{"paper_id":"0d99bdf2-1dda-4d42-957a-3a9c0e794030","body_text":"Atypical tetracyclines promote longevity and ferroptotic neuroprotection via translation attenuation \n \nKhalyd J. Clay1,2, Manuel Sanchez-Alavez1,2, Ian Newman1,2, Na Na2, Ana P . Verduzco Espinoza2, Alan To1,2, Shannon Saad1, \nHollis T. Cline2, Michael Petrascheck1,2 \n \nAffiliations:  \n1Department of Molecular and Cellular Biology, The Scripps Research Institute, La Jolla, CA 92037, USA \n2Department of Neuroscience, The Scripps Research Institute, La Jolla, CA 92037, USA \nAbstract (Words: 143) \nPreclinical and clinical studies have reported neuroprotective and geroprotective effects of tetracyclines that are \nindependent of their antibiotic activity, but the underlying mechanisms remain unclear. Here, we systematically profile \nwidely used tetracyclines, including impurities and degradation products, and identify translation attenuation as the shared \ndriver of their neuroprotective and longevity-promoting effects, independent of classical tetracycline mechanisms. Instead, \nwe uncover two mechanistica lly distinct classes of tetracyclines. Mitochondrial-targeting tetracyclines (MitoTets), \nexemplified by doxycycline, inhibit the mitochondrial ribosome and attenuate cytosolic translation through activation of \nthe Integrated Stress Response (ISR). In contrast, atypical tetracyclines such as 4-epiminocycline and 12-aminominocycline \nact as cytosolic -targeting tetracyclines (CytoTets), directly inhibiting the cytosolic ribosome, bypassing the ISR, and \nprotecting neurons from ferroptotic cell death. CytoTets are non-antibiotic, brain-penetrant, and neuroprotective in mouse \nand human neurons, establishing the tetracyclines as a tunable chemical scaffold for selectively targeting translation in \naging and neurodegeneration. \n \nHighlights \n• The tetracyclines broadly attenuate translation in multiple eukaryotic models \n• Translation attenuation results from both ISR-dependent and ISR-independent mechanisms \n• Discovery of atypical, cytosolic targeting tetracyclines (CytoTETs) that protect from ferroptosis ISR-independently \n• CytoTETs inhibit translation and are neuroprotective in human-derived neurons and mouse hippocampus \n \nKeywords \nTranslation inhibition, cytosolic ribosome, mitochondrial ribosome, integrated stress response, proteostasis, ferroptosis, \nneuroprotection, longevity, aging, tetracyclines \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nIntroduction \n \nThere have long been tantalizing observations in the literature of tetracyclines eliciting anti -inflammatory and \nneuroprotective effects  in age -associated disease s such as rheumatoid arthritis or neurodegeneration 1-9. Similarly, \ntetracyclines have been found to extend lifespan in model organisms such as C. elegans or D. Melanogaster, suggesting \nthat they have genuine geroprotective effects10-13. Despite clinical evidence that tetracyclines have non -antibiotic effects \nthat are beneficial to humans, their antibiotic activity limits dosing and confounds trial results due to gastrointestinal side \neffects9,14. Consequently, non -antibiotic tetracyclines that retain their therapeutic and geroprotective profile would be \ndesirable. However, the multitude of different tetracycline  analogs, disease indications, and proposed mechanisms \nobscures whether tetracyclines  cohesively act through one central or several distinct mechanisms, hindering the \ndevelopment of geroprotective compounds. \n \nThe primary  antibiotic mechanism of tetracyclines is the inhibition of the bacterial ribosome, preventing translation\n15. \nOlder studies, comparing the activity of tetracyclines in E. coli  and S. cerevisiae , showed that tetracyclines target  \nmitochondrial ribosomes in S. cerevisiae16. These findings, in light of the ancestral origin of mitochondria 17, provide a \nconvincing narrative for  their mechanism in eukaryotes. Indeed, doxycycline was shown to target the mitochondrial \nribosome in several eukaryotes18-22 through a mechanism that  involves the activation of the mitochondrial unfolded \nprotein response (UPR mt) and the I ntegrated Stress Response (ISR) 23-25. However, unbiased chemoproteomics and \nbiochemical methods have also shown that some tetracyclines directly bind to the cytosolic ribosome 11,26,27. Inhibition of \nthe ribosome and reducing translation  is well established genetically to increase lifespan 28-32. T he most ubiquitous \nmechanism cited for the beneficial effects of tetracyclines observed in mammals is the inhibition of Matrix \nMetalloproteinase 9 (MMP9)33-35 originating from observations in diabetes–induced gingivitis2, arthritis27, and cancer28. \n \nThe non -antibiotic beneficial effects of the tetracyclines have been investigated in a variety of separate models using \nvarying analogs, which raises the possibility that the various beneficial effects reported for different  tetracyclines are the \nresult of different mechanisms dependent on the tetracycline or the result of  specific disease contexts. To elucidate the \nmechanism(s) by which the tetracyclines elicit their neuroprotective and geroprotective effects, we systematically profiled \n21 widely used tetracyclines, including their known impurities. We established their neuroprotective and geroprotective \nproperties using a combination of C. elegans, mammalian cell lines, primary neurons, and neurons derived from human \ninduced pluripotent s tem cells (iPSCs ) and  evaluated the validity of several proposed mechanisms of action. We \ndemonstrate that the neuroprotective and geroprotective effects do not require inhibition of matrix metalloproteinase 9 \n(MMP9). Instead, we find t hat tetracyclines attenuate translation in eukaryotes, through both ISR-dependent and ISR -\nindependent mechanisms, revealing an unexpected mechanistic diversity responsible for their neuroprotective and \nlifespan extending effects despite their close structural similarities. \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nResults \n \nAntibiotic activity is unrelated to tetracycline-induced neuro- or geroprotection \n \nTo survey the neuroprotective and geroprotective36,37 properties of tetracyclines, we generated a library of the most widely \nstudied tetracyclines, including several synthesis impurities and degradation products. To compare their neuroprotective \nefficacy, we measured their ability to protect hippocampal-derived HT22 cells from oxytosis/ferroptosis -mediated cell \ndeath. Oxytosis/ferroptosis is  a lipid peroxidation- mediated cell death linked to neurodegenerative and other age-\nassociated diseases. Experimentally, it is induced by high concentrations of glutamate, erastin, or RSL338,39. Fourteen of the \n21 tetracyclines tested were neuroprotective, exhibiting EC50 values of 1.5— 11 µM (Figure 1A). \n \nTo evaluate the geroprotective properties of tetracyclines, we determined their ability to extend the lifespan of wild-type \nC. elegans (N2). We identified 11 tetracyclines that increase lifespan, eight of which were also neuroprotective (Figure 1B, \nbold names; Supplementary Table 1). Five additional tetracyclines showed a tendency to extend lifespan but did not reach \nsignificance. These may be true negatives or their effect size might fall below our power of detection, which we estimated \nto be 80% at an α = 0.05 for a 15% increase in lifespan\n40 (Figure 1B).  \n \nNext, we screened our library for antibiotic activity against the Gram-negative bacterium E. coli (Figure 1C). We identified \nfive non- or reduced-antibiotic tetracyclines. These included: 12-aminominocycline and 4-epiminocycline\n41, two common \nminocycline impurities, β-apo-oxytetracycline, a degradation product of oxytetracycline42, sarecycline, a narrow-spectrum \nantibiotic with low activity against E. col i43, and Col-3. The two minocycline impurities, 4-epiminocycline and 12 -\naminominocycline, along with Col-3, also exhibited neuroprotective and geroprotective properties . Thus, the  beneficial \neffects of tetracyclines in eukaryotes are separable from their antibiotic activity.  \n \nMMP9 inhibition is not required for tetracycline-induced neuroprotection \n \nTetracycline neuroprotection is often attributed to MMP9 inhibiton via Zn2+ chelation4,44-46, but two analogs we identified \n(12-aminominocycline and R464) lack the C11-C12 β-diketone structure required for chelation (Extended Figure 1A) 47. We \ntherefore tested if MMP9 inibiton was responsible for neuroprotection. As expected, they failed to chelate Zn2+, and even \nclassical tetracyclines did so only weakly, with millimolar IC50 values (Extended Figure 1B). In a recombinat MMP9 assays, \n5 of 8 tetracyclines inhibited the enzyme at 100 µM  (Extended Figure 1C), and dose response curves showed that 12 -\naminominocycline and 4-epiminocycline did not inhibit MMP9 despite being neuroprotective (Extended Figure 1D). Thus, \ntetracycline–induced neuroprotection can be uncoupled from MMP9 inhibiton.  \n \nTetracyclines attenuate eukaryotic translation \n \nWe previously showed that the geroprotective effect of minocycline is the result of attenuate d translation\n11. Excluding \nMMP9 inhibition as the primary mechanism raised the question of whether the attenuation of translation drives \nneuroprotection as well. We therefore screened our tetracycline library for the reduction of de novo translation in HEK293 \ncells through o -propargyl puromycin incorporation (OPP)48. The OPP -labeled proteins are visualized by conjugating a \nfluorophore using click chemistry49, and quantified by comparing the green probe signal to the nuclear DAPI signal. Of the \n21 tetracyclines, 17 reduced de novo  translation in HEK293 cells  ( Figure 2 A). The efficacy of tetracyclines in reducing \ntranslation is weaker than that  of cycloheximide. Thus, w e refer to the reduction in translation by tetracyclines in \neukaryotes as attenuation. Our data show a ttenuation of translation is a general feature  of tetracyclines in eukaryot es, \ncontradicting the long-held belief that tetracyclines specifically inhibit translation in bacteria.  \n \nW\ne further  determined if all geroprotective tetracyclines attenuate translation in wild-type C. elegans. We selected  6 \ngeroprotective tetracyclines and one inactive negative control (tigecycline ) (Figure 2B ) and measured translation in \ntetracycline-treated C. elegans  using the SUnSET protocol, which visualizes the incorporation of puromycin into newly \nsynthesized proteins via immunoblot50,51. All 6 geroprotective tetracyclines att enuated translation, while tigecycline did \nnot (Figure 2 C). Notably, tigecycline attenuated translation and exhibited neuroprotective effects  in mammalian cells , \nsuggesting that its binding site is sufficiently  divergent between species . Thus, the attenuation of translation by \ntetracyclines is likely to be responsible for their geroprotective effects.  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nTetracyclines elicit neuroprotection by ISR-dependent and independent attenuation of translation \n \nWe next set out to characterize the mechanisms by which tetracyclines attenuate translation. Tetracyclines, like \ndoxycycline, that target the mitochondrial ribosome activate the ISR via the phosphorylation of eIF2α to attenuate cytosolic \ntranslation, and activate  ATF418,20-22. Alternatively, tetracyclines that target the cytosolic ribosome directly attenuate \ntranslation, independently of the ISR11. The two mechanisms can be distinguished by co-treatment with the ISR Inhibitor \n(ISRIB)52,53, which prevents tetracycline -mediated attenuation of translation in cells treated with ISR -dependent \ntetracyclines, but not in cells treated with ISR-independent tetracyclines. \n \nTo determine if the tetracycline-mediated attenuation of translation is prevented by co-treatment with ISRIB, we repeated \nthe OPP-based translation assay in HT22 cells in the presence or absence of ISRIB. As a positive control for an ISR-dependent \ninhibition of translation, we included thapsigargin54. As a positive control for the ISR-independent inhibition of translation, \nwe included cycloheximide (CHX)55.  \n \nThese experiments revealed the existence of two tetracycline classes, ISR-dependent (Figure 3A, C) and ISR-independent \ntetracyclines(Figure 3B, C). The ISR-dependent tetracyclines, such as doxycycline,  were previously shown to act on the \nmitochondrial ribosome\n18,22 while the ISR-independent tetracyclines, such as minocycline, where shown to directly bind to \nthe cytoplasmic ribosome to lower translation11,23. We thus will refer to mitochondrial targeting tetracyclines as MitoTets \n(e.g. doxycycline) and to those targeting the cytoplasmic ribosome (minocycline derivatives) as CytoTets. The ability of  \nMitoTets such as doxycycline to attenuate translation can be blocked by ISRIB, allowing us to directly test whether the \nneuroprotection depends on the attenuation of translation. Co-treatment of doxycycline-treated cells with ISRIB rescued \ntranslation and abolished the neuroprotective effect, indicating that attenuation of translation leads to neuroprotection. \n \nOf the 7 tetracyclines, we classified doxycycline, Col -3, and tigecycline as MitoTets (Figure 3D); Minocycline and R464 as \nintermediate ( Figure 3E); and 4 -epiminocycline and 12 -aminominocycline as CytoTets ( Figure 3F). Doxycycline is the \nclearest MitoTet representative, as ISRIB co-treatment reduced its neuroprotective properties the most. 4-epiminocycline \nand 12-aminominocycline were the clearest CytoTet  representatives and entirely resistant to ISRIB co-treatment (Figure \n3D, F). However, ISRIB co-treatment never completely rescued translation or completely abolished neuroprotection, even \nfor MitoTets, suggesting that all tetracyclines inhibit the cytosolic ribosome to varying degrees and that ISR activation \noccurs in addition. In summary, tetracyclines diverge into MitoTets and CytoTets based on the mechanism underlying their \nneuroprotective effect. We were unable to pinpoint unequivocal structural motifs responsible for the divergence between \nMitoTets and CytoTets, but modification in the central pharmacophore important to antibiotic activity (e.g., C11-12) tended \nto result in ISR-independent CytoTets. Similarly, inversion of the stereochemistry at the C4 position reduced antibiotic \nactivity and increased potency.  \n \nT\netracyclines extend lifespan by ISR-dependent and independent mechanisms in C. elegans \nWe next asked if the classification of tetracyclines into MitoTets and CytoTets also applies to their geroprotective effects. \nConsistent with this classification, the MitoTet doxycycline is known to inhibit the mitochondrial ribosome, triggering the \nmitochondrial unfolded protein response (UPRmt), which in turn activates the ISR to attenuate translation and upregulate \nATF-4 to extend the lifespan of C. elegans56-58. Conversely, CytoTets are expected to extend lifespan independently of the \nISR or of any stress response, such as the heat shock response (HSR) or the UPR mt, as stress response signaling converges \non the ISR 59. Whether tetracyclines exist that extend  lifespan in C. elegans  independently of the ISR,  as seen for the \nCytoTets in cell culture, is unknown.  \nWe previously showed that hyper-translation is a defining phenotype of the HSR-deficient hsf-1(sy441) mutant11,60. Hyper-\ntranslation makes the hsf-1(sy441) mutant especially resistant to many longevity paradigms, including UPR ER activation, \nreduced of mTOR activity, reduced IIS signaling , dietary restriction or hormesis 61-65, while sensitizing animals to  \ninterventions that lower translation. Thus, we tested 21 tetracyclines for their ability to extend the lifespan of hsf-1(sy441) \nmutants11,60. Eleven tetracyclines extended the lifespan of hsf-1(sy441) (Figure 4A). These 11 tetracyclines included both \nMitoTets and CytoTets, based on their classification in mammalian cells. Thus, both MitoTets and CytoTets extend lifespan \nindependently of the HSR by lowering the excessive protein synthesis of the hsf-1(sy441) mutant. This conclusion was \nfurther corroborated by evaluating the induction of the hsp-16.2::GFP heat shock reporter after treatment with 2 MitoTets \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nand 2 CytoTets, followed by a heat shock. Both MitoTets and CytoTets inhibit rather than activate the HSR (Extended Figure \n4A—4C ). \nTo probe the involvement of the UPR mt, we evaluated the ability of  the most typical MitoTet, doxycycline, and the most \ntypical CytoTets, 4-epiminocycline and 12 -aminominocycline, to induce the UPRmt reporter hsp-6::GFP66. Consistent with \ntheir classification as MitoTets and CytoTet s, d oxycycline induced the UPR mt while 12-aminominocycline did not. \nSurprisingly, 4-epiminocycline induced the UPRmt (Figure 4B & C).  \nThe induction of the UPRmt by 4-epiminocycline suggested it might act as a MitoTet in C. elegans  but as a CytoTet in \nmammals. To classify 4-epiminocycline we determined its dependence on the  ISR to extend lifespan. We measured the \nability of 4-epiminocycline and 12 -aminominocycline, to extend the lifespan of the ISR-deficient eIF2⍺(rog3)67 and \neIF2⍺(qd338)68 strains. These strains  carry phospho -deficient S49A and S49P mutations  that prevent eIF2⍺ \nphosphorylation and thus ISR activation .  A s  a  c o n t r o l ,  w e  i n c l u d e d  s a l u b r i n a l ,  a n  I S R-dependent translation initiation \ninhibitor that extends the lifespan of C. elegans60,69. 4-epiminocycline and 12-aminominocycline treatment extended the \nlifespan of ISR-deficient strains (Figure 4D, E, Extended Figure 4D) and attenuated translation in the eIF2α(qd338) mutant \n(Figure 4F, Extended Figure 4E), while  salubrinal treatment  did neither. We conclude  that 4-epiminocycline and 12 -\naminominocycline extend lifespan and attenuate translation independently of the ISR. \nTo further corroborate  4-epiminocycline and 12 -aminominocycline as CytoTet s and of doxycycline as a MitoTet , we \ndetermined their dependency on  ATF-4, an ISR transcription factor downstream of eIF2⍺ phosphorylation. Both 4 -\nepiminocycline and 12-aminominocycline extended the lifespan of atf-4(ok576) mutants (Figure 4G), while the effect of \ndoxycycline was severely blunted (Extended Figure 4F & G). Thus, the mechanistic classification of 4-epiminocycline and \n12-aminominocycline as CytoTets and of doxycycline as a MitoTet applies to geroprotection as well (Figure 4H). \n \nThe CytoTet 4-epiminocycline is neuroprotective and inhibits translation in mouse and human neurons. \nPrevious studies have shown that inhibiting the ISR can improve age-related cognitive decline in mice70 and extend lifespan \nin C. elegans 71. ISR-independent, non -antibiotic neuroprotective tetracyclines would make attractive therapeutics for \ntreating aging and age-related diseases in cases where continued ISR activation may be problematic.  \n \nA\nlthough the CytoTet profile of 12-aminominocycline was cleaner than that of 4-epiminocycline, as it had no MMP9 activity \nand did not activate the UPRmt in C. elegans, we focused on 4-epiminocycline as 12-aminominocycline was less stable. The \npharmacokinetic (PK) evaluation of 4-epiminocycline in rats established significant exposure in both the periphery and \nCNS. A single 50 mg/kg dose (i.p) resulted in plasma and brain concentrations of 1887 ng/mL and 410 ng/g, respectively, \nwhile a 25 mg/kg dose resulted in plasma and brain concentrations of 764 ng/mL and 172 ng/g, respectively (Figure 5A), \n8 hours post-injection (Figure 5B). Thus, 4-epiminocycline is brain-penetrant and has a low clearance in both plasma and \nthe CNS. \n \nWe then measured the ability of 4-epiminocycline to attenuate translation in a mouse brain in vivo by Fluorescent Non-\ncanonical Amino Acid Tagging (FUNCAT).  FUNCAT measures the incorporation of azidohomoalanine (AHA) into newly \nsynthesized proteins and visualization by bioconjugation of a TAMRA-alkyne using click chemistry (Extended Figure 5A)72. \nWe treated male C57BL/6 mice with 4-epiminocycline in the drinking water for three days, followed by an intraperitoneal \n(i.p.) co-injection with 4-epiminocycline plus AHA to label newly synthesized proteins . Since neuronal activity promote s \ntranslation72, we injected the mice with seizure-inducing pentylenetetrazol (PTZ) 30 minutes after the AHA injection. Two \nhours post-PTZ injection, the mice were sacrificed, and hippocampal proteins were extracted to quantify translation . \nTreatment with 4-epiminocycline significantly attenuated hippocampal translation by ~25% (Figure 5C & D), indicating that \n4-epiminocycline inhibits translation in vivo to the same degree as observed in all other paradigms. \n \nWe further determined whether 4-epiminocycline acts as a CytoTet or MitoTet in vivo. The MitoTet doxycycline disrupts \nthe mitonuclear translational balance, which can be visualized by the stoichiometric ratio between  nuclear-encoded \nelectron transport chain components (ETC), such as ATP5A, and mitochondrial-encoded ETC components like MT -CO1\n12. \nWe did not observe any selective suppression of the mitochondrial-encoded MT-CO1, as would be expected for a MitoTet, \nindicating that 4-epiminocycline does not inhibit mitochondrial translation (Extended Figure 5B & C). It is essential to note \nthat we normalized each lane by protein amount  and the loading control actin. This method  obfuscates a cytosolic \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nribosome-specific inhibitor since  both Actin and ATP5A are equally reduced. Taken together, these data show that 4-\nepiminocycline is a CytoTet and not a MitoTet like doxycycline.  \n \nWe then  pre-treated primary mouse neurons with 4-epiminocycline and measured survival following induction of  \noxytosis/ferroptosis73 in the presence and absence of ISRIB. As expected for an ISR -independent CytoTet, the dose -\ndependent improvement of survival of 4 -epiminocycline-treated neurons was unaffected by  ISRIB co-treatment (Figure \n5E). ISRIB activ ity was independently confirmed in primary neurons by blocking tunicamycin-induced translational \nrepression (Extended Figure 5D)54,74. \n \nFinally, we validated the ability of 4-epiminocycline to protect human neurons from oxytosis/ferroptosis. We used human \n(iPSC) line HE463#7 to differentiate into neurons (iNeurons) (Extended Figure 5E)\n75. We confirmed the neuronal identity \nof iNeurons by expression of the neuronal markers NeuN, Cux1,  and TUBB3  (Figure 5F). We also confirmed that 4-\nepiminocycline was nontoxic to iNeurons (Figure 5G). We then pre-treated iNeurons with 4-epiminocycline and induced \noxytosis/ferroptosis with RSL3. 4-epiminocycline dose-dependently increased the survival of human iNeurons (Figure 5H). \nWe then measured translation using FUNCAT and found a dose- dependent attenuation of translation by a maximum of \n40% at the highest concentration (Figure 5I). Together, we conclude that non-antibiotic CytoTets are brain penetrant and \nneuroprotective in mammals, providing an interesting new strategy for therapeutic development. \n \nDiscussion \n \nThe neuroprotective, anti-inflammatory, and geroprotective effects of tetracyclines are well-documented in the pre-clinical \nand clinical trial literature including for many age-associated diseases. While the literature on the potential therapeutic \neffects of different tetracyclines is extensive, no systematic study has been conducted to  compare and evaluate the  \nunderlying molecular mechanisms.  Here, w e profiled the most widely used tetracyclines  and their impurities to \ncharacterize their geroprotective and neuroprotective activities and mechanisms of action.  \n \nOur data establish: (i) Modifications in the C4 or C11,12 positions of tetracyclines weaken their antibiotic and MMP9 \ninhibitory activity and uncouple t heir neuroprotective and geroprotective effects from antibiotic activity (Figure 1 & \nExtended Figure 1). (ii) Instead, the neuroprotective and geroprotective effects of tetracyclines are the result of attenuation \nof translation (Figures 2 & 3). (iii) The mechanism of action  by which tetracyclines attenuate translation classifies them \ninto ISR-dependent ( MitoTets) and ISR -independent tetracyclines (CytoTets) ( Figures 3 & 4 ). MitoTets target the \nmitochondrial ribosome, resulting in ISR -dependency, while CytoTets target the cytosolic ribosome directly . (iv) \nMinocycline degradation products like 4-epiminocycline show reduced antibiotic activity, are brain penetrant, and protect \nboth mouse and human neurons from oxytosis/ferroptosis (Figure 5).  \n \nWe first uncoupled the neuroprotective effects from MMP9 inhibition  by showing that structurally related minocycline \nanalogs fail to inhibit MMP9 but remain neuroprotective (Extended Figure 1, 3E, F). This evidence challenges the pervasive \nassertion that MMP9 inhibition is the causal mechanism underlying tetracycline neuroprotection, consistent with past \nfailures of specific MMP inhibitors to recapitulate the properties of tetracyclines. \n \nWe then focused on 7 diverse tetracyclines  to evaluate the attenuation of translation as the underlying mechanism for \nneuroprotection\n11-13,76. All 7 neuroprotective tetracyclines lower protein synthesis but  did so through  ISR-dependent \n(MitoTets) and ISR-independent  ( CytoTets) mechanisms ( Figure 3). The ISR  dependency of doxycycline enabled us to \ndemonstrate that the attenuation of translation directly causes neuroprotection , since t he ISR inhibitor ISRIB  restored \ntranslation and abolished neuroprotection during doxycycline treatment, revealing a causal relatio nship (Figure 3). The \nISR-based classification of tetracyclines into MitoTets and CytoTets also applies to their effect on longevity. (Figure 2)77. \nMitoTets, such as doxycycline, induce the UPRmt and ISR and require ATF-4 to extend lifespan18-22,25,56. Conversely, CytoTets \nsuch as 12-aminominocycline and 4-epiminocycline neither require the ISR nor ATF -4 (Figure 4). However, our data also \nsuggest that many tetracyclines fall in between the MitoTet and CytoTets extremes and are partially ISR dependent. \n \nM\nany antibiotics that inhibit bacterial ribosomes also inhibit mitochondrial ribosomes by identical mechanisms78,79. Cryo-\nEM structures of tigecycline bound to the mitoribosome showed binding in the peptidyl transferase center that overlapped  \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nwith the binding site for structurally related tetracyclines in bacterial ribosomes19. A recent screen aimed at identifying \nmitochondrial-targeting tetracyclines with limited antibiotic activity identified 9-t-butyl-doxycycline as a therapeutically \ndevelopable compound for influenza19. Notably, none of the 18 tetracyclines derived from the minocycline scaffold \nactivate the UPRmt, confirming that doxycycline-related tetracyclines tend to target mitochondrial ribosomes, whereas \nminocycline-related tetracyclines tend to target cytosolic ribosomes11. Thus, tetracyclines represent a molecular scaffold \nwhose neuroprotective and geroprotective properties can be tuned to be ISR-dependent or independent by either \ntargeting the mitochondrial or cytosolic ribosome.  \n \nWe examined the potential of the CytoTet 4-epiminocycline as a neuroprotective agent . 4 -epiminocycline is the \nepimerization product of minocycline and only differs in the stereochemistry of the C4 position, resulting in low antibiotic \nactivity, while remaining  brain penetr ant and improving neuroprotection  efficacy ( Figures 1 & 5). 4-epiminocycline \nattenuated translation in the hippocampus of mice as well as in human iNeurons  by ~25% (Figures 5D & 5I)80. Translation \nwas reduced without disrupting the mito-nuclear balance of ETC proteins  (Extended Figure Figure 5B), as would be \nexpected for doxycycline but not minocycline derivatives 12. Importantly, 4-epiminocycline protected primary mouse and \nhuman iNeurons from RSL3-induced oxytosis/ferroptosis (Figure 5E & 5H) independent of the ISR18. \n \nMinocycline, doxycycline, and other tetracyclines have been tested for their beneficial effects across various indications \nrelated to aging but unrelated to their antibiotic activity\n59. However, the results in human clinical trials have been mixed9. \nOne of the critical limitations mentioned in many studies is the antibiotic activity, which is both dose-limiting and a liability \nfor chronic use due to associated gastroin testinal side-effects and the important role of the microbiome in aging 9,14,81,82 \nSeveral recent studies, including this one, have demonstrated that the beneficial pre-clinical effects can be uncoupled from \nthe antibiotic activity, as illustrated by 4 -epiminocycline. In addition, the surprising mechanistic diversity of tetracyclines \nhas clear implications for the interpretation of past and future clinical trials. The results of both failed and successful trials \nneed to be reconsidered in light of the ISR dependency of individual tetracyclines. Activation of the ISR can have both \nbeneficial and detrimental effects , depending on the disease indication, and thus drive the failure or success of a trial. \nSimilarly, when generating novel non-antibiotic tetracycline derivatives, it will be crucial to consider the ISR dependency \nof the parent scaffold and to match it in the non-antibiotic derivatives. Finally, our identification of minocycline impurities \nwith superior efficacy and potency may explain at least some of the mixed or disappointing results in clinical studies, as it \nrepresents a variable that was poorly controlled for in any study. Pharmacologically active metabolites with higher potency \nthan the parent compound are well-documented in drug discovery 83. Overall, our study suggests that tetracyclines, with \ntheir extensive safety profile offer the possibility to attenuate translation in humans and thus chemically target one of the \nmost potent lifespan extending mechanisms identified in model organisms.  \n \nAcknowledgments \nSome strains were provided by the CGC, funded by the NIH Office of Research Infrastructure Programs (P40 OD010440). \nThis work was supported by grants from the NIH (1RF1AG079517-01 to H.T.C. and 1R21NS107951-01 t o  M . P. ) ,  a  \nFellowship from the Helen Dorris Foundation to K.C. \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n \nFigure 1: Antibiotic activity is not required for tetracycline-induced neuro - or geroprotection. Top panels depict \nexperimental strategy to phenotypically profile 21 commercially available tetracyclines across multiple species for \nneuroprotection (A), geroprotection (B), and antibiotic activity (C). (A) Graph shows Log10(EC50) for neuroprotection. EC50 \nvalues were derived from dose response curves measuring cell viability of hippocampal-derived HT22 cells subjected to \noxytosis/ferroptosis. HT22 cells were pre-treated for 4 hrs with increasing concentrations of tetracyclines, followed by the \ninduction of oxytosis/ferroptosis with glutamate (5 mM) and determination of cell survival 20 hrs later. Error bars indicated \nmean ± SD. Data representative of three independent trials. * indicates compounds that were cytotoxic at the highest \nconcentration. (B) Bar graphs show % change in lifespan for N2 C. elegans treated with the indicated tetracycline starting \non day 1 of adulthood. Doses tested were 33 or 100 µM, except for 12 -aminominocycline, which was tested at 200 µM. \nSignificance was determined by the long-rank test. Black bars: p <  0.05; grey bars: p > 0.05. N > 50 animals per treatment. \nSee Supplementary Data for more details. The dotted line indicates the estimated power of detection. We expect to identify \n80% of all tetracyclines that extend lifespan by 15% or more at an α = 0.05. (C) Bacterial growth as a function of time for \nE. coli (OP50) in the pr esence of 100 µM of each t etr acycline. T etr acyclines that allowed mor e than 50% gr owth were \nconsidered reduced -antibiotic, while tetracyclines that had the same growth curve as DMSO were considered non -\nantibiotic. DMSO is omitted as it is identical to that of β-appo-oxytetracyline. DMSO and kanamycin were used as negative \nand positive controls, respectively. Total of 3 independent experiments.  \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n \nExtended Figure 1: MMP 9 inhibition is not required for tetracycline-induced neuroprotection. (A) Shown is the general \nstructure of the tetracyclines with the keto -enol system (pink) at C11 & C12 that is responsible for the chelation of Zn 2+ \nand other divalent ions. Structural modifications in 12 -aminominocycline and R464 disrupt the chelation center (blue & \norange circles). (B) Colorimetric assay measuring the % remaining Zn2+ as a function of tetracycline concentration, indicative \nof the chelation effect of each tetracycline. 12-aminominocycline does not chelate ions at concentrations up to 1800 µM, \nwhereas 300 µM is required for R464. Other tetracyclines tested (grey) include: 4 -epiminocycline, minocycline, Col-3, \nDoxycycline, Tigecycline, & Evaracycline. EDTA was used as a positive control. (C) Bar graph shows the % remaining \nrecombinant MMP9 activity after tetracycline treatment (100 µM). The assay measures proteolytic cleavage of a \nfluorogenic substrate, released upon cleavage. NNGH is a broad-spectrum inhibitor of matrix metalloproteinases and was \nused as a positive control. Significance was determined by one-way ANOVA with Dunnett's multiple comparisons, where \n*** = p < 0.001 and **** = p < 0.0001. Error bars indicate mean ± SD from three independent trials. (D) Dose response \ncurve of four neuroprotective tetracyclines. Tetracyclines with IC50 greater than 300 µM are indicated as \"not determined\" \n(n.d.). \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n \nFigure 2: Tetracyclines attenuate translation in eukaryotes. (A) Graph shows fold change in de novo translation in HEK293 \ncells treated with individual tetracyclines. Each tetracycline was tested at a concentration of 100 µM. Translation was \nquantified by the OPP incorporation assay determining the mean OPP incorporation -intensity relative to DAPI. \nCycloheximide (CHX) serves as a positive control. Total of 3 independent trials with at least 4 images quantified per trial.  \n(B) Chemical structures of 7 diverse tetracyclines, representing an array of structural modifications. (C) A puromycin \nincorporation assay (SUnSET) was used to monitor the effect of tetracycline treatment on translation in C. elegans (100 \nµM, for 4 hrs). Immunoblot of protein extracts stained with puromycin antibody reveals reduced puromycin incorporation \nand hence translation (top). GAPDH is used as a loading control. Quantification of three independent SUnSET trials shows \nbroad translation inhibition  of most tetracyclines. Significance was determined by one-way ANOVA using Dunnett's \nmultiple comparisons correction, where black represents p < 0.05 and grey represents p > 0.05. \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n \nFigure 3: Tetracyclines protect neurons by ISR-dependent and ISR-independent attenuation of translation. \n(A) Representative fluorescent micrographs of HT22 cells stained for OPP incorporation as a readout for de novo  \ntranslation. Translation is attenuated by ISR-dependent tetracyclines (MitoTets) and restored by co-treatment with ISRIB. \nThapsigargin (Thaps., 1 µM) is included as a control for an ISR-dependent translation inhibitor. Tetracyclines were screened \nat 100 µM. (B) Same fluorescent micrographs as in A, but for ISR-independent tetracyclines (CytoTets) whose translation \nattenuation is unaffected by ISRIB co -treatment. Cycloheximide (CHX, 500 nM ) is included as a control for an ISR -\nindependent translation inhibitor. (C) Quantification of the OPP incorporation-intensity relative to the DAPI signals shown \nin Figure 4A & 4B. Several tetracyclines attenuate translation independently of the ISR (black vs. grey bars). Significance \nwas determined by two-way ANOVA with Šídák multiple comparisons test. ** = p < 0.01 and **** = p < 0.0001. (D) Graphs \nshow the % survival of HT22 cells as a function of tetracycline dose. The % survival is calculated relative to non-glutamate-\ntreated HT22 control cells (100%). HT22 cells were pre-treated with ISR -dependent tetracyclines (MitoTets), alone or in \ncombination with ISRIB followed by the induction of oxytosis/ferroptosis. Thapsigargin (Thaps.) is used as a positive, ISR -\ndependent control. (E) Same as D, but for partially ISR -dependent tetracyclines. (F) Same as D, but for ISR-independent \n(CytoTets) tetracyclines. For all figures: Cells were treated for 2 hrs with ISRIB (300 nM, grey bar, or dotted line) or vehicle \ncontrol (DMSO, black bar/line) prior to co -incubation with the indicated tetracycline or control compound. Error bars \nindicate mean ± SD from three independent trials. \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n \nFigure 4: 4-epiminocycline and 12-aminocycline extend lifespan independently of the ISR. (A) Bar graphs show % change \nin lifespan for hsf -1(sy441) mutants treated with the indicated tetracycline starting on day 1 of adulthood. Doses tested \nwere 33 or 100 µM, except for 12-aminominocycline, which was tested at 200 µM. Black bars p < 0.05, grey bars p > 0.05. \nN > 50 animals per treatment. See Supplementary Data for more details. The dotted line indicates the estimated power of \ndetection. We expect to identify 80% of all tetracyclines that extend lifespan by 15% or more at an α = 0.05. For additional \ndata on the HSR, see Extended Figure 5A—5C.  (B) Representative fluorescent images of 50 randomly selected hsp-6p::GFP \nUPRmt reporter animals treated with the indicated tetracycline. Images were taken in parallel following measurement in \nFigure 5C. Images were inverted for clarity. (C) Scatter plot shows the fold induction relative to DMSO of the hsp-6::GFP \nUPRmt reporter in response to 12-hrs tetracycline treatment initiated at the L2 stage. Significance was determined by one-\nway ANOVA with Dunnett's multiple comparisons, where **** = p < 0.0001. Error bars indicate mean ± SD of 3 independent \ntrials. (D) Survival plot of wild-type (N2) animals treated with tetracyclines or the ISR -dependent translation inhibitor \nsalubrinal. Tetracycline and salubrinal treatment extend the lifespan of N2. (E) Survival plot of ISR-deficient eIF2⍺(qd338) \nmutants treated with tetracyclines or salubrinal. Only tetracyclines, but not salubrinal treatment, extend the lifespan of \neIF2⍺(qd338) mutants. For similar data on eIF2⍺(rog3), see Extended Figure 5D.  (F) A puromycin incorporation assay \n(SUnSET) was used to monitor the effect of tetracycline or salubrinal treatment on translation in the ISR -deficient \neIF2⍺(qd338) mutant. Only tetracyclines, but not salubrinal treatment, reduce translation in the ISR-deficient eIF2⍺(qd338) \nmutant. GAPDH was used as a loading control. For quantification, see Extended Figure 5E. (G) Survival plot of the partially \nISR-deficient atf-4(ok576) mutants treated with tetracyclines. Tetracyclines extend lifespan mostly independent of ATF-4. \nFor corresponding survival data on doxycycline, see Extended Figure 5F & 5G. (H) Schematic outlining distinguishing \nfeatures between CytoTets and MitoTets. CytoTets, as exemplified by 12-aminominocycline and 4-epiminocycline, directly \ninhibit translation to extend lifespan, while MitoTets, as exemplified by doxycycline, activate the mitochondrial UPR to \ninhibit translation and subsequently extend lifespan in an atf-4 dependent manner. The significance of all survival data was \ndetermined by the log-rank test. In B–G all compounds were tested at 100 µM or 200 µM for 12-aminominocycline. \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n \nExtended Figure 4: Tetracyclines elicit stress response-dependent and independent longevity mechanisms \n(A) Scatter plot shows the fold induction relative to DMSO of the hsp16.2::GFP heat shock response reporter after \ntetracycline treatment, followed by a 1.5 hr, 35°C heat shock (HS) . (B)  qRT-PCR quantification of relative GFP  mRNA \nexpression from untreated or tetracycline-treated animals with and without a 1.5 hr, 35°C HS. Tetracycline treatment \nsuppresses the HS-induced GFP fluorescence of the hsp-16.2p::GFP reporter at the protein (A) but not at the mRNA level \n(B). (C) qRT-PCR quantification of endogenous hsp16.2 mRNA expression following HS with or without pre -incubation of \ntetracycline treatment of wild-type (N2) animals. ( D) Survival plot of ISR -deficient eIF2⍺(rog3) mutants, which lack the \neIF2⍺ phosphorylation site. Only tetracyclines, but not salubrinal treatment, extend lifespan. (E) Quantification of 3 \nbiological replicates from the SUNSET experiment in Figure 5F shows that 4 -epiminocycline and 12-aminominocycline do \nnot depend on eIF2 ⍺ phosphorylation for translation inhibition, while salubrinal does. (F) Survival plot of N2 and the \npartially ISR -deficient atf-4(ok576) mutant treated with DMSO or doxycycline.  ATF-4 is partially required for lifespan \nextension by doxycycline. The survival curve is one of the three trials quantified in Extended Figure 5G. (G) Comparison of \nthe mean lifespan extension of N2 and atf-4(ok576) animals treated with the indicated tetracycline. Doxycycline specifically \nloses efficacy in atf-4(ok576) mutants across 3 biological replicates. Statistics for B, C, G: Significance was determined by \none-way ANOVA with Dunnett's multiple comparisons, where ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001. All error \nbars indicate mean ± SD from three independent trials. Significance for all survival data (D, E) was determined by the log-\nrank test. \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n \nFigure 5: 4-epiminocycline is a non-antibiotic brain penetrant CytoTet that protects human neurons from ferroptosis.(A) \nPlasma concentrations of 4 -epiminocycline over 8 hr s following a single intraperitoneal injection (i.p.) of male Sprague –\nDawley rats. (B) Brain concentrations of 4 -epiminocycline after 8 hrs following a single intraperitoneal injection of male \nSprague–Dawley rats. (C ) Fluorescent-scan (top) and Coomassie blue stain (bottom) gel. In -gel fluorescence detects de \nnovo translation by the amount of incorporated fluorescently labelled AHA (top) relative to total protein (bottom). 4 -\nepiminocycline reduced de novo translation in the hippocampus of mice. (D) Quantification of FUNCAT experiments shows \n4-epiminocycline reduces de novo translation by 25%. Significance was determined by a two-tailed Student's t-test, where \n*** = p < 0.001 and N = 6. (E) Survival of primary hippocampal neurons as a function of 4 -epiminocycline concentration \nafter ferroptosis induction by RSL3 (300 nM). Co-treatment with ISRIB (300 nM) does not abolish the neuroprotective effect \nof 4-epiminocycline, confirming its classification as a CytoTet. (F) (Top) Cropped representative confocal images of \nimmunofluorescent labeling of generated iNs. Nuclei (DAPI, white), cortical marker CUX1 (Cux1, red), and neuronal marker \n(NeuN, green) stains are shown. (Bottom) Validation of differentiation into iNeurons, by staining for nuclei (DAPI, white), \nthe pan neuronal marker TUBB3 (βIII-Tub, red), and the neuronal marker (NeuN, green). (G) No cytotoxicity was detected \nin iNeurons 48 hrs after 4-epiminocycline addition. (H) Survival of iNeurons as a function of 4-epiminocycline concentration \nafter ferroptosis induction by RSL3 ( 2 µM). (I) Quantification of de novo translation in iNeurons treated with increasing \nconcentration of 4 -epiminocycline using FUNCAT. iNeurons were co -incubated with (AHA, 8mM) and increasing \nconcentrations of 4 -epiminocycline for 4 hr s, followed by lysis and bioconjugation of an Alexa -488 alkyne to the \nincorporated AHA by a click reaction. The protein synthesis inhibitor cycloheximide (CHX, 250nM) was used as a positive \ncontrol. Significance was determined by one-way ANOVA with Dunnett's multiple comparisons, where * = p < 0.05, *** = \np < 0.001 and **** = p < 0.0001. All error bars indicate mean ± SD from 3-6 independent trials or animals. Cell survival was \nmeasured using Cell Titer Glo 20 hrs after treatment in all experiments. \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n \nExtended Figure 5: 4-epiminocycline targets translation in mammalian neurons \n(A) Experimental scheme to measure de novo  translation by FUNCAT in the brains of male C57B6 mice. Following 4 -\nepiminocycline i.p. injections (100 mg/kg), the animals were injected with AHA (50 mg/kg) and then with \npentylenetetrazole (50 mg/kg) 30 minutes later. The hippocampus was micro -dissected and lysed to fluorescently label \nnewly synthesized proteins that incorporated AHA by click chemistry. (B ) 4- epiminocycline did not alter the ratio of \nnuclear–encoded (ATP-5A) to mitochondrial -encoded (MT-CO1) electron transport chain proteins. (C ) Quantification of \nExtended Figure 5B. Actin was used as a loading control for normalization between samples before comparison of \nnuclear/mitochondrial–encoded proteins. ns = p > 0.05. (D) Validation of ISRIB activity in Figure 5E. Primary neurons were \nco-t r e a t e d  w i t h  1  µ M  t u n i c a m y c i n  ( T m )  t o  c o n fi r m  t h e  a b i l i t y  o f  I S R I B  (300 nM) to rescue translation inhibited by \ntunicamycin. The 18 hr  timepoint shows significant inhibition of translation by Tm, which is rescued with ISRIB. (E) \nSchematic outlining the generation of iNeurons from APOE3 (HE463#7) iPSCs and experimental procedure.\n \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n \nReferences \n \n1 Chopra, I. Tetracycline analogs whose primary target is not the bacterial ribosome. Antimicrob Agents \nChemother 38, 637–640 (1994). https://doi.org/10.1128/aac.38.4.637 \n2 Golub, L. M. et al.  