© 1999 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Nanomedicine, Volume I: Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999

10.4.2.2 Mechanical Cytoskeletolysis and Monkeywrenching

It may have occurred to the alert reader that a simple way to kill a eukaryotic cell might be to cytopenetrate and then motor around inside the cytoplasm at higher-than-recommended velocities, thus indiscriminately bursting lysosomes, shredding the thin membranes of the endoplasmic reticulum (ER) and the Golgi, and trashing other delicate cellular structures, depending upon the motility mechanism employed (Section 9.4). Such a course would almost certainly result in an unwanted necrotic cellular lysis.

Rupture of lysosomes (Section 8.5.3.8) should especially be avoided because lysosomal lipases and other corrosive enzymes would be released into the cytosol, possibly provoking some autolysis, local disintegration of the plasma membrane, and inflammation. Peroxisome (Section 8.5.3.9) rupture may produce similar effects, since these organelles catalyze some of the fatty acid breakdown in the cell. Selective destruction of ribosomes (Section 8.5.3.4), the ER (Section 8.5.3.5), the Golgi complex (Section 8.5.3.6), the mitochondria (Section 8.5.3.10), or the nucleus (Section 8.5.4) will more slowly kill the cell but again this may be a nonapoptotic death if the transcription, protein synthesis, or energizing mechanisms necessary to sustain the apoptotic cascade are disabled. Loss of power due to mitochondrial destruction eliminates the cell's ability to maintain long-term membrane recycling; the cell will eventually come apart due to the accumulated damage in the membrane.

However, elements of the cytoskeleton probably can be safely destroyed without disabling apoptosis, while simultaneously severely reducing the chances of post-attack cell survival, particularly in cells that are incapable of further division. Chemical cytoskeletolysis by caspases (Section 10.4.1.1) and various mitotic inhibitors such as the microtubule inhibitor vincristine (Sections 9.4.7.4 and 10.4.1.3) have been described earlier. We can speculate that mechanical cytoskeletolysis may trigger apoptosis in eukaryotic cells because the entire cytoskeletal structure per se does not appear to be a crucial operational component of the apoptotic cascade.

How best to perform the cytoskeletolysis? Microtubules in cyto preferentially absorb light at ~350 nm,¹⁰⁷⁰ so in theory a sufficiently intense laser source at this frequency could selectively disrupt the microtubule network. However, proteolytic or mechanical chopping should be more energy efficient (especially if used to activate a few of the molecules at the upstream end of the cascade). For example, p56 severs microtubules slowly in an ATP-independent manner;³¹⁴⁷ katanin is a heterodimeric protein that severs and disassembles microtubules in an ATP-dependent manner.³¹⁴⁸ Human elongation factor 1a (EF-1a, ~48 kilodaltons) rapidly severs taxol-stabilized fluorescent microtubules in vitro, and induces rapid fragmentation of cytoplasmic microtubule arrays when microinjected into fibroblasts, at concentrations of >~15 microgram/cm³ (>~200 molecules/micron³).¹⁰⁸³ In theory, a pair of fiber cleavage tools modeled after EF-1a operating at ~500 Hz (Section 9.4.6) would lyse ~1000 fibers/sec. If the cytoskeleton of a typical tissue cell has a total of ~10⁶ intermediate filament segments (Table 8.17), then 10% of all such filaments can be mechanochemically cleaved by a single tool pair in ~100 sec, ignoring travel time.

The cytoskeleton can also be mechanically cleaved. Individual microfilaments have a tensile tearing strength of ~0.1 nN, microtubules ~1 nN, and intermediate filaments (IFs) ~100 nN. Consider a cutting blade of length 50 nm rotating at ~1 KHz (blade tip speed ~ 0.3 mm/sec). The energy required to cut a single 10-nm thick IF is ~ 0.001 pJ (Section 9.3.5.1); the power required to cut 10% of all cellular IFs in ~100 sec is ~1 pW. The continuous drag power of the freely rotating blade in plasma during the 100-sec cytoskeletolysis program is ~0.1 pW (Eqn. 9.75). A 1-micron² cross-section nanorobot traversing the entire 8000 micron³ volume of a 20-micron tissue cell traces an 8000-micron path, a 100-sec travel time at a mean ~80 microns/sec velocity; continuous nanorobot drag power at this velocity even in a cytoskeleton-rich cytoplasm of viscosity ~10 kg/m-sec (Section 9.4.6) is ~1 pW (Eqn. 9.73).

Another simple technique for mechanical disruption of the cell may informally be termed "monkeywrenching." As examples, active DNA cutters (e.g., restriction enzymes), or in situ manufactured H₂O₂ along with some iron borrowed from local ferritin molecules, could be injected into the eukaryotic nucleus, rapidly generating enough double-strand breaks to activate the fatal DNA damage response. R. Bradbury notes that a nanorobot could seize a single mitochondrion (~0.4% of nuclear volume; Table 8.17), transport it to the nuclear envelope and then forcibly insert it into the nucleus -- if there are sufficient oxygen and other small reactant molecules present, the mitochondrion would probably produce enough local free radicals to initiate the DNA-damage (apoptotic) cascade. Methods for synthesizing superoxide, hydrogen peroxide, and hypochlorite from locally available materials are described in Chapter 19. In the case of prokaryotes, medical nanorobots could introduce restriction enzyme molecules that are alien to the target strain of bacterium, followed by some DNase (in case the strain has none). R. Bradbury believes that ribonucleases and peptidases may be unnecessary because some endogenous ribonucleases should always be present and the remaining lipid bag of proteins would not represent much of a threat.

Last updated on 24 February 2003