Microbivores: Artificial Mechanical Phagocytes

April 11, 2002 by Robert A. Freitas Jr.

Nanorobotic “microbivores” traveling in the bloodstream could be 1000 times faster-acting than white blood cells and eradicate 1000 times more bacteria, offering a complete antimicrobial therapy without increasing the risk of sepsis or septic shock (as in traditional antibiotic regimens) and without release of biologically active effluents. They could also quickly rid the blood of nonbacterial pathogens such as viruses, fungus cells, or parasites.

Originally published April 2001 in IMM Report Number 25: Nanomedicine, in conjunction with Foresight Update 44. Published on KurzweilAI.net April 11, 2002.

Nanomedicine [1] offers the prospect of powerful new tools for the treatment of human diseases and the augmentation of human biological systems. Diamondoid-based medical nanorobotics may offer substantial improvements in capabilities over natural biological systems, exceeding even the improvements possible via tissue engineering and biotechnology. For example, the respirocytes [2] – artificial red blood cells comprised of microscopic diamondoid pressure tanks that are operated at up to 1000 atm of pressure – could carry >200 times more respiratory gases than an equal volume of natural red blood cells. The clottocytes [3] are artificial platelets that could stop human bleeding within ~1 second of physical injury, but using only 0.01% the bloodstream concentration of natural platelets – in other words, nanorobotic clottocytes would be ~10,000 times more effective as clotting agents than an equal volume of natural platelets.

In this article I’d like to examine the future nanorobotic equivalent of the third major class of natural blood cells – the white cells. This paper summarizes the results of a recently completed scaling study of artificial mechanical phagocytes of microscopic size, called "microbivores." Microbivores constitute a potentially large class of medical nanorobots intended to be deployed in human patients for a wide variety of antimicrobial therapeutic purposes, for example, as a first-line response to septicemia. The analysis focuses on a relatively simple device: an intravenous (I.V.) microbivore whose primary function is to destroy microbiological pathogens found in the human bloodstream, using the "digest and discharge" protocol first described by the author elsewhere [1]. The full technical paper describing the microbivore scaling design study is already available online [4].

Septicemia, also known as blood poisoning, is the presence of pathogenic microorganisms in the blood. If allowed to progress, these microorganisms can multiply and cause an overwhelming infection. Bacteremia is the presence of bacteria in the human bloodstream. Although bacterial nutrients are plentiful in blood, the healthy human bloodstream is generally considered a sterile environment. Major antimicrobial defenses include the circulating neutrophils and monocytes (white cells) capable of phagocytosis (engulfing and digesting other cells) and the supporting components of humoral immunity including complement and immunoglobulins.

Still, it is not unusual to find a few bacteria in blood. Normal activities like chewing, brushing or flossing teeth causes movement of teeth in their sockets, infusing a burst of commensal oral microbes into the bloodstream [5]. Bacteria can enter the blood via an injury to the skin, the lining of the mouth or gums, or from gingivitis and other minor infections in the skin and elsewhere [6]. Bacteria can also enter the blood during surgical, dental, or other medical procedures such as the insertion of I.V. lines (providing fluids, nutrition or medications), cystoscopy (a viewing tube inserted to examine the bladder), colonoscopy (a viewing tube inserted to view the colon), or heart valve replacement with a prosthetic [6]. Such bacteria are normally removed by circulating leukocytes (along with fixed reticuloendothelial phagocytes in the spleen, liver, and lungs), but a few species of bacteria are unusually virulent and can overwhelm the natural defenses. The Center for Disease Control estimates that ~25,000 U.S. patients die each year from bacterial sepsis. Current therapies often involve multiple antibiotics administered simultaneously in multi-gram quantities per day. These treatments can sometimes take weeks or even months to bring under control certain hardy infectious microorganisms like Pseudomonas aeruginosa or enterobacteria such as Escherichia coli and Salmonella.

A nanorobotic device that could safely provide quick and complete eradication of bloodborne pathogens using relatively low doses of devices would be a welcome addition to the physician’s therapeutic armamentarium. Such a machine is the microbivore, an artificial mechanical phagocyte.