Minocycline reduces gingival collagenolytic activity during diabetes. Preliminary \nobservations and a proposed new mechanism of action. J Periodontal Res 18, 516– 526 (1983). \nhttps://doi.org/10.1111/j.1600-0765.1983.tb00388.x \n3 Sapadin, A. N. & Fleischmajer, R. Tetracyclines: nonantibiotic properties and their clinical implications. J \nAm Acad Dermatol 54, 258–265 (2006). https://doi.org/10.1016/j.jaad.2005.10.004 \n4 Rudek, M. A.  et al. Phase I clinical trial of oral COL -3, a matrix metalloproteinase inhibitor, in patients \nwith refractory metastatic cancer. J Clin Oncol  19, 584– 592 (2001). \nhttps://doi.org/10.1200/jco.2001.19.2.584 \n5 Xu, X.  et al.  Doxycycline improves clinical outcomes during cystic fibrosis exacerbations. European \nRespiratory Journal 49, 1601102 (2017). https://doi.org/10.1183/13993003.01102-2016 \n6 Nettis, M. A. et al. Augmentation therapy with minocycline in treatment -resistant depression patients \nwith low -grade peripheral inflammation: results from a double -blind randomised clinical trial. \nNeuropsychopharmacology 46, 939–948 (2021). https://doi.org/10.1038/s41386-020-00948-6 \n7 Liu, F.  et al.  Minocycline supplementation for treatment of negative symptoms in early -phase \nschizophrenia: a double blind, randomized, controlled trial. Schizophrenia research 153, 169–176 (2014). \nhttps://doi.org/10.1016/j.schres.2014.01.011 \n8 Champigny, C. et al. Combining Lovastatin and Minocycline for the Treatment of Fragile X Syndrome: \nResults From the LovaMiX Clinical Trial. Frontiers in Psychiatry  12 (2022). \nhttps://doi.org/10.3389/fpsyt.2021.762967 \n9 Howard, R. et al. Minocycline at 2 Different Dosages vs Placebo for Patients With Mild Alzheimer Disease: \nA Randomized Clinical Trial. JAMA Neurology  77, 164– 174 (2020). \nhttps://doi.org/10.1001/jamaneurol.2019.3762 \n10 Lee, G. J., Lim, J. J. & Hyun, S. Minocycline treatment increases resistance to oxidative stress and extends \nlifespan in Drosophila via FOXO. Oncotarget  8, 87878– 87890 (2017). \nhttps://doi.org/10.18632/oncotarget.21224 \n11 Solis, G. M. et al. Translation attenuation by minocycline enhances longevity and proteostasis in old post-\nstress-responsive organisms. Elife 7 (2018). https://doi.org/10.7554/eLife.40314 \n12 Houtkooper, R. H. et al. Mitonuclear protein imbalance as a conserved longevity mechanism. Nature 497, \n451–457 (2013). https://doi.org/10.1038/nature12188 \n13 Ye, X., Linton, J. M., Schork, N. J., Buck, L. B. & Petrascheck, M. A pharmacological network for lifespan \nextension in Caenorhabditis elegans. Aging cell 13, 206–215 (2014). https://doi.org/10.1111/acel.12163 \n14 Haïk, S. et al. Doxycycline in Creutzfeldt-Jakob disease: a phase 2, randomised, double -blind, placebo-\ncontrolled trial. The Lancet. Neurology  13, 150– 158 (2014). https://doi.org/10.1016/s1474-\n4422(13)70307-7 \n15 Brodersen, D. E. et al. The structural basis for the action of the antibiotics tetracycline, pactamycin, and \nhygromycin B on the 30S ribosomal subunit. Cell 103, 1143–1154 (2000). https://doi.org/10.1016/s0092-\n8674(00)00216-6 \n16 Clark-Walker, G. D. & Linnane, A. W. In vivo differentiation of yeast cytoplasmic and mitochondrial protein \nsynthesis with antibiotics. Biochemical and biophysical research communications  25, 8 –13 (1966). \nhttps://doi.org/10.1016/0006-291x(66)90631-0 \n17 Chatzispyrou, I. A., Held, N. M., Mouchiroud, L., Auwerx, J. & Houtkooper, R. H. Tetracycline antibiotics \nimpair mitochondrial function and its experimental use confounds research. Cancer Res 75, 4446–4449 \n(2015). https://doi.org/10.1158/0008-5472.Can-15-1626 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n18 Mottis, A. et al. Tetracycline-induced mitohormesis mediates disease tolerance against influenza. J Clin \nInvest 132 (2022). https://doi.org/10.1172/jci151540 \n19 Shao, Q. et al. T cell toxicity induced by tigecycline binding to the mitochondrial ribosome. Nat Commun \n16, 4080 (2025). https://doi.org/10.1038/s41467-025-59388-9 \n20 Perry, E. A.  et al. Tetracyclines promote survival and fitness in mitochondrial disease models. Nature \nMetabolism 3, 33–42 (2021). https://doi.org/10.1038/s42255-020-00334-y \n21 Molenaars, M. et al. A Conserved Mito -Cytosolic Translational Balance Links Two Longevity Pathways. \nCell Metabolism (2020). https://doi.org/https://doi.org/10.1016/j.cmet.2020.01.011 \n22 Ronayne, C. T. et al. Tetracyclines activate mitoribosome quality control and reduce ER stress to promote \ncell survival. EMBO reports  24, e57228 (2023). \nhttps://doi.org/https://doi.org/10.15252/embr.202357228 \n23 Mortison, J. D. et al. Tetracyclines Modify Translation by Targeting Key Human rRNA Substructures. Cell \nChem Biol 25, 1506–1518 e1513 (2018). https://doi.org/10.1016/j.chembiol.2018.09.010 \n24 Brüning, A., Brem, G. J., Vogel, M. & Mylonas, I. Tetracyclines cause cell stress-dependent ATF4 activation \nand mTOR inhibition. Exp Cell Res 320, 281–289 (2014). https://doi.org/10.1016/j.yexcr.2013.11.012 \n25 Vendramin, R. et al. Activation of the integrated stress response confers vulnerability to mitoribosome-\ntargeting antibiotics in melanoma. Journal of Experimental Medicine  218, e20210571 (2021). \nhttps://doi.org/10.1084/jem.20210571 \n26 Osterman, I. A. et al. Tetracenomycin X inhibits translation by binding within the ribosomal exit tunnel. \nNature Chemical Biology 16, 1071–1077 (2020). https://doi.org/10.1038/s41589-020-0578-x \n27 Protasoni, M., Kroon, A. M. & Taanman, J. W. Mitochondria as oncotarget: a comparison between the \ntetracycline analogs doxycycline and COL -3. Oncotarget 9, 33818– 33831 (2018). \nhttps://doi.org/10.18632/oncotarget.26107 \n28 Hansen, M.  et al. Lifespan extension by conditions that inhibit translation in Caenorhabditis elegans. \nAging cell 6, 95–110 (2007). https://doi.org/10.1111/j.1474-9726.2006.00267.x \n29 Pan, K. Z. et al. Inhibition of mRNA translation extends lifespan in Caenorhabditis elegans. Aging cell 6, \n111–119 (2007). https://doi.org/10.1111/j.1474-9726.2006.00266.x \n30 Kapahi, P . et al. Regulation of lifespan in Drosophila by modulation of genes in the TOR signaling pathway. \nCurr Biol 14, 885–890 (2004). https://doi.org/10.1016/j.cub.2004.03.059 \n31 Rogers, A. N. et al. Life span extension via eIF4G inhibition is mediated by posttranscriptional remodeling \nof stress response gene expression in C. elegans. Cell Metab 14, 55 –66 (2011). \nhttps://doi.org/10.1016/j.cmet.2011.05.010 \n32 Curran, S. P . & Ruvkun, G. Lifespan Regulation by Evolutionarily Conserved Genes Essential for Viability. \nPLOS Genetics 3, e56 (2007). https://doi.org/10.1371/journal.pgen.0030056 \n33 Golub, L. M., Ramamurthy, N. S., McNamara, T. F., Greenwald, R. A. & Rifkin, B. R. Tetracyclines inhibit \nconnective tissue breakdown: new therapeutic implications for an old family of drugs. Crit Rev Oral Biol \nMed 2, 297–321 (1991). https://doi.org/10.1177/10454411910020030201 \n34 Vandenbroucke, R. E. & Libert, C. Is there new hope for therapeutic matrix metalloproteinase inhibition? \nNature Reviews Drug Discovery 13, 904–927 (2014). https://doi.org/10.1038/nrd4390 \n35 Bruno, M. A.  et al. Amyloid beta-induced nerve growth factor dysmetabolism in Alzheimer disease. J \nNeuropathol Exp Neurol 68, 857–869 (2009). https://doi.org/10.1097/NEN.0b013e3181aed9e6 \n36 Domercq, M. & Matute, C. Neuroprotection by tetracyclines. Trends in pharmacological sciences 25, 609–\n612 (2004). https://doi.org/10.1016/j.tips.2004.10.001 \n37 Bortolanza, M.  et al.  Tetracycline repurposing in neurodegeneration: focus on Parkinson's disease. J \nNeural Transm (Vienna) 125, 1403–1415 (2018). https://doi.org/10.1007/s00702-018-1913-1 \n38 Lewerenz, J., Ates, G., Methner, A., Conrad, M. & Maher, P . Oxytosis/Ferroptosis-(Re-) Emerging Roles for \nOxidative Stress-Dependent Non-apoptotic Cell Death in Diseases of the Central Nervous System. Front \nNeurosci 12, 214 (2018). https://doi.org/10.3389/fnins.2018.00214 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n39 Sarparast, M.  et al.  Dihydroxy-Metabolites of Dihomo -γ-linolenic Acid Drive Ferroptosis -Mediated \nNeurodegeneration. ACS Cent Sci 9, 870–882 (2023). https://doi.org/10.1021/acscentsci.3c00052 \n40 Petrascheck, M. & Miller, D. L. Computational Analysis of Lifespan Experiment Reproducibility. Front \nGenet 8, 92 (2017). https://doi.org/10.3389/fgene.2017.00092 \n41 Nelis, H. J. & Leenheer, A. P . D. Metabolism of minocycline in humans. Drug Metabolism and Disposition \n10, 142 (1982).  \n42 Halling-Sørensen, B., Sengeløv, G. & Tjørnelund, J. Toxicity of tetracyclines and tetracycline degradation \nproducts to environmentally relevant bacteria, including selected tetracycline -resistant bacteria. Arch \nEnviron Contam Toxicol 42, 263–271 (2002). https://doi.org/10.1007/s00244-001-0017-2 \n43 Zhanel, G., Critchley, I., Lin, L. -Y . & Alvandi, N. Microbiological Profile of Sarecycline, a Novel Targeted \nSpectrum Tetracycline for the Treatment of Acne Vulgaris. Antimicrob Agents Chemother  63, \n10.1128/aac.01297–01218 (2018). https://doi.org/10.1128/aac.01297-18 \n44 Lindeman, J. H., Abdul -Hussien, H., van Bockel, J. H., Wolterbeek, R. & Kleemann, R. Clinical trial of \ndoxycycline for matrix metalloproteinase -9 inhibition in patients with an abdominal aneurysm: \ndoxycycline selectively depletes aortic wall neutrophils and cytotoxic T cells. Circulation 119, 2209–2216 \n(2009). https://doi.org/10.1161/circulationaha.108.806505 \n45 Dezube, B. J., Krown, S. E., Lee, J. Y ., Bauer, K. S. & Aboulafia, D. M. Randomized phase II trial of matrix \nmetalloproteinase inhibitor COL -3 in AIDS -related Kaposi's sarcoma: an AIDS Malignancy Consortium \nStudy. J Clin Oncol 24, 1389–1394 (2006). https://doi.org/10.1200/jco.2005.04.2614 \n46 Chang, J. J. et al. Minocycline and matrix metalloproteinase inhibition in acute intracerebral hemorrhage: \na pilot study. Eur J Neurol 24, 1384–1391 (2017). https://doi.org/10.1111/ene.13403 \n47 Hidalgo, M. & Eckhardt, S. G. Development of matrix metalloproteinase inhibitors in cancer therapy. J \nNatl Cancer Inst 93, 178–193 (2001). https://doi.org/10.1093/jnci/93.3.178 \n48 Liu, J., Xu, Y ., Stoleru, D. & Salic, A. Imaging protein synthesis in cells and tissues with an alkyne analog of \npuromycin. Proceedings of the National Academy of Sciences 109, 413– 418 (2012). \nhttps://doi.org/10.1073/pnas.1111561108 \n49 Kolb, H. C., Finn, M. G. & Sharpless, K. B. Click Chemistry: Diverse Chemical Function from a Few Good \nReactions. Angew Chem Int Ed Engl 40, 2004– 2021 (2001). https://doi.org/10.1002/1521-\n3773(20010601)40:11<2004::Aid-anie2004>3.0.Co;2-5 \n50 Arnold, A. et al. Functional characterization of C. elegans Y -box-binding proteins reveals tissue -specific \nfunctions and a critical role in the formation of polysomes. Nucleic acids research 42, 13353– 13369 \n(2014). https://doi.org/10.1093/nar/gku1077 \n51 Yan, J.  et al.  Eukaryotic Elongation Factor 2 Kinase EFK -1/eEF2K promotes starvation resistance by \npreventing oxidative damage in C. elegans. Nat Commun 16, 1752 (2025). \nhttps://doi.org/10.1038/s41467-025-56766-1 \n52 Sidrauski, C. et al. Pharmacological brake-release of mRNA translation enhances cognitive memory. Elife \n2, e00498 (2013). https://doi.org/10.7554/eLife.00498 \n53 Zyryanova, A. F.  et al.  ISRIB Blunts the Integrated Stress Response by Allosterically Antagonising the \nInhibitory Effect of Phosphorylated eIF2 on eIF2B. Mol Cell  81, 88– 103.e106 (2021). \nhttps://doi.org/10.1016/j.molcel.2020.10.031 \n54 Sidrauski, C., McGeachy, A. M., Ingolia, N. T. & Walter, P . The small molecule ISRIB reverses the effects of \neIF2α phosphorylation on translation and stress granule assembly. Elife 4 (2015). \nhttps://doi.org/10.7554/eLife.05033 \n55 Schneider- Poetsch, T.  et al.  