By Looped Hook and Targeting Crook: Potential Game-Changer for "Allotopic" Mitochondrial RNA

Posted by Michael Rae on March 15, 2012 | Chief Science Officer's Team

UCLA Researchers have exploited a recently-discovered mammalian system for the mitochondrial import of nuclear-encoded RNA to import, express, and demonstrate functional protein translation from engineered mRNA and tRNA constructs. They used this system, with modifications for mitochondrial targeting and orthotopic translation, to rescue respiration in human mitochondriopathy cells. While further characterization and extension is clearly needed, this approach appears offer great promise for the correction of age-related mitochondrial DNA mutations.

In the summer of 2010, we posted an entry about a significant advance in basic mitochondrial biology that held within it a tantalizing promise: a potentially new approach to obviating the problem of age-related accumulation of mutations in mitochondrial DNA (mtDNA), which are widely suspected to play an important role in the age-related rise in oxidative stress, Parkinson's disease, and age-related muscle degeneration. The report (1)-- from a group led by UCLA's Dr. Michael Teitell, and including graduate student Carla Koehler, Newcastle University's Dr. Robert Lightowlers, and others -- identified and provided initial characterization of polynucleotide phosphorylase (PNPase) as a mammalian-specific system for the mitochondrial import of nuclear-encoded RNA. which appeared to offer some potential as the linchpin of a novel route to overcoming this biomedical challenge.

From that initial report,(1) it appeared that Teitell's group were focused on further basic biology questions, such as "what other pathway components are involved and what the RNA sequence or structure rules tell us about how PNPASE may decipher between processing and import." But at his presentation at the fifth Strategies for Engineered Negligible Senescence (SENS) biomedical conference, Teitell  revealed that their focus had shifted: "our studies show an unanticipated role for PNPASE in mediating the translocation of RNAs into mitochondria and provide a potential therapeutic route for halting or reversing the decline in mitochondrial function with aging [emphasis mine]." Now, the full details of that presentation, along with new findings, have been published -- and the results are sufficiently promising as to suggest that it may overtake existing strategies for mitochondrial rejuvenation.

Import, Processing, and Expression
In their original report,(1) Teitell's group had identified 20-nucleotide stem-loop import sequences in RNase P and MRP RNA that greatly facilitated their PNPase-dependent import into mitochondria. When fused to the 5′-terminus of the RNA for nucleotides that are not physiologically imported, these import sequences enabled their import into isolated yeast and human mitochondrial matrix, in a PNPase-dependent fashion. Now, in a new report, they have provided evidence of its ability to facilitate "allotopic" mitochondrial import of mRNA for wild-type versions of mitochondrially-encoded proteins, and to import corrective tRNA into cells bearing mutations in the same ribonucleotides. The imported tRNA restore protein expression in mutant cells, and the newly-generated mitochondrial proteins appear to integrate into the Complexes of the electron transport system, and substantially rescue oxidative phosphorylation.(2)

The mutations that underlie the mitochondriopathies Myoclonic Epilepsy with Ragged Red Fibers (MERRF) and some variants of Mitochondrial Encephalomyopathy, Lactic Acidosis, and Strokelike Episodes (MELAS) create defective versions of particular mitochondrial amino acid tRNAs, leading to inefficient translation of electron transport system (ETS) subunit proteins. The investigators synthesized wild-type versions of the defective mitochondrial tRNA precursors that contained or lacked the RNase P import sequence (RP) appended to their 5′ ends. Isolated wild-type mouse liver mitochondria efficiently imported those constructs that included the appended RP sequence, and cleaved their transcript terminus sequences in line with normal mRNA maturation. Constructs lacking the RP sequence were barred from the mitochondria, and even constructs with appended RP were reduced to ~13-17% of normal import into mitochondria from mice with a liver-specific, two-thirds reduction in levels of the murine PNPase homolog.(2) In mitochondria derived from mitochondriopathy cells, similarly, incubation with WT tRNAs partially normalized the steady-state abundance of mitochondrial polypeptides only when those tRNAs bore an appended RP sequence.(2)

In a further test that would more accurately reflect such constructs' ability to function under physiologic conditions, HeLa cells and human fibroblasts were transfected with constructs for mtRNA for (respectively) mouse and human cytochrome c oxidase 2 (COX2), under control of the physiological promoter for the RP gene, and with or without the RP import sequence appended to their 5′ terminus. Transcripts of the introduced COX2 mtRNAs only appeared in the mitochondria of cells in which the introduced constructs included the RP import sequence. Similarly, in the mitochondria of fibroblasts transfected with human COX2 mtRNA, the protein was expressed -- and could be localized to the inner membrane -- when and only when the import sequence was imported. Importantly, this also demonstrated the ability of the import sequence to facilitate import of mRNA, and of RNA substantially (~8.5-10 times) larger than tRNA.(2)

Modifications for Mitochondrial MEDEVAC
However, when they then went on to attempt to use their system to introduce WT mitochondrial tRNAs into MELAS and MERRF cybrids -- and thus rescue oxidative phosphorylation (OXPHOS) -- Teitell's group encountered a series of hurdles requiring additional modification of their constructs. In the first setback, synthetic WT tRNA constructs introduced and stably expressed within the nucleus of mitochondriopathy cybrids were prematurely processed by removal of their 5' presequences within the nuclei, rather than intramitochondrially as required for orthotopic expression. To overcome this problem, the investigators first tried eliminating the mismatch of unpaired ribonucleotides in the  aminoacyl stem of the constructs' import sequences by replacement of several adjacent ribonucleotides. As desired, this reduced intranuclear cleavage of tRNA presequences; however, the improved constructs still failed to rescue OXPHOS in the cells.(2)

The investigators reasoned that, while their constructs' appended RP import sequences can facilitate mitochondrial import, that ability cannot be exploited if the transcripted constructs are not actually shuttled from the nuclear membrane to the mitochondrial surface for import. Therefore, Teitell's group opted to further modify their constructs with a 3′-UTR mitochondrial targeting sequence (MTS) from human mitochondrial ribosomal protein S12 -- the same essential approach used to optimize the import of allotopically-expressed proteins by Dr. Marisol Corral-Debrinski,(3,4) and subsequently advanced for multiple additional ETS subunit proteins by Dr. Matthew O'Connor's group at the SENS Foundation Research Center (RC).(5)

Testing eight different combinations of the corrective tRNAs, each designed with or without the RP import signal, its modified aminoacyl stem, or the MTS, they found that with the inclusion of all three modifications -- and only with their inclusion -- expression of their constructs induced a substantial increase in mitochondrially-encoded protein synthesis (6-8-fold) and respiration (~2.5-fold) in mutant cybrid cells. The restoration of the ETS structure and function was confirmed by in-gel activity assay of complex I, which demonstrated an increase in activity to 30–40% of that in WT cells in cybrids transfected by constructs containing all three modifications; it was also supported by the "marked" increase in levels of COX2 and ND6, both of which are mitochondrially-encoded ETS subunit component proteins. Importantly, moreover, the same constructs did not interfere with respiration when transfected into WT cells.(2)

Respect to the Ancestors; Goads for the Innovators
The UCLA group's approach is highly promising. Their work builds upon and may potentially supersede  several previous approaches to the problem of mitochondrial mutations that occur as a result of the degenerative aging process, including allotopic protein expression,(6) its optimization using an MTS,(3-5) and the exploitation of the multiprotein RNA import complex (RIC) of the protozoal parasite Leishmania tropica(7) (which the investigators characterize as "requir[ing] the introduction of nonnative tRNAs with foreign protein factors or the transfer of a large multisubunit aggregate into cells, which is of low efficiency and difficult to reproduce in desirable disease-relevant settings"(2)).

As compared to allotopic protein expression, an RNA-based approach has the theoretical advantage of abrogating the difficulties encountered thus far with the mitochondrial import of large and hydrophobic proteins. But as we suggested in discussion of their earlier, more discovery-phase research, allotopic protein and RNA approaches are not mutually exclusive: different mitochondrially-encoded proteins could be either allotopically expressed, or their mRNAs generated allopically and imported for in situ translation, depending on the ease or efficiency of each approach for the protein in question. The use of a dual-track approach might be speculated to have an additional advantage, in avoiding any hypothetical "saturation" of the relevant mitochondrial import machinery (PNPase or TIM/TOM complex) if only one approach is used for all 13 mitochondrially-encoded proteins.