The microbivore is an oblate spheroidal nanomedical device consisting of 610 billion precisely arranged structural atoms plus another ~150 billion mostly gas or water molecules when fully loaded. The nanorobot measures 3.4 microns in diameter along its major axis and 2.0 microns in diameter along its minor axis, thus ensuring ready passage through even the narrowest of human capillaries which are ~4 microns in diameter [1]. Its gross geometric volume of 12.1056 micron^3 includes two normally empty internal materials processing chambers totalling 4 micron^3 in displaced volume. The nanodevice consumes 100-200 pW of continuous power while in operation and can completely digest trapped microbes at a maximum throughput of 2 micron^3 per 30-second cycle, large enough to internalize a single microbe from virtually any major bacteremic species in a single gulp. As in previous designs [2], to help ensure high reliability the microbivore has tenfold redundancy in all major components, excluding only the largest passive structural elements. The microbivore has a dry mass of 12.2 picograms.

Here’s how the nanorobot works. During each cycle of operation, the target bacterium is bound to the surface of the microbivore like a fly on flypaper, via species-specific reversible binding sites [1]. Telescoping robotic grapples emerge from silos in the device surface, establish secure anchorage to the microbe’s plasma membrane, then transport the pathogen to the ingestion port at the front of the device where the pathogen cell is internalized into a 2 micron^3 morcellation chamber. After sufficient mechanical mincing, the morcellated remains of the cell are pistoned into a 2 micron^3 digestion chamber where a preprogrammed sequence of 40 engineered enzymes are successively injected and extracted six times, progressively reducing the morcellate ultimately to monoresidue amino acids, mononucleotides, glycerol, free fatty acids and simple sugars. These simple molecules are then harmlessly discharged back into the bloodstream through an exhaust port at the rear of the device, completing the 30-second digestion cycle. This "digest and discharge" protocol [1] is conceptually similar to the internalization and digestion process practiced by natural phagocytes, except that the artificial process should be much faster and cleaner. For example, it is well known that macrophages release biologically active compounds during bacteriophagy [7], whereas well-designed microbivores need only release biologically inactive effluent.

Natural phagocytic cells are 100-1000 times larger in volume than microbivores but may consume almost as much power during comparable activities. For instance, heat production rises from 9 pW in unstimulated human neutrophils up to 28 pW during phagocytosis, with the rise proportional to the number of particles ingested [8]. The basal rate for a resting ~400 micron^3 T-cell lymphocyte is ~20 pW, rising to ~65 pW during antigen response [9].

Microbe ingestion times for natural professional phagocytes can be quite rapid, often a matter of minutes, but full digestion and excretion of the target pathogen may require hours. While macrophages can ingest up to ~25% of their volume per hour [10], microbivores can process ~2000% of their volume per hour, thus are ~80 times more efficient as phagocytic agents. In other words, a given volume of microbivores can digest bacterial pathogens 80 times faster than an equal volume of white cells or macrophages could digest them.

Many natural professional phagocytic cells such as neutrophils also have a maximum capacity for phagocytosis during their short lifetime, typically a few hours in blood or a few days in tissue. In one experiment [11], 1-100 S. aureus or S. faecalis bacteria were presented to each neutrophil, which digested more of them at the higher concentrations. At the highest concentration (100:1), neutrophils from normal patients could only kill a mean of 9 S. aureus bacteria per phagocyte, while neutrophils from carriers of chronic granulomatous disease could kill a mean of 14 S. faecalis bacteria per phagocyte. By comparison, a single microbivore could completely digest up to ~3000 microbes/day of P. aeruginosa bacteria with no well-defined maximum lifetime capacity for phagocytosis.

In the accompanying technical paper [4] a simple mathematical model for microbivore pharmacokinetics quantifies the activity of a specific dose of nanorobots injected into the human bloodstream, with the conclusion that a 1-terabot (10^12-device) dose of microbivores employed in the treatment of a mild bacteremia (0.1 x 10^6 colony-forming units (CFU) per ml) can reduce the initial whole-bloodstream bacterial load of 5.4 x 10^8 CFU down to <1 CFU in 460-5400 sec (8-90 min), if 1-10 bacterium-microbivore collisions are required for the bacterium to stick.

Similarly, a severe bacteremia (100 x 106 CFU/ml) is eliminated in 620-7300 sec (10-120 min). Note that a single 1-terabot intravenous dose (a ~12 cm^3 injection) of microbivores produces a nanocrit of Nct ~ 0.2% in the blood of a normal adult human male patient and could liberate up to 100-200 watts of systemic waste heat which is very near the maximum thermogenic limit for in vivo medical nanorobot systems [1].

