Drought bottlenecks strip Enterobacteriaceae from wild insect guts, and compositional recovery outpaces functional proof

Question: How do wild insect populations' gut microbiomes respond to drought-induced host population bottlenecks, and how quickly do these communities recover in subsequent generations?

Answer

The retrieved literature supports a narrow, provocative answer: drought-induced host bottlenecks cause severe but taxon-selective and apparently transient microbiome disruption, with compositional recovery measurable within a few host generations — but the evidence base for this is a single wild system, and "recovery" there means resemblance of community structure, not demonstrated restoration of strain identity, function, or host benefit. In the Åland Glanville fritillary metapopulation, the 2018 extreme drought that bottlenecked the butterfly host was associated with severe microbiota disruption and loss of key Enterobacteriaceae, yet the bacterial community appeared to return toward pre-drought structure in subsequent years, which the authors interpret as resilience to acute climatic perturbation [1]. Taken together with work showing that insect microbiome composition can turn over rapidly while its phenotypic legacy persists for multiple host generations [7], and that microbiome composition is often governed by environmental reservoirs and colonization opportunity rather than faithful maternal transfer [12], the defensible position is that drought bottlenecks produce fast compositional rebounds that are likely driven by re-acquisition from the environment and by host filtering, and that these rebounds should not be read as functional or genetic restoration of the pre-bottleneck symbiosis.

The only direct test of the question shows severe, taxon-selective disruption followed by apparent rebound

The 28-year field survey of Melitaea cinxia on the Åland Islands is the one retrieved study that explicitly asks how a wild insect's microbiome responds to a drought-triggered host bottleneck: the 2018 extreme drought was associated with severe disruption of the larval microbiota and the loss of key Enterobacteriaceae taxa [1]. The same study reports that the bacterial community subsequently appeared to recover toward its pre-drought structure, which the authors frame as resilience to acute climatic perturbation [1]. Crucially, the disruption was not community-wide: M. cinxia harbours a set of stable resident bacteria including Pseudomonas, Telluria, and Enterobacteriaceae, and it was the Enterobacteriaceae that were lost [1].

The bottleneck perturbation is mechanistically distinct from the slow warming trend it sits on

The same dataset separates two signals that are easy to conflate: a gradual shift in the associated bacterial community across three decades of rising temperature and declining precipitation, and an abrupt disruption coincident with the drought and its host bottlenecks [1]. The gradual shift was not unidirectional, with the dominant Telluria and Pseudomonas showing opposing responses to the same environmental trends, which means "climate response" cannot be summarised as a single directional trajectory [1]. This matters for interpreting bottlenecks: an acute demographic crash and a chronic environmental trend can push the same community in different directions and on different timescales.

Taxon-selective loss, not wholesale collapse, is the consistent thermal-stress signature across insects

Where the core community is conserved, thermal stress tends to leave it intact while reshuffling peripheral taxa: in Bombus impatiens, core genera such as Bombilactobacillus and Snodgrassila were unaffected by heat and infection treatments, whereas Apilactobacillus kunkeei rose and Citrobacter freundii dominated in hot infected bees [2]. Bacterial communities in blue orchard bee pollen provisions were resilient in composition, diversity, and abundance across a simulated heatwave and recovery period [9]. House fly microbiomes differed between warm and cool rearing temperatures without detectable dysbiosis within that range, and temperature outweighed host genotype [6]. Host identity, not temperature or geography, dominated the structure of adult butterfly microbiomes sampled across two continents, indicating strong host filtering that could buffer communities against environmental shocks [13].

Recovery is most plausibly re-colonisation from environmental reservoirs, not restored transmission

Recovery speed in the fritillary is best explained by re-acquisition rather than by faithful inheritance of a damaged vertically transmitted community, because insect gut communities with weak maternal transmission drift substantially: in Tribolium castaneum, five years of sampling revealed dramatic fluctuations between a few dominant taxa and hundreds of rare taxa, attributed to a lack of maternal microbial transmission and to microbial load fluctuations in the external reservoir, which itself varied with host population density and stage structure [12]. Similar horizontal acquisition dominates other wild and semi-wild systems: Drosophila suzukii acquires core bacterial and niche-specific yeast components from host fruits [14], and phylogenetic host signal rather than environment structures adult butterfly microbiota [13]. If the reservoir, not the host lineage, supplies the recovering taxa, then a bottleneck that thins the host population also thins the faecal and resource-borne inoculum — a coupling that density-dependent load variation in flour beetles makes explicit [12].

Community-level resilience is partly predictable from baseline network properties

Baseline microbiome network and functional properties predict community-level structural and functional stability under drought, warming, and salinity across roughly 29,000 microbiomes and over 440 matched network pairs, with keystone microbes high in connectivity, functional redundancy, and metabolic flexibility disproportionately likely to survive stress [10]. This offers a testable explanation for why the fritillary community rebounded while retaining a stable resident set: resilience may be a property of the pre-disturbance architecture rather than of the disturbance magnitude [10] [1].

Conflicts and gaps

The central conflict is between resilience and durable legacy, and the retrieved papers support both. Apparent compositional recovery coexists with multi-generational carryover of microbiome effects: duckweed hosts co-introduced with non-local microbes suffered reduced performance that persisted for multiple host generations despite rapid turnover in microbiome composition during colonisation [7]. In tropical tree seedlings, bacterial and fungal leaf communities did not converge on control composition during recovery, and no recovery network mirrored the control network regardless of drought duration [5]. At the same time, common bean root and rhizosphere bacterial microbiomes showed neither a drought response nor a notable cross-generational drought legacy in one genotype, arguing that some host-microbiome systems are simply resistant [8]. In insects specifically, serial rearing on a novel substrate produced divergent, stochastic, lineage-specific gut communities across four generations with low narrow-sense heritability and the strongest compositional shifts late in the series, implying that post-bottleneck trajectories may be unpredictable rather than convergent [3].

Three gaps are decisive. First, the recovery claim rests on community structure from 16S metabarcoding and metagenomics, so strain-level replacement by functionally divergent taxa cannot be excluded, and resolution itself matters: full-length 16S sequencing revealed cohort-level compositional differences in medfly guts that V4 region sequencing did not detect [1] [4]. Second, no retrieved study links drought-bottleneck microbiome disruption to host fitness in a wild insect, so whether the Enterobacteriaceae loss mattered is unknown [1]. Third, the possibility that host bottlenecks erode the genetic variation underpinning microbiome filtering is untested in this context, even though microbiome α-diversity metrics show high heritability across Drosophila Genetic Reference Panel lines [11].

What would change the answer

Strain-resolved metagenomics on archived and freshly sampled Åland M. cinxia larvae spanning pre-2018, drought, and post-drought years, with metagenome-assembled genomes tracked individually, would determine whether "recovery toward pre-drought structure" means return of the same strains or substitution by different ones. Absolute quantification by qPCR alongside relative abundances would separate genuine loss and regain from proportional shifts. Gnotobiotic reconstitution of pre-drought versus post-drought larval communities into axenic hosts, with measured correlates of host performance, would test functional equivalence directly. A controlled demographic bottleneck experiment in a tractable insect with a known external reservoir — Tribolium populations with manipulated census size and flour inoculum, sequenced longitudinally over ten or more generations — would yield an actual recovery half-life rather than a qualitative "subsequent years" [12]. Finally, coupling host genome resequencing to microbiome time series across bottlenecked versus unmanipulated patches would test whether bottleneck-reduced host genetic variation, rather than microbial supply, sets the recovery trajectory.

References

[1] Sun, L., van Dis, N. E., Davrinche, A., Saastamoinen, M., Ekroos, J., Duplouy, A.. "The microbiota of a wild butterfly species over three decades of climate change". bioRxiv. 2026-10-19T00:00:00.000Z. URL: https://doi.org/10.64898/2026.06.08.730914

[2] Van Wyk, J., Beirne, L., Bowder, S., Campbell, E., Disharoon, M., Dreyer, M., Frolichstein-Appel, N., Gill, A., Jones-Ducharme, A., Kinkaid, A., Broussard Korr, G., McCabe, M., McDowell, E., Perez, F., Villarreal-Mentz, R., Johnston, J.. "Hot and Bothered, Bees Gut Microbiome Shifts Under Thermal Stress and Pathogen Infection.". bioRxiv. 2025-11-07T00:00:00.000Z. URL: https://doi.org/10.1101/2025.11.06.687019

[3] Silvaraju, S., Lim, A. C. H., Tang, Y. J., Kittelmann, S., Puniamoorthy, N.. "Transgenerational dynamics of gut microbiota in black soldier fly larvae (Hermetia illucens) reared on a novel substrate". bioRxiv. 2025-05-28T00:00:00.000Z. URL: https://doi.org/10.1101/2025.05.28.656550

[4] Mason, C., Hansen, T., Newell, N., Geib, S. M.. "Cryptic Insect Microbiome Compositions Unveiled with Full-Length 16S Sequencing". bioRxiv. 2026-01-23T00:00:00.000Z. URL: https://doi.org/10.64898/2026.01.22.701149

[5] BOISSEAUX, M., Goret, J.-Y., Burban, B., Troispoux, V., Bordes, A., Cazal, J., Cazal, S.-O., Coste, S., Stahl, C., Schimann, H.. "Longer projected droughts will impair the recovery of tropical seedlings and their leaf microbiota". bioRxiv. 2026-08-27T00:00:00.000Z. URL: https://doi.org/10.64898/2026.08.26.746266

[6] Sanjee, S. A., Adhikari, K., Meisel, R.. "Genotype and environmental effects shape the house fly microbiome (Musca domestica)". bioRxiv. 2026-04-07T00:00:00.000Z. URL: https://doi.org/10.64898/2026.04.06.716741

[7] Usui, T., Yu, J., Frederickson, M. E.. "For colonization success, should hosts and microbes travel alone, together, or swap partners along the way?". bioRxiv. 2025-10-28T00:00:00.000Z. URL: https://doi.org/10.1101/2025.10.27.684839

[8] Bintarti, A. F., Sulesky-Grieb, A., Colovas, J., Marolleau, B., Boureau, T., SIMONIN, M., BARRET, M., Shade, A.. "Evaluating the legacy of drought exposure on root and rhizosphere bacterial microbiomes over two plant generations". bioRxiv. 2025-09-19T00:00:00.000Z. URL: https://doi.org/10.1101/2025.09.19.677321

[9] Martin, A. N. N., Williams, N. M., Vannette, R. L.. "Heatwaves do not impact bacteria within pollen provisions, despite accelerating blue orchard bee (Osmia lignaria) larval development". bioRxiv. 2026-08-27T00:00:00.000Z. URL: https://doi.org/10.64898/2026.08.26.747351

[10] Rawstern, A., Reyes, A., Li, V., Fowler, J., Hay, E., Hernandez, D., Katula, A., Gebriel, A., Afkhami, M.. "A global signature of microbiome resilience to climate stress". bioRxiv. 2026-10-02T00:00:00.000Z. URL: https://doi.org/10.64898/2026.09.30.755659

[11] Bak, N. K., Ostergaard, S. K., Schacksen, P. S., Nielsen, J. L., Rohde, P. D., Kristensen, T. N.. "Host genetics and diet jointly shape the microbiome of Drosophila melanogaster but do not predict lifespan or age-related traits". bioRxiv. 2025-10-21T00:00:00.000Z. URL: https://doi.org/10.1101/2025.10.21.683603

[12] Sanjenbam, P., Guruswamy, H. K., Deb, R., Singhal, S., Garge, S., Buddh, S., Dhanush, A., Haque, A. M., Govindarajan, S., Agashe, D.. "Ecological drift in the flour beetle microbiome and associated instability in host benefits". bioRxiv. 2025-11-08T00:00:00.000Z. URL: https://doi.org/10.1101/2025.09.16.676705

[13] Weinhold, A., Pinos, A., Correa-Carmona, Y., Holzmann, K. L., Alonso-Alonso, P., Yon, F., Steffan-Dewenter, I., Peters, M. K., Brehm, G., Keller, A.. "Host-structured adult butterfly microbiomes share a common core across continents". bioRxiv. 2026-09-22T00:00:00.000Z. URL: https://doi.org/10.64898/2026.02.27.708436

[14] Dunis, S., Lapegue, M., Deschamps, C., Cesari, L., Loiseau, A., Facon, B., Rode, N.. "Core Bacterial and Host Fruit-Specific Yeast Microbiota in a Polyphagous Fly Pest". bioRxiv. 2026-09-14T00:00:00.000Z. URL: https://doi.org/10.64898/2026.04.24.720762

How this survey was made

This survey answers one question: “How do wild insect populations' gut microbiomes respond to drought-induced host population bottlenecks, and how quickly do these communities recover in subsequent generations?”

We keep our own collection of recent preprints and refresh it every day from bioRxiv, medRxiv, arXiv, PsyArXiv, SocArXiv and ChemRxiv. For each survey, we search that collection for recent work and keep the strongest matches: papers that use the same terms as the question, and papers on the same topic in different words.

The text was written from the titles and abstracts of those papers; the full texts were not read. Statements that draw on a particular paper are marked with a number in brackets. The reference list just above is the 14 papers whose work this text draws on, chosen from the 87 papers those searches returned; the rest of the shortlist is not printed. It is not a complete review of everything published on the question.

Treat this as a starting point and a reading list, not a peer-reviewed review. The searches behind this survey ran on October 21, 2026 and covered work from the previous two years.