Inhibition of eukaryotic translation elongation by cycloheximide and \nlactimidomycin. Nat Chem Biol 6, 209–217 (2010). https://doi.org/10.1038/nchembio.304 \n56 Borankova, K. et al.  Mitoribosomal synthetic lethality overcomes multidrug resistance in MYC -driven \nneuroblastoma. Cell Death Dis 14, 747 (2023). https://doi.org/10.1038/s41419-023-06278-x \n57 Lima, T., Li, T. Y ., Mottis, A. & Auwerx, J. Pleiotropic effects of mitochondria in aging. Nature Aging 2, 199–\n213 (2022). https://doi.org/10.1038/s43587-022-00191-2 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n58 Nargund, A. M., Pellegrino, M. W., Fiorese, C. J., Baker, B. M. & Haynes, C. M. Mitochondrial import \nefficiency of ATFS -1 regulates mitochondrial UPR activation. Science 337, 587– 590 (2012). \nhttps://doi.org/10.1126/science.1223560 \n59 Derisbourg, M. J., Hartman, M. D. & Denzel, M. S. Perspective: Modulating the integrated stress response \nto slow aging and ameliorate age -related pathology. Nat Aging 1, 760– 768 (2021). \nhttps://doi.org/10.1038/s43587-021-00112-9 \n60 Clay, K. J., Yang, Y ., Clark, C. & Petrascheck, M. Proteostasis is differentially modulated by inhibition of \ntranslation initiation or elongation. eLife 12, e76465 (2023). https://doi.org/10.7554/eLife.76465 \n61 Seo, K. et al. Heat shock factor 1 mediates the longevity conferred by inhibition of TOR and insulin/IGF-1 \nsignaling pathways in C. elegans. Aging cell 12, 1073–1081 (2013). https://doi.org/10.1111/acel.12140 \n62 Steinkraus, K. A. et al. Dietary restriction suppresses proteotoxicity and enhances longevity by an hsf-1-\ndependent mechanism in Caenorhabditis elegans. Aging cell 7, 394– 404 (2008). \nhttps://doi.org/10.1111/j.1474-9726.2008.00385.x \n63 Hsu, A. L., Murphy , C. T . & Kenyon, C. Regulation of aging and age-related disease by DAF-16 and heat-\nshock factor. Science 300, 1142–1145 (2003). https://doi.org/10.1126/science.1083701 \n64 Kumsta, C. & Hansen, M. Hormetic heat shock and HSF-1 overexpression improve C. elegans survival and \nproteostasis by inducing autophagy. Autophagy 13, 1076– 1077 (2017). \nhttps://doi.org/10.1080/15548627.2017.1299313 \n65 Howard, A. C., Rollins, J., Snow, S., Castor, S. & Rogers, A. N. Reducing translation through eIF4G/IFG -1 \nimproves survival under ER stress that depends on heat shock factor HSF -1 in Caenorhabditis elegans. \nAging cell 15, 1027–1038 (2016). https://doi.org/10.1111/acel.12516 \n66 Yoneda, T. et al.  Compartment-specific perturbation of protein handling activates genes encoding \nmitochondrial chaperones. Journal of Cell Science  117, 4055 –4066 (2004). \nhttps://doi.org/10.1242/jcs.01275 \n67 Rollins, J., Lind, N. & Rogers, A. N. Loss of eif-2alpha phosphorylation on S49 (mammalian S51) associated \nwith the integrated stress response hastens development in C. elegans. MicroPubl Biol 2017 (2017). \nhttps://doi.org/10.17912/w2bm1s \n68 Kulalert, W., Sadeeshkumar, H., Zhang, Y . K., Schroeder, F. C. & Kim, D. H. Molecular Determinants of the \nRegulation of Development and Metabolism by Neuronal eIF2 α Phosphorylation in Caenorhabditis \nelegans. Genetics 206, 251–263 (2017). https://doi.org/10.1534/genetics.117.200568 \n69 Boyce, M. et al. A selective inhibitor of eIF2alpha dephosphorylation protects cells from ER stress. Science \n307, 935–939 (2005). https://doi.org/10.1126/science.1101902 \n70 Krukowski, K. et al. Small molecule cognitive enhancer reverses age-related memory decline in mice. Elife \n9 (2020). https://doi.org/10.7554/eLife.62048 \n71 Derisbourg, M. J., Wester, L. E., Baddi, R. & Denzel, M. S. Mutagenesis screen uncovers lifespan extension \nthrough integrated stress response inhibition without reduced mRNA translation. Nat Commun 12, 1678 \n(2021). https://doi.org/10.1038/s41467-021-21743-x \n72 Xie, Y . et al. Activity-dependent synthesis of Emerin gates neuronal plasticity by regulating proteostasis. \nCell Reports 44 (2025). https://doi.org/10.1016/j.celrep.2025.115439 \n73 Yang, W. S. & Stockwell, B. R. Synthetic lethal screening identifies compounds activating iron-dependent, \nnonapoptotic cell death in oncogenic -RAS-harboring cancer cells. Chem Biol 15, 234– 245 (2008). \nhttps://doi.org/10.1016/j.chembiol.2008.02.010 \n74 Szaruga, M.  et al.  Activation of the integrated stress response by inhibitors of its kinases. Nature \nCommunications 14, 5535 (2023). https://doi.org/10.1038/s41467-023-40823-8 \n75 Verduzco Espinoza, A. P . et al. Microglia-to-neuron signaling links APOE4 and inflammation to enhanced \nneuronal lipid metabolism and network activity. Proceedings of the National Academy of Sciences 122, \ne2516103122 (2025). https://doi.org/10.1073/pnas.2516103122 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\n76 Lombard, D. B.  et al.  High-throughput small molecule screening reveals Nrf2 -dependent and -\nindependent pathways of cellular stress resistance. Science Advances 6, eaaz7628 (2020). \nhttps://doi.org/doi:10.1126/sciadv.aaz7628 \n77 Anisimova, A. S., Alexandrov, A. I., Makarova, N. E., Gladyshev, V. N. & Dmitriev, S. E. Protein synthesis \nand quality control in aging. Aging 10, 4269–4288 (2018). https://doi.org/10.18632/aging.101721 \n78 Marks, J., Young, E. & Hafner, M. Determining the off-target activity of antibiotics and novel translation \ninitiation sites in mitochondria. Elife (2024). https://doi.org/10.7554/elife.103699.1 \n79 Almeida, L.  et al.  Ribosome-Targeting Antibiotics Impair T Cell Effector Function and Ameliorate \nAutoimmunity by Blocking Mitochondrial Protein Synthesis. Immunity 54, 68– 83.e66 (2021). \nhttps://doi.org/10.1016/j.immuni.2020.11.001 \n80 Greenberger, N. J. Inhibition of Protein Synthesis in Rat Intestinal Slices by Tetracycline. Nature 214, 702–\n703 (1967). https://doi.org/10.1038/214702b0 \n81 Partridge, L., Fuentealba, M. & Kennedy, B. K. The quest to slow ageing through drug discovery. Nature \nReviews Drug Discovery 19, 513–532 (2020). https://doi.org/10.1038/s41573-020-0067-7 \n82 Ghosh, T. S., Shanahan, F. & O’Toole, P . W. The gut microbiome as a modulator of healthy ageing. Nature \nReviews Gastroenterology & Hepatology  19, 565– 584 (2022). https://doi.org/10.1038/s41575-022-\n00605-x \n83 Fura, A. Role of pharmacologically active metabolites in drug discovery and development. Drug Discovery \nToday 11, 133–142 (2006). https://doi.org/https://doi.org/10.1016/S1359-6446(05)03681-0 \n \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nExperimental Models \n \nMouse studies \nAll procedures are approved by TSRI IACUC committee. C57BL/6 female mice (Stock #000664) were purchased from \nJackson Labs at 9 weeks of age. Mice were group-housed and maintained on a 12 hrs light/dark cycle with ad libitum access \nto food and water. \n \nC. elegans strains \nThe Bristol strain (N2) was used as the wild -type strain. In addition, the following worm strains used in this study were \nobtained from the Caenorhabditis Genetics Center (CGC; Minneapolis, MN, USA): SJ4100 [zcIs13 [hsp- 6p::GFP ]], CL2070 \n[dvIs70 [hsp-16.2p::GFP + rol-6(su1006) ]], PS3551 [ hsf-1(sy441)], RB790 [atf-4(ok576)], ZD1866 [ eIF2a(qd338)], ANR165 \n[eIF2a(rog3)]. Strains were backcrossed at least three times prior to experimental analysis. \n \nCell Culture \nHEK293 cells (ATCC CRL -1573) were directly obtained from ATCC. HT22 neuronal cells were obtained as a gift from Dr. \nPamela Maher. Cells were grown in Dulbecco’s Modified Eagle Medium (DMEM, ThermoFisher, Cat#11995065) \nsupplemented with 10% fetal bovine serum (FBS, ThermoFisher, Cat#16000044) and 1% penicillin-streptomycin (P/S, \nThermoFisher, Cat#15140-122). Cells were subcultured every two  o r  t h r e e  d a y s  a t  3 7  ° C  w i t h  5 %  C O\n2 in a humidified \nincubator. Mycoplasma testing was performed every 6 months through TSRI.  \n \nPrimary neurons were harvested as previously described in Chassefeyre et al1. Briefly, the hippocampi or cortices of male \nand female C57BL/6 P0- P1 mouse pups were dissected in Hanks’ balanced salt solution (HBSS) (ThermoFisher, \nCat#24020117) supplemented with 0.08% d -glucose (Sigma -Aldrich, Cat#G6152), 0.17% HEPES (Sigma -Aldrich, \nCat#H7006), and 1% P/S; filter-sterilized; and adjusted to pH 7.3. Dissected brain tissues were washed twice with cold HBSS \nand individually incubated at 37°C for 15 to 20 minutes in a sterile solution of 45 U of papain (Worthington, Cat#LS003119), \n0.01% deoxyribonuclease (DNase) (Sigma -Aldrich, Cat#D4527), 1 mg of DL-cysteine (Sigma -Aldrich, Cat#C9768), 1 mg of \nbovine serum albumin (BSA) (Sigma -Aldrich, Cat#A7906), and 25 mg of D -glucose in phosphate- buffered saline (PBS) \n(ThermoFisher, Cat#10010049). Tissues were washed twice with DMEM supplemented with 10% FBS (preheated to 37°C) \nand disrupted by 10 to 12 cycles of aspiration through a micropipette tip. Dissociated neurons were th en resuspended in \nwarm DMEM supplemented with 10% FBS, counted and plated in either 96-well plates (Millipore Sigma, Cat#CLS3603) or \n10 cm dishes (Genesee Scientific, Cat#25 -202) pretreated with poly -L-lysine (50 μ g/mL) (Sigma -Aldrich, Cat#P5899) in \nborate buffer [1.24 g of boric acid (Thermo Fisher Scientific, Cat#BP168-1) and 1.90 g of borax (Sigma-Aldrich, Cat#B9876) \nin 500 mL of cell culture–grade water, adjusted to pH 8.5, and filter-sterilized]. Plating densities were 2 x 10\n4 cells per well \nor 7.75 x 10 6 cells per dish. After 3 hrs, medium was replaced with Neurobasal-A media (ThermoFisher, Cat#10888022), \nsupplemented with 2% antioxidant free (ThermoFisher, Cat#10889038) or complete (ThermoFisher, Cat#17504044) B27 \nand 0.25% GlutaMAX (ThermoFisher, Cat#35 050061). Cultures were maintained in culture media at 37 °C with 5% CO\n2. \nFour days after plating (DIV4), 5 -Fluoro-2’-deoxyuridine (FUDR, Millipore Sigma, Cat#F0503) was added to a final \nconcentration of 5 µM in culture media.  \n \nInduced neurons (iNeurons) were previously generated in Verduzco Espinoza et al\n2. In short, iNeurons were generated \nfrom APOE3/E3 (HE463#6) iPSCs, which is available on WiCell in a collection called Topol Lab’s Next Gen Cell Lines with the \naccession number of SCRP2307i. iPSCs were thawed in mTeSR Plus with 10 µM Y -27632 (Stem Cell Technologies, \nCat#72304) and seeded on 6-well dishes coated with Matrigel. For maintenance, iPSCs were fed every 1-3 days, depending \non confluence, and passaged 1 -2 times per week  as small clusters using 1 µM EDTA (Invitrogen, Cat#15575-020). 1X P/S \nwas added to all culture media. Neurons were generated from iPSCs via direct induction with NGN2. iPSCs were dissociated \nwith accutase (Stem Cell Technologies, Cat#7920) and seeded on Matrigel-coated 6 -well plates at 1.5 x 105 cells per well \nin mTeSR Plus with 10 µM Y-27632. The next day (day 0), each well was transduced with 125 µL of tetO-NGN2 and RTTA \nlentivirus for 2 hrs in mTeSR Plus with 10 µM Y-27632 at 37 °C. After 2 hrs, the viral media was aspirated and replaced with \nmTeSR Plus with 2 µg/mL doxycycline (Stem Cell Technologies, Cat#72742) to start NGN2 expression. On day 1, cells were \nfed with half mTeSR plus half Neuronal media [Neurobasal A, 1X B27, NEAA, Glutamax (ThermoFisher, Cat#10888022, \n17504044, 350500661, 11140050] with doxycycl ine. NGN2-expressing cells were selected on day 2 by media exchange \nwith doxycycline and puromycin (Gibco, Cat#A11138-03). On day 3, cells were replated at 6.25 x 10\n4 cells per cm2 on plates \ncoated with PDL and Matrigel. 40 nM BRDU was added to select against the remaining dividing iPSCs. Cells underwent a \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nfull media change on days 7 and 10 to remove doxycycline and add neuronal differentiation factors (BDNF, GDNF, NT3 at \n10 ng/mL, and laminin at 1 μg/mL). Experiments in this manuscript begin on day 14 for FUNCAT and day 24 for ferroptosis \ncell survival. For ferroptosis cell survival experiment, cells underwent half-media change on days 14 and 17.  \n \nCompounds \nTetracyclines: Minocycline (MP Biomedicals, Cat#155718), 4 -epiminocycline (Toronto Research Chemicals, Cat#TRC -\nE588540), Doxycycline (Sigma, Cat#D9891), Col-3 (Incyclinide, MedChemExpress, Cat#HY-13648), R464449 (Sigma-Aldrich, \nCat#R464449), 12-aminominocy cline (Toronto Research Chemicals, Cat#TRC-A618285), Tigecycline (LKT Laboratories, \nCat#T3324). Other translation inhibitors: Cycloheximide (CHX, EMD Millipore, Cat#239763), Tunicamycin (Tm, Cayman \nChemical, Cat#11445), Thapsagargin (Thaps., MedChemExpress, Cat#HY-13433). Following the initial hit, 4-epiminocycline \nand 12-aminominocycline were synthesized from minocycline to confirm purity and compound identity. \n \n4-epiminocycline: \n1H NMR (400 MHz, DMSO-d6): δ 15.06 (s, 1H), 11.48 (s, 1H), 9.48-9.29 (m, 2H), 7.49-7.40 (m, 1H), 7.38 \n(s, 1H), 6.85-6.80 (m, 2H), 4.71 (s, 1H), 3.05-2.65 (m, 8H), 2.65-2.750 (m, 2H), 2.20-2.00 (m, 1H), 1.55-1.41 (m, 1H). LCMS: \ncalc. for C23H27N3O7: 457.18, found: [M+H]+ 458.1. HPLC: 99.2% (254 nm). SFC: 97.7%. \n \n12-aminominocycline: 1H NMR (400 MHz, DMSO-d6): δ 13.17 - 13.03 (m, 1H), 10.72 - 10.13 (m, 1H), 9.69 - 9.46 (m, 1H), \n9.10 - 8.44 (m, 1H), 7.16 (d, J = 8.8 Hz, 1H), 6.76 (d, J = 8.8 Hz, 1H), 6.46 - 6.21 (m, 1H), 5.95 (br s, 1H), 3.23 (dd, J = 4.0, 15.2 \nHz, 1H), 2.98 (d, J = 1.6 Hz, 1H), 2.93 - 2.74 (m, 1H), 2.65 (br t, J = 2.4 Hz, 1H), 2.62 (t, J = 2.4 Hz, 1H), 2.57 (s, 6H), 2.47 (s, \n4H), 2.39 (br s, 2H), 2.09 (br t, J = 14.4 Hz, 1H), 1.93 (br dd, J = 5.2, 7.6 Hz, 1H), 1.46 - 1.35 (m, 1H). LCMS: calc. for C22H28N4O6: \n456.20, found: [M+H]+ 457.2. HPLC: 96.4% (254 nm). \n \nMethod Details \n \nLifespan assay \nAge-synchronized C. elegans were prepared in liquid medium [S -complete medium with 50 mg/mL carbenicillin and 0.1 \nmg/mL fungizone in flat-bottom, optically clear 96-well plates (Corning, Cat#351172) containing 150 µL total volume per \nwell, as previously described.3 Plates contained ~10 animals per well in 6 mg/mL OP50. All experiments were performed \nwith X-ray-irradiated OP50. Age-synchronized animals were seeded as L1 larvae and grown at 20 °C. Plates were covered \nwith sealers to prevent evaporation. To prevent self-fertilization, FUDR (0.12 mM final) was added 42 – 45 hr. after seeding. \nDrugs were added on the days indicated and survival was scored manually by visualizing worm movement using an inverted \nmicroscope 3x/ week. When used, DMSO was kept to a final concentration of 0.33% v/v. Statistical analysis was performed \nusing the Mantel–Haenszel version of the log-rank test as outlined in Petrascheck and Miller\n4. \n \nO-propargyl puromycin translation \nHEK293 (6 x 10\n5) or HT22 (5 x 103) cells were seeded into black 96-well plates (Costar, Cat#3603) with complete media (100 \nµL/well) and incubated overnight. For drug treatment, drugs were prepared at 20 mM stock in 100% DMSO. This was \nfurther diluted with 12.5% DMSO/PBS to make 12X working solutions at the indicated concentrations. 4 µL were added for \na final concentration of 0.5% DMSO/PBS in each well. 0.5% DMSO/PBS served as a control. Drugs and control were added \nin triplicate and incubated for 2 hrs. Protein synthesis rates were determined using EZClick Global Protein Synthesis Assay \nKit (BioVision, Cat#76305 -296) according to the manufacturer’s protocol. In short, nascent polypeptides were pulse-\nlabeled by addition of 1 µL 100X “protein label” and incubated for 30 minutes at 37 °C. We included a 1 µL addition of \nDMSO as a “no -protein label” control to subtract background fluorescence. Then culture media was removed and cells \nwere washed once with 100 µL PBS. Cells were fixed by addition of 100 µL “fixative solution” and incubated for 15 minutes \non ice in the dark. This was aspirated off and 100 µL “permeabilization buffer” was added and incubated for 10 minutes at \nroom temperature. Buffer was removed and 20 µL fresh “permeabilization buffer” was added. The 1X EZclick reaction \ncocktail was prepared according to the protocol: 3 µL PBS, 1 µL copper reagent, 1 µL fluorescent azide, 5 µL reducing agent, \nscaled for number of reactions needed. 100 µL 1X EZclick reaction cocktail was added to each well and incubated. After 30 \nminutes, the reaction mixture was aspirated and cell were washed three times with 100 µL “wash buffer.” Wells were then \nincubated with 1X DAPI stain for 15 minutes and washed three times with “wash buffer”. As a positive control, 0.5 µM \ncycloheximide (CHX) was added 30 minutes prior to addition of “protein label.” When thapsagargin was used, it was added \nat the same time as the tetracyclines at 250 nM. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nEach well was imaged using the ImageExpress HT.ai confocal high-content imaging system (IXM, Molecular Devices). A 20X \nApo LWD 9na 0.95 water immersion objective was used with a confocal 50 µm slit setting. Laser power was set to 2% and \neach image frame was exposed for 100ms for GFP channel and 10ms for DAPI channel. Each well was imaged in 4 center \nlocations. Images were analyzed using a custom Cell Profiler pipeline that autodetects an ROI b ased on DAPI staining of \nnuclei, then expands ROI by 20 pixels, and then measures GFP signal. Measuring GFP/DAPI signal gives an approximation \nof total protein synthesis relative to total cells seeded in each well. \n \nOxytosis Ferroptosis survival assay— HT22 cells \n4 x 10\n3 cells were seeded into sterile black 96-well plates with complete media and incubated overnight (100 µL/well). For \ndrug treatment, drugs were prepared at 10 mM stock in 100% DMSO. This was further diluted with 5% DMSO/PBS to make \n25X working solutions. 4 µL were added for a final concentration of 0.5% DMSO in each well. 0.2% DMSO served as a \nc o n t r o l .  D r u g s  a n d  c o n t r o l  w e r e  a d d e d  2  h r s  p r i o r  t o  a d d i ti o n  o f  o x y t o s i s / f e r r o p t o s i s  i n d u c e r s .  5  m M  g l u t a m a t e  \n(ThermoFisher, Cat#156212500) was added to each w ell and cell viability was measured using CellTiter -Glo (Promega, \nCat#G9243). Luminescence was measured using BioTek Cytation 5 (Agilent). Results are shown as a percentage of \nuntreated control cells. In experiments where ISRIB was used to block the ISR, 3 00 nM ISRIB was added 2 hrs prior to the \naddition of drugs/DMSO controls. When thapsagargin was used, it was added at the same time as the tetracyclines at 250 \nnM. \n \nOxytosis Ferroptosis survival assay— Primary and iNeurons \nFor primary neurons: Hippocampal primary neurons were prepared in sterile black 96-well plates coated with borate buffer \nas described above. At DIV7, drugs were added at increasing concentrations to each well and incubated for 4 hrs. Then, \n300 nM RSL3 (Selleck Chem, Cat#S8155) was added to each well and incubated for 24 hrs. Viability was determined using \nCellTiter-Glo as above. When ISRIB was used, 300 nM was added to each well and incubated for 4 hrs prior to drug/control \ntreatment. \n \nFor iNeurons: On day 3, 24 hrs after puromycin selection, cells from a 6-well plate were replated equally into 96-well plates, \nat 2 x 10\n4, leaving the edge wells empty. 40 nM BRDU was added to select against the remaining dividing iPSCs. The media \nwas changed on days 7 and 10 with neuronal differentiation factors (BDNF, GDNF, NT3, Laminin). Cells were maintained \nwith half media changes twice a week and were then allowed to grow until day 24. Then 0, 0.14, 0.41, 1.2, 3.7, 11, 33, or \n100 µM 4-epiminocycline was added to 6 wells of the 96 -well plate. After 2 hrs, 2 µM RSL3 was added to each well to \ninduce ferroptosis. 24 hrs later, viability was determined using CellTiter Glo as above. \n \nSurface Sensing of Translation (SUNSET) \nFor C. elegans: Day 1 adult N2 worms were bleached, and eggs were allowed to hatch in S-complete by shaking overnight. \nOn the next day , 10,000 L1 worms were seeded in a 10 cm plate containing a total volume of 20 mL S -complete with 6 \nmg/mL OP50 bacteria, 50 μg/mL carbenicillin, and 0.1 μ g/mL amphotericin B. 4 mL FUDR (0.6 mM stock in S -complete) \nwere added to worms at L4 stage in each plate. 2 hrs later, 100 μM of each compound was added to worms. After 12 hrs, \nworms were transferred into a 15 mL corning tu be containing a total volume of 4 mL S -complete with 500 μL 6 mg/mL \nOP50 bacteria, 0.5 mg/mL puromycin (ThermoFisher, Cat#A11138-03), and 100 μM tetracyclines. After rotating the corning \ntubes for 4 hrs, worms were collected into 2 mL cryotubes by washing them with M9 once and with cold PBS three times. \nWorms were flash-frozen in liquid nitrogen and 150 µL of cold lysis buffer [20 mM Tris base, 100 mM NaCl, 1 mM MgCl\n2, \npH = 7.4, with protease inhibitors (Roche, Cat#11836153001)] was added and samples subseq uently broken with a beak \nmill homogenizer (Fisher). Protein concentrations were determined by the Bradford protein assay (Bio-Rad, Cat#5000006). \n50 µg protein from each sample was loaded for western blot analysis using antibodies against puromycin (Millipore, \nCat#MABE343) and GAPDH (Proteintech, Cat#10494-1-AP). Primary antibodies were diluted 1:5,000 in 5% non-fat milk in \nTBST, and secondary antibodies were diluted 1:10,000. To determine the relative intensities of each blot, the integrated \nintensity was measured for each full anti -puromycin lane using ImageJ. A similar sized band with no signal was used to \ncalculate and subtract background, and then each intensity was normalized to corresponding GAPDH loading control. \nFinally, each condition was normalized to DMSO control for quantification and statistics. \n \nFor primary neurons: Cortical primary neurons were prepared in 10 cm dishes as described above. On DIV7, dishes were \ntreated with 300 nM ISRIB or DMSO. Immediately after, tunicamycin was added to a final concentration of 3 ug/mL. Either \n30 minutes or 18 hr s after tunicamycin addition, puromycin was added to a final concentration of 2ug/mL. 2 hrs after \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\npuromycin addition, total protein was extracted with RIPA buffer [150 mM NaCl (Sigma -Aldrich, Cat#S7653), 5 mM EDTA \n(Applichem, Cat#A1104), 50 mM Tris (ThermoFisher, Cat#BP152-5), 1% NP-40 (Millipore Sigma, Cat#492016), 0.5% sodium \ndeoxycholate (ThermoFisher, Cat#BP349-100), and 0.1% SDS] supplemented with protease and phosphatase inhibitor \ncocktail (ThermoFisher, Cat#78442). Protein concentrations were determined by the Bradford protein assay. 15 μg from \neach sample was loaded for western blot analysis using antibodies against puromycin diluted 1:1000 in 5% non-fat milk in \nTBST. Relative intensities were determined as described above. Total protein was visualized with Coomassie stain solution \n(0.1% Coomassie Brilliant Blue R stain (ICN Biomedicals, Cat#02190343-CF), 7% acetic acid, 50% methanol). \n \nAntibiotic activity assay  \nOne colony of OP50 was inoculated overnight in a 37 °C shaker in 5 mL LB media containing 50 μg/mL ampicillin. 14-16 hrs \nlater, the pre-inoculum was diluted 1:200 in 5 mL LB containing 50 μ g/mL ampicillin. In a flat bottom, clear 96 -well plate, \n100 μL of t he diluted inoculum was added to each well being tested, one column (8 wells) per condition. DMSO or the \nindicated drugs were added and a 0 time point was recorded for baseline OD\n600 values using a plate reader (Tecan Safire \nII). The plate was returned to the 37°C shaker and OD600 was re-measured every hr. OD600 values for each time point were \naveraged across each well per condition. \n \nZn\n2+ chelation assay \nChelation assay was preformed using the Abchem Zinc Assay Kit (Abcham, Cat#ab102507). First, the included zinc standard \nwas added to each well according to the manufacturers protocol. In a clear 96-well plate, samples were preincubated with \n0, 100, 300, 780, 1200, 1800 μM of each tetracycline in duplicate. Samples were allowed to incubate for 30 minutes before \ndetermination of free Zn\n2+. . Absorbance was measured at OD 560nm on the T ecan Safire II. EDTA was included as a positive \ncontrol and data was normalized with 0% remaining free Zn2+ defined by assay buffer with no zinc standard, and 100% free \nzinc being the DMSO control. \n \nMMP9 Inhibition \nMMP9 inhibition was determined using a quenched fluorogenic peptide included in the MMP9 Inhibitor Screening Assay \nKit following the manufacturer’s protocol (Abcham, Cat#ab139449). N-hydroxy -2-phenylethanamide (NNGH, 1 μ M) was \nincluded as a positive control. In the screening assay, 100 μM of each tetracycline in triplicate was added to a well of a 96-\nwell plate containing MMP enzyme and assay buffer and allowed to incubate for 10 minutes. MMP9 substrate was then \nadded to each well using a multi -channel pipet, the plate was incubated for 60 minutes at 37 °C, and fluorescence was \nmeasured using the BioTek Cytation 5. An identical procedure was followed for the generation of dose response curves for \n12-aminominocycline, 4-epiminocycline, minocycline, and tigecy cline except concentrations were: 0, 1, 10, 30, 100, 300 \nμM. 4 technical replicates were used for each concentration and assay was repeated twice. \n \nUPR\nMT stress imaging \n20,000 SJ4100 [zcIs13 [hsp- 6p::GFP ]], age synchronized animals were prepared as in the SUNSET method. At the \ndevelopmental stage of L2, 100 μM of each tetracycline or DMSO was added and allowed to incubate for 12 hrs. After the \ntreatment window, animals were washed 3x with S-complete to remove any bacteria and suspended in 10 mL S-complete. \nGFP intensity was quantified using the COPAS Biosorter (Union Biometric). Integral values for GFP channel were used to \ndetermine fluorescent values. The animals were also sorted into 96 -well plates and imaged using an ImageXpress Micro \nXL High-Content screening system (Molecular Devices) with a 2x objective. \n \nHSR stress imaging \n20,000 CL2070 [dvIs70 [hsp-16.2p::GFP+rol-6(su1006)]] animals were prepared as in the SUNSET method. 100 μM of each \ntetracycline or DMSO was added to liquid culture at the late L4 stage. On day 1 of adulthood, after 12 hrs treatment, the \nliquid culture was transferred to a 50 mL falcon tube and worms were se ttled by gravity. The supernatant was aspirated \noff leaving the concentrated worm pellet. The pellet was washed twice with S-complete, pelleted by gravity again, and \nthen transferred to 10 cm NGM plate. Once the liquid was completely dry on the NGM plate, the plates were transferred \nto a 36 °C incubator, upside down, and incubated for 1.5 hrs. Following heat shock, NGM plates were transferred to 20 °C \nincubator and allowed to recover for 8 hrs. Worm GFP intensity was determined as in the hsp-6::GFP  reporter using the \nCOPAS Biosorter, but instead of being sorted into 96-well plates, 5,000 animals were bulk-sorted into 2 mL cryotubes and \nflash frozen for RT-qPCR analysis.  \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nQuantitative real-time PCR (qRT-PCR) and data analysis \nAll qRT-PCR experiments were conducted according to the MIQE guidelines, except that samples were not tested in a bio-\nanalyzer, but photometrically quantified using a Nanodrop. Worm pellets were obtained from the HSR stress imaging \nexperiment above. To extract RNA, frozen animals were suspended in ice-cold Trizol (Qiagen, Cat#79306), then zirconium \nbeads were added and animals were broken open with a beak mill homogenizer (Fisherbrand). Following chloroform \nextraction, RNA was precipitated using isopropanol and washed once with 75% ethanol followed by DNAse (Sigma-Aldrich, \nCat#AMPD1-1KT) treatment. Reverse transcription was carried out using iScript RT-Supermix (Bio-Rad, Cat#170–8841) at \n42°C for 30 minutes. Quantitative PCR reactions were set up in 384 -well plates (Bio-Rad, Cat#HSP3901), which included \n2.5 µl Bio-Rad SsoAdvanced SYBR Green Supermix (Cat#172–5264), 1 µl cDNA template (2.5 ng/µl, to final of 0.5 ng/µl in \n5 µl PCR reaction), 1 µl water , and 0.5 µl of forward and reverse primers (150 nM final co ncentration). Quantitative PCR \nwas carried out using a Bio-Rad CFX384 Real-Time thermocycler (95°C, 3 minutes; 40 cycles of 95°C 10 s, 60°C 30 s; Melting \ncurve: 95°C 5 s, 60– 95°C at 0.5°C increment, 10 s). Gene expression was normalized to three reference genes for  C. \nelegans samples, act-1, xpg-1 and rpl-6.  \n \nFluorescent Non-canonical Amino Acid Tagging (FUNCAT) \nFor iNeurons: On day 3, 24 hrs after puromycin selection, cells from a 6-well plate were replated equally into 6-well palates, \nat 6 x 10\n5 . on day 14, 0, 10, 30, or 100 µM 4-epiminocycline was added to each well of a 6 well plate (3 mL media). 250 \nnM CHX was added as a positive control in one of the wells. Drugs were prepared from 30 mM stock solutions in 100% \nDMSO. Working solutions were prepared at 300X in 15% DMSO and 20 µL was added to each well for a final concentration \nof 0.1% DMSO and allowed to incu bate for 2 hrs. 200 mM azidohomoalanine (AHA, Vector Laboratories, Cat#CCT -1066-\n1000) stock solution was prepared with Ultrapure H2O and brought to pH = 7 with 10N NaOH. 200 mM AHA was diluted in \ncomplete culture media pre-warmed to 37 °C for a final concentration of 8 mM. Media was replaced with AHA and pulse \nlabeled for 2 hrs after which labeling was stopped by removal of AHA media. \n \niNeurons were washed once 1X DBPS and then lysed in dish with the addition of 100 µL 10% RIPA/DPBS buffer + 1X \nprotease inhibitors (Pierce Protease Inhibitor Mini Tablets, EDTA -free, ThermoFisher, Cat#A32955) on ice. Cells were \nscrapped into an Eppendorf tube and centrifuged at 5000 x g for 2 minutes to remove debris. Lysate was collecte d and \nprotein concentration was determined by Pierce BCA assay kit (ThermoFisher, Cat#23235) following manufacturer’s \ninstructions. \n \nCell lysates were adjusted to 1.0 mg/mL using cold DPBS. To each sample (50 μL) in a PCR Eppendorf tube, 6 μL of freshly \nprepared click reaction mixture was added. The cocktail was prepared by addition of 3 μL THPTA (1.7 mM in H 2O), 1 μL \nCuSO4 (50 mM in H2O), 1 μL TAMARA-azide (1.25 mM in DMSO),and 1 μ L ascorbate (50 mM in H2O, prepared last). After \naddition of the mixture, each reaction was vortex, briefly centrifuged, and incubated at 40 °C for 1 hr . The reaction was \nquenched by addition of addition of 17 μL of 4X SDS loading buffer (Bio -Rad, Cat#1610747), and heated at 95 ° C for 5 \nminutes. 30 μg protein was loaded per lane of a polyacrylamide gel (Bio -Rad, Cat#4569034) and visualized by in-gel \nfluorescence on a ChemiDoc MP flatbed fluorescence scanner (Bio -Rad, Cat#12003154). After imaging, gels were stained \nwith Coomassie blue to determined total protein. Gels were quantified in ImageJ by dividing the integrated intensity of \neach lane of the fluorescent image by the integrated intensity of the corresponding total protein lane after subtracting a \nsimilar area background lane,  then normalizing to the DMSO control. Abbreviations: THPTA ((tris -\nhydroxypropyltriazolylmethyl)amine), Combi-blocks, Cat#QH-3278) CuSO\n4, (Copper(II) sulfate, Sigma-Aldrich, Cat#C1297), \nAzide-fluor 545 (5-carboxytetramethylrhodamine-azide, Millipore Sigma,  Cat#760757), and ascorbate ((+) Sodium L-\nascorbate, Sigma-Aldrich, Cat#A7631). \n \nFor in vivo translation: We utilized the in vivo translation measurement method developed in Xie et al\n5 but adapted for gel-\nbased FUNCAT. 10-week old male C57BL6 were treated with either 4 -epiminocycline or control in drinking water (0.6 \nmg/mL) for 3 days (n = 6 per group). This dose was selected because it is well tolerated in mice and we determined a \nplasma level of approximately 7 μM for the analog minocycline, which is comparable to circulating levels in human patients \nof 2–11 μM6,7. On the 4th day, 100 mg/kg 4-epiminocycline or control (saline) was administered via intraperitoneal (i.p.) \ninjection. 30 minutes later, AHA (dissolved in PBS) was i.p. injected at 50 mg/kg. Again, 30 minutes later 50 mg/kg \npentylenetetrazol (PTZ, dissolved in PBS) was i.p. injected. 90 minutes later the animals were sacrificed, PBS-perfused, the \nhippocampus was harvested, and samples snap frozen in liquid nitrogen. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint \n\nSamples were resuspended in 150 μL 1x DPBS and homogenized with a hand -held tip sonicator (Fisher Scientific, sonic \ndismembrator model 100). Protein concentration of the homogenate was determined by BCA protein assay kit. Protein \nlysate with AHA-labeled nascent proteins was normalized by total protein to 2.0 mg/mL and the click reaction with TAMRA-\nazide was performed as described above. \n \nRat pharmacokinetic determination \nPharmacokinetic testing for 4-epiminocycline was conducted by Pharmaron. 4-epiminocycline was formulated in PBS and \nwas i.p. injected at a low dose of 25 mg/kg and high dose of 50 mg/kg. A total of 3/3 male rats were assessed at 0.083, \n0.25, 0.5, 1, 2, 4, and 8 hr timepoints for blood. At the 8 hr timepoint, the mice were sacrificed after blood collection, and \nbrains were weighed then homogenized. Analytes from each collection were extracted and run on an LC -MS/MS (AB API \n55000) to quantify compound concent ration. Data were collected as plasma (ng/mL) and brain (ng/g) concentrations \ndetermined by an internal standard method. \n \nReferences \n \n1 Chassefeyre, R. et al. Endosomal sorting drives the formation of axonal prion protein endoggresomes. Science \nAdvances 7, eabg3693 (2021). https://doi.org/doi:10.1126/sciadv.abg3693 \n2 Verduzco Espinoza, A. P . et al. Microglia-to-neuron signaling links APOE4 and inflammation to enhanced neuronal \nlipid metabolism and network activity. Proceedings of the National Academy of Sciences 122, e2516103122 (2025). \nhttps://doi.org/10.1073/pnas.2516103122 \n3 Clay, K. J. & Petrascheck, M. in Aging: Methods and Protocols   (ed Sean P . Curran)  77–89 (Springer US, 2020). \n4 Petrascheck, M. & Miller, D. L. Computational Analysis of Lifespan Experiment Reproducibility. Front Genet 8, 92 \n(2017). https://doi.org/10.3389/fgene.2017.00092 \n5 Xie, Y . et al.  Activity-dependent synthesis of Emerin gates neuronal plasticity by regulating proteostasis. Cell \nReports 44 (2025). https://doi.org/10.1016/j.celrep.2025.115439 \n6 Perez-Gomez, A.  et al.  A phenotypic Caenorhabditis elegans screen identifies a selective suppressor of \nantipsychotic-induced hyperphagia. Nature Communications  9 (2018). https://doi.org/10.1038/s41467-018-\n07684-y \n7 Agwuh, K. N. & MacGowan, A. Pharmacokinetics and pharmacodynamics of the tetracyclines including \nglycylcyclines. J Antimicrob Chemother 58, 256–265 (2006). https://doi.org/10.1093/jac/dkl224 \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 11, 2026. ; https://doi.org/10.64898/2026.01.09.698733doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}