While microbivores can fully eliminate septicemic infections in minutes to hours, natural phagocytic defenses – even when aided by antibiotics – can often require weeks or months to achieve complete clearance of target bacteria from the bloodstream. Thus microbivores appear to be up to ~1000 times faster-acting than either unaided natural or antibiotic-assisted biological phagocytic defenses.

Another useful comparative perspective is that the administration of antibacterial agents (e.g., against E. coli) typically may increase the LD50 of that pathogen by ~500-fold using antibiotics [12] or ~850-fold using monoclonal antibodies [13]. For example, the mammalian LD50 for E. coli is ~0.1-1 x 10^6 CFU/ml, rising to ~10^8 CFU/ml with the administration of antibiotics. By employing a suitable dose of microbivores, a bloodstream bacterial concentration up to the theoretical maximum of ~10^11 CFU/ml (~20% of blood volume assuming ~2 micron^3 organisms) could be controlled, bringing another ~1000-fold improvement using nanomedicine and at last extending the therapeutic competence of the physician to the entire range of potential bacterial threats, including locally dense infections.

With minor additions to the basic design, microbivores could be used to combat toxemia, the distribution throughout the body of poisonous products of bacteria growing in a focal or local site, and other biochemical sequelae of sepsis. For instance, E. coli-induced septicemic shock in vervet monkeys occurred at 425 x 10^6 CFU/ml and bacterial lipopolysaccharide (LPS) endotoxin rose from normal at 0.076 ng/ml to a maximum of 1.130 ng/ml blood concentration [14]. In another study, endotoxin levels during a gram-negative bacterial infection rose from 0.2 to 2 ng/ml in pig blood [15]. Eliminating a bloodstream concentration of ~2 ng/ml of ~8 kDa LPS endotoxin [16] would require the extraction and enzymatic digestion of ~8 x 10^14 LPS molecules from the ~5400 cm3 human blood compartment, a mere ~800 LPS molecules per nanorobot assuming a 1-terabot dose of modified microbivores.

The high mortality (up to 30%-50%) associated with gram-negative sepsis is due in large measure to the patient’s reaction to LPS, an endotoxin which induces the production of cytokines such as IL-1beta and IL-6 which leads to an uncontrolled inflammatory reaction resulting in tissue damage and organ failure [17]. We’ve already noted that small quantities (~ng/ml) of LPS are released by living and growing bacteria, but the killing of bacteria using traditional antibiotic regimens often liberates large quantities of additional LPS, potentially up to ~10^5 ng/ml [17]. Such massive releases as occur with the use of antibiotics will not accompany the use of microbivores, because all bacterial components (including all cell-wall LPS) are internalized and fully digested into harmless nonantigenic molecules prior to discharge from the device. Microbivores thus represent a complete antimicrobial therapy without increasing the risk of sepsis or septic shock.

If the patient presents with a septic condition before the microbivores are introduced, a substantial preexisting concentration of inflammatory cytokines will likely be present and must be extracted from the blood in concert with the primary antibacterial microbivore treatment. All unwanted cytokine molecules may be rapidly and systemically extracted from the blood using a modest dose of respirocyte-class nanodevices [2] such as pharmacytes [1], a combination-treatment approach previously suggested elsewhere [1, 18]. Specifically, a 1-terabot intravenous dose of micron-size pharmacytes [1] each having ~10^5 cytokine-specific molecular sorting rotors and ~0.5 micron^3 of onboard storage capacity could reduce the blood concentration of ~20 kDa IL-1beta and IL-6 cytokines from LPS-elevated levels of ~100 ng/ml [210] (~3 x 10^-9 molecules/nm^3) down to normal serum levels of ~10 pg/ml [211] (~3 x 10^-13 molecules/nm^3) after only ~200 sec of diffusion-limited pumping, using just ~0.1% of the available onboard storage volume. (Extracting an additional ~10^5 ng/ml of LPS from the bloodstream would take a similar amount of time and use ~100% of the available onboard storage volume.)

Microbivores could also be useful for treating infections of the meninges or the cerebrospinal fluid (CSF) and respiratory diseases involving the presence of bacteria in the lungs or sputum, and could also digest bacterial biofilms. These handy nanorobots could quickly rid the blood of nonbacterial pathogens such as viruses (viremia), fungus cells (fungemia), or parasites (parasitemia). Outside the body, microbivore derivatives could help clean up biohazards, toxic biochemicals or other environmental organic materials spills, as in bioremediation.


The author thanks C. Christopher Hook, M.D., Stephen S. Flitman, M.D., Ronald G. Landes, M.D., and also Forrest Bishop, Robert J. Bradbury, and Ralph C. Merkle, for helpful comments on the technical paper from which this summary article has been abstracted.


1. Robert A. Freitas Jr., Nanomedicine, Volume I: Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999. See at: http://www.nanomedicine.com.

2. Robert A. Freitas Jr., "Exploratory Design in Medical Nanotechnology: A Mechanical Artificial Red Cell," Artificial Cells, Blood Substitutes, and Immobil. Biotech. 26(1998):411-430. See at: http://www.foresight.org/Nanomedicine/Respirocytes.html.

3. Robert A. Freitas Jr., "Clottocytes: Artificial Mechanical Platelets," Foresight Update No. 41, 30 June 2000, pp. 9-11. See at: http://www.imm.org/Reports/Rep018.html.

4. Robert A. Freitas Jr., "Microbivores: Artificial Mechanical Phagocytes using Digest and Discharge Protocol," March 2001; see at: http://www.zyvex.com/Publications/articles/Microbivores.html.

5. John Heritage, "Septicaemia and Endocarditis," Laboratory and Scientific Medicine, MICR3290 Medical Microbiology, University of Leeds, November 1996; see at: http://www.leeds.ac.uk/mbiology/ug/med/septendo.html.

6. Robert Berkow, Mark H. Beers, Andrew J. Fletcher, eds., The Merck Manual of Medical Information, Merck Research Laboratories, Whitehouse Station NJ, 1997.

7. E.F. Fincher, L. Johannsen, L. Kapas, S. Takahashi, J.M. Krueger, "Microglia digest Staphylococcus aureus into low molecular weight biologically active compounds," Am. J. Physiol. 271(July 1996):R149-R156.

8. H. Hayatsu, T. Miyamae, M. Yamamura, "Heat production as a quantitative parameter of phagocytosis," J. Immunol. Methods 109(9 May 1988):157-160.

9. Augusto Cogoli, Birgitt Bechler, Marianne Cogoli-Greuter, Sue B. Criswell, Helen Joller, Peter Joller, Elisabeth Hunzinger, Ottfried Muller, "Mitogenic signal transduction in T lymphocytes in microgravity," J. Leukocyte Biol. 53(May 1993):569-575; Dennis A. Carson, "Chapter 94. Composition and biochemistry of lymphocytes and plasma cells," in Ernest Beutler, Marshall A. Lichtman, Barry S. Coller, Thomas J. Kipps, eds., William’s Hematology, Fifth Edition, McGraw-Hill, New York, 1995, pp. 916-921.

10. Robert K. Murray, Daryl K. Granner, Peter A. Mayes, Victor W. Rodwell, Harper’s Biochemistry, 23rd Edition, Appleton & Lange, Norwalk CT, 1993.

11. J.E. Repine, C.C. Clawson, "Quantitative measurement of the bactericidal capability of neutrophils from patients and carriers of chronic granulomatous disease," J. Lab. Clin. Med. 90(September 1977):522-528.

12. S.E. Bucklin, D.C. Morrison, "Bacteremia versus endotoxemia in experimental mouse leukopenia – role of antibiotic chemotherapy," J. Infect. Dis. 174(December 1996):1249-1254.

13. J. Colino, I. Outschoorn, "The form variation of the capsular polysaccharide K1 is not a critical virulence factor of E. coli in a neonatal mouse model of infection," Microb. Pathog. 27(October 1999):187-196.

14. B.C. Wessels, M.T. Wells, S.L. Gaffin, J.G. Brock-Utne, P. Gathiram, L.B. Hinshaw, "Plasma endotoxin concentration in healthy primates and during E. coli-induced shock," Crit. Care Med. 16(June 1988):601-605.

15. O. Rokke, A. Revhaug, B. Osterud, K.E. Giercksky, "Increased plasma levels of endotoxin and corresponding changes in circulatory performance in a porcine sepsis model: the effect of antibiotic administration," Prog. Clin. Biol. Res. 272(1988):247-262.

16. L. Aussel, R. Chaby, K. Le Blay, J. Kelly, P. Thibault, M.B. Perry, M. Caroff, "Chemical and serological characterization of the bordetella hinzii lipopolysaccharides," FEBS Lett. 485(17 November 2000):40-46.

17. J.T.M. Frieling, J.A. Mulder, T. Hