Reports three bat sarbecoviruses from northern Laos (BANAL-52, -103, -236), each >95% identical to SARS-CoV-2 genome-wide and closer than RaTG13 in the receptor-binding domain specifically; BANAL-236 binds human ACE2 with affinity comparable to or exceeding early SARS-CoV-2 isolates, and its spike lacks a furin cleavage site — sharpening the question of when and by what route the FCS was acquired. relevance_note: currently the closest known wild-bat relatives to SARS-CoV-2; used on both sides to argue what a “natural” precursor genome looks like and whether the FCS’s absence there is informative.
Observations
O-68 - Three new Lao bat sarbecoviruses; BANAL-52 is 96.8% identical to SARS-CoV-2, exceeding RaTG13
Three new sarbecoviruses were isolated from cave bats in northern Laos - BANAL-52 (Rhinolophus malayanus), BANAL-103 (R. pusillus) and BANAL-236 (R. marshalli); BANAL-52 is 96.8% identical to SARS-CoV-2 across the whole genome, higher than RaTG13’s 96.1%, making it the closest known full-genome relative.
Link to original
O-69 - BANAL RBDs match 15-16 of 17 hACE2-contact residues vs only 11-17 for RaTG13
Of the 17 RBD residues that contact human ACE2, 16/17 are conserved in BANAL-52/103 (one mismatch, H498Q) and 15/17 in BANAL-236 (K493Q, H498Q), versus only 13/17 for the Cambodian R. shameli virus and 11/17 for RaTG13 - the BANAL RBDs are far closer to SARS-CoV-2 at the receptor interface than RaTG13.
Link to original
O-70 - BANAL RBDs bind human ACE2 ~3x more strongly than early SARS-CoV-2; BANAL-236 RBD-hACE2 crystal near-identical
By biolayer interferometry, the BANAL-52/103 and BANAL-236 RBDs bind human ACE2 with a dissociation constant three times lower (higher affinity) than the early Wuhan SARS-CoV-2 RBD; the BANAL-236 RBD-hACE2 crystal structure (PDB 7PKI, 2.9 A) is essentially identical to SARS-CoV-2 (Ca RMSD 0.36 A).
Link to original
O-71 - A wild, unmodified BANAL-236 enters and replicates in human cells via hACE2
BANAL-236 spike-pseudotyped lentivirus enters hACE2-expressing HEK-293T cells (but not ACE2-negative controls) and is blocked by SARS-CoV-2-neutralizing sera, and live BANAL-236 virus replicates in human Calu-3 and Caco-2 cells expressing endogenous ACE2 - a wild bat virus infecting human cells without adaptation.
Link to original
O-72 - None of the BANAL viruses has a furin cleavage site
None of the Laotian BANAL sarbecoviruses carries a furin cleavage site at the spike S1/S2 junction; the internal fusion peptide is conserved.
Link to original
O-73 - SARS-CoV-2 is a recombinant mosaic of -5 wild bat sarbecoviruses, its RBD fragment nesting in the BANAL clade
A GARD recombination analysis found 14 breakpoints (15 fragments); SARS-CoV-2’s genome is a mosaic contributed to by more than five bat sarbecoviruses (RmYN02, RpYN06, RaTG13, BANAL-52, BANAL-103), with a breakpoint 7 nt upstream of the RBD and the RBD-containing fragment grouping with the wild BANAL-52/103/236 clade; no pangolin coronavirus was immediately associated with a recombination event at the origin of SARS-CoV-2.
Link to original
Hypotheses
H-36 - Wild bat sarbecoviruses able to infect humans via hACE2 without adaptation circulate in the Indochinese peninsula
SARS-CoV-2-like bat sarbecoviruses that bind human ACE2 and enter/replicate in human cells with no laboratory adaptation circulate naturally in Rhinolophus bats of the Indochinese peninsula (northern Laos).
Link to original
H-37 - SARS-CoV-2's human-ACE2-binding RBD and backbone arose by natural recombination among wild bat viruses
SARS-CoV-2’s genome apart from the furin cleavage site - including its human-ACE2-optimized RBD - is explained by natural recombination among wild bat sarbecoviruses of this Indochinese pool, requiring no laboratory engineering or adaptation of the RBD.
Link to original
H-38 - SARS-CoV-2's furin cleavage site was acquired separately, by a route not represented in the closest bat relatives
Because the closest wild relatives all lack a furin cleavage site, SARS-CoV-2 acquired its FCS by a distinct later event whose route (natural or otherwise) is not accounted for by these bat genomes.
Link to original
Arguments
A-65 - A wild bat RBD binding hACE2 as well as SARS-CoV-2 shows no lab adaptation is needed for human-ACE2 use
Because an unmodified wild bat virus (BANAL-236) binds human ACE2 with affinity equal to or greater than early SARS-CoV-2 and enters/replicates in human cells, high-affinity human-ACE2 tropism does not require passage or engineering, undercutting the premise that SARS-CoV-2’s RBD needed lab adaptation to use hACE2.
Reasoning
A central plank of lab-adaptation arguments is that a bat virus would need to be adapted (by serial passage or engineering) before binding human ACE2 efficiently. Temmam et al. present a counterexample from nature: BANAL-236, isolated from a wild bat with no laboratory manipulation, has an RBD whose 17-residue ACE2 interface differs from SARS-CoV-2 by only two residues, binds hACE2 with a Kd three times lower than early SARS-CoV-2, and mediates entry and replication in human cell lines through hACE2 (specifically neutralised by anti-SARS-CoV-2 sera). Since a naturally occurring, unadapted virus already achieves human-cell infectivity comparable to the pandemic strain, that phenotype provides no evidence for - and removes the necessity of - a lab-adaptation step for the RBD.
