthcare) was used to scan the screen and ImageJ was used to analyze the signal intensity.

For MS analysis, 1 µM nsp12-His6 and 10 µM nsp9 (without hexahistidine tag) were used in NMPylation analysis (SI appendix, Table S1) and the increased concentration of 500 µM UTP and GTP were used. Depending on their concentration and expected protein quality, a Waters ACQUITY H-Class HPLC system equipped with a MassPrep column (Waters) was used to desalt 1 to 10 µL of buffered protein solutions online. The desalted protein is eluted into the electrospray ion source of Synapt G2Si mass spectrometer (Waters) through the following gradient of buffer A (water/0.05% formic acid) and buffer B (acetonitrile/0.045% formic acid), and the column temperature is 60 ° C and a flow rate of 0.1 mL/min: elution isocratically with 5% A for 2 minutes, then a linear gradient to 95% B within 8 minutes, and maintain 95% B for another 4 minutes.

Positive ions with a mass range of 500 to 5000 m/z are detected. Glu-fibrinopeptide B is measured every 45 seconds for automatic mass drift correction. Use MassLynx instrument software with MaxEnt1 extension to deconvolve the average spectrum after deducting the baseline and smoothing.

UMPylated HCoV-229E nsp9 was digested by adding sequencing-grade modified trypsin (Serva) and incubated overnight at 37 °C. A Chromabond C18WP spin column (part number 730522; Macherey-Nagel) was used to desalt and concentrate the peptides. Finally, the peptide was dissolved in 25 µL of water, which contained 5% acetonitrile and 0.1% formic acid.

The samples were analyzed by MS using an Orbitrap Velos Pro mass spectrometer (Thermo Scientific). The ultimate nanoâ HPLC system (Dionex), equipped with a custom end-mounted 50 cm?? 75 μm C18 RP column packed with 2.4 μm magnetic beads (Dr. Albin Maisch High Performance LC GmbH) Connect to the mass spectrometer online through the Proxeon nanospray source; inject 6 µL of trypsin digestion solution into a 300 µm inner diameter ×?? 1 cm C18 PepMap pre-concentration column (Thermo Scientific). Using water/0.05% formic acid as the solvent, the sample was automatically trapped and desalinated at a flow rate of 6 µL/min.

The following gradients of water/0.05% formic acid (solvent A) and 80% acetonitrile/0.045% formic acid (solvent B) were used to achieve the separation of tryptic peptides at a flow rate of 300 nL/min: 4% B for 5 minutes, then 30 A linear gradient to 45% B within minutes, and a linear increase to 95% solvent B within 5 minutes. Connect the chromatographic column to a stainless steel nano-emitter (Proxeon), and spray the eluent directly to the heated capillary of the mass spectrometer using a potential of 2,300 V. The survey scan with a resolution of 60,000 in the Orbitrap mass analyzer is associated with at least three data MS/MS scans, dynamically excluded for 30 seconds, using linear ion trap collision induced dissociation or higher energy collision dissociation combined with orbitrap detection , The resolution is 7,500.

Post time: Aug-03-2021