There is clearly more work to do to probe the limits of this approach. Thus far, even the largest RNA that Teitell's group have imported (that for human COX2) is relatively small, and the efficiency relatively low, so larger proteins' mRNA must be trialed. Individual mRNA constructs should be evaluated for the dispensibility, and for the relative efficiency, of including (or not) of the modification of the aminoacyl stem of the RP import sequence, and of the MTS. The complete translation of "allotopic" mRNA for COX2 and for future candidates into their encoded proteins, and their structural integrity, should be more rigorously tested, and a more ironclad ascertainment be provided of their functional integration into the ETS.

For their part, Dr. Teitell's group is evidently optimistic, and have clearly moved beyond the basic science focus of their earlier report(1):

this approach may generalize to mtDNA mutations in mt-tRNAs, mt-rRNAs, and protein-encoding mtRNAs as well as to heteroplasmic mtDNA populations, where ribozymes can be targeted. Thus, this rational transcript engineering approach may represent a unique therapeutic opportunity for a wide range of diseases caused by mutations in the mitochondrial genome for which current effective therapies are lacking(2)

... a range which, as they note elsewhere in their report, extends to "muscular and neuronal diseases and ... decline of organ function with aging." They will doubtless be exploring many of the questions we have poised above -- and are hereby put on notice that others are now considering the theft of their thunder. In response to this report, SENS Foundation CSO Dr. Aubrey de Grey has said that "If this is as good as it looks, I think it could be a real game-changer", and Dr. O'Connor and his team at the SENS Foundation RC are considering testing a construct based on Teitell's methods in a system that the RC has already generated and used for testing of allotopic expression of cytochrome B.

The race is on -- as it should be, for the stakes are large. Large, age-related deletions in mtDNA are likely responsible for the systemic rise in oxidative stress with aging, and for localized but terrible pathologies of skeletal muscle and substantia nigra dopaminergic neurons in aging bodies. The obviation of these mutations is a desperate medical need, and biomedicine is shamed for every day that a solution is delayed. This new method must be tested and exploited to its limits, and all approaches must be trialed, until the fires of life are once again burning in rejuvenated cells, in bodies restored to their youthful prime.


1: Wang G, Chen HW, Oktay Y, Zhang J, Allen EL, Smith GM, Fan KC, Hong JS, French SW, McCaffery JM, Lightowlers RN, Morse HC 3rd, Koehler CM, Teitell MA. PNPASE Regulates RNA Import into Mitochondria. Cell. 2010 Aug 6;142(3):456-467. PubMed PMID: 20691904.

2: Wang G, Shimada E, Zhang J, Hong JS, Smith GM, Teitell MA, Koehler CM. Correcting human mitochondrial mutations with targeted RNA import. Proc Natl Acad Sci U S A. 2012 Mar 12. [Epub ahead of print] PubMed PMID: 22411789.

3: Bonnet C, Augustin S, Ellouze S, Bénit P, Bouaita A, Rustin P, Sahel JA, Corral-Debrinski M. The optimized allotopic expression of ND1 or ND4 genes restores respiratory chain complex I activity in fibroblasts harboring mutations in these genes. Biochim Biophys Acta. 2008 Oct;1783(10):1707-17. Epub 2008 May 6. PubMed PMID: 18513491.

4: Ellouze S, Augustin S, Bouaita A, Bonnet C, Simonutti M, Forster V, Picaud S, Sahel JA, Corral-Debrinski M. Optimized allotopic expression of the human mitochondrial ND4 prevents blindness in a rat model of mitochondrial dysfunction. Am J Hum Genet. 2008 Sep;83(3):373-87. Epub 2008 Sep 4. PMID: 18771762 [PubMed - indexed for MEDLINE]

5: O'Connor MS, Swaminathan G, Fazal S, Jones T, de Grey AD. MitoSENS: Allotopic expression of mitochondrial genes using a co-translational import strategy. Rejuvenation Res. 2011 Aug;14(Suppl1):S34(Abs 77). Presentation video here.

6: de Grey AD. A mechanism proposed to explain the rise in oxidative stress during aging. J Anti-Aging Med 1998;1(1):53-66. 

7: Mahata B, Mukherjee S, Mishra S, Bandyopadhyay A, Adhya S. Functional delivery of a cytosolic tRNA into mutant mitochondria of human cells. Science. 2006 Oct 20;314(5798):471-4. PubMed PMID: 17053148.