Validity verdict — approved (checked)
This is a clean necessity-refutation (modus tollens on a “must be adapted” claim). The premise being attacked is “a bat sarbecovirus RBD would need serial passage or engineering before it can bind human ACE2 with high affinity and enter human cells.” Premises granted: BANAL-236 is a wild, unmanipulated isolate; its RBD differs from SARS-CoV-2 at only 2 of 17 hACE2-contact residues, binds hACE2 with a ~3x lower Kd, and mediates hACE2-dependent entry/replication in human cell lines. A single genuine natural instance of the phenotype falsifies a universal necessity claim, so “high-affinity hACE2 tropism does not require lab adaptation” follows validly. Probed undercutting defeaters: (a) BANAL-236 is not identical to SARS-CoV-2’s RBD — irrelevant, since the refuted claim is about whether any unadapted bat RBD can achieve high hACE2 affinity, which BANAL-236 settles; (b) BANAL-236 lacks the furin cleavage site — irrelevant, because the statement scopes the conclusion narrowly to RBD/hACE2 use, not to the FCS. No defeater survives against the scoped conclusion. Note the conclusion is correctly modest — it removes a necessity/adaptation-required premise; it does not affirmatively establish natural origin (absence of necessity ≠ evidence of natural spillover), and the statement stays within that bound. Conditional on premises, the inference holds. Approved; checked via the reconstruction above.
Link to original
A-66 - Mosaic recombination placing SARS-CoV-2's RBD in the wild bat clade supports a natural recombinant origin
The GARD-detected mosaic structure - with SARS-CoV-2’s RBD-region fragment nesting within the wild BANAL-52/103/236 clade and more than five identified bat donors - shows SARS-CoV-2’s components (apart from the FCS) are present and recombining in wild bats, supporting a natural recombinant origin of the backbone-plus-RBD over a constructed chimera.
Reasoning
Recombination detection across sarbecovirus genomes resolves SARS-CoV-2 into 15 fragments whose closest donors are wild bat viruses; notably a breakpoint 7 nt upstream of the RBD separates a downstream fragment (RBD plus start of S2) that groups with the Laotian BANAL-52/103/236 clade. Thus the very region argued by lab-origin proponents to be a grafted ‘pangolin-like’ RBD instead maps onto a highly supported wild bat clade, and the rest of the genome likewise finds wild bat donors. Because the mosaic can be assembled entirely from naturally circulating, sympatric bat viruses that are known to recombine in shared caves, an ordinary natural recombination history is sufficient to produce SARS-CoV-2’s backbone and RBD without invoking deliberate construction; no pangolin virus was needed as an immediate recombination donor.
Validity verdict — approved (checked)
Reconstruction. Premises: (i) GARD resolves SARS-CoV-2 into ~15 fragments whose closest donors are wild bat viruses; (ii) the RBD-region fragment nests within the sympatric BANAL-52/103/236 clade; (iii) these bats co-circulate and are known to recombine in shared caves. Load-bearing step: from “every component except the FCS maps onto co-circulating, recombining wild bat clades” to “a natural recombination pathway is demonstrably sufficient, which supports a natural recombinant origin of backbone+RBD over the constructed (pangolin-graft) chimera.”
Evaluation. Conditional on the premises the step holds: exhibiting a sufficient wholly-natural assembly route from sympatric recombining viruses raises P(natural) and specifically removes the RBD from the list of features needing a special (pangolin-graft) construction story. The conclusion is appropriately comparative and hedged (“supporting…over”), not a proof. Probed undercutting defeater: “a lab could equally have constructed a chimera from those same wild bat viruses.” This does not survive against the stated conclusion — it does not deny that a sufficient natural pathway now exists, which is exactly what a “supports natural over constructed” (evidential) claim requires; it would only bear on the magnitude of the tilt, priced at step 8. No defeater breaks the direction. Approved; the chain is elementary and traced here, hence checked.
Link to original
A-67 - The furin cleavage site is the one feature not bridged by the closest wild relatives
Since none of the closest wild relatives carries a furin cleavage site while sharing nearly everything else, the FCS is a discriminating feature whose acquisition is a separate event not explained by these bat genomes, leaving its route to be determined.
Reasoning
The BANAL viruses match SARS-CoV-2 closely across the genome and at the ACE2 interface, yet uniformly lack the polybasic furin cleavage site at S1/S2. Therefore the wild-relative pool can account for the backbone and the human-ACE2-binding RBD but not the FCS. Logically the FCS must have been gained after SARS-CoV-2’s lineage diverged from these bats, as an independent event. This sharpens rather than resolves the origins question: the closest natural genomes remove the RBD from the list of features requiring special explanation but leave the FCS’s mode of acquisition (recombination with an as-yet-unsampled virus, in-host insertion, or otherwise) open.
Validity verdict — approved (checked)
Reconstruction. Premise: the BANAL viruses match SARS-CoV-2 closely across the genome and at the ACE2 interface yet uniformly lack the polybasic FCS. Load-bearing step: from “closest relatives lack the FCS while sharing nearly everything else” to “the FCS is unaccounted for by these genomes and so was acquired as a separate event, route undetermined.”
Evaluation. Near-deductive and holds conditional on the premise: if the closest sampled relatives lack the feature, those genomes cannot supply it, so its acquisition is a distinct event in the lineage. Probed defeater: “the FCS could be ancestral and lost in the BANAL lineages rather than gained in SARS-CoV-2.” This does not defeat the stated, deliberately modest conclusion — under a loss scenario these particular genomes still fail to explain the FCS and its route still remains to be determined, which is all the argument asserts (it stops short of claiming natural vs lab). No defeater survives against the conclusion as stated. Traced here → checked.
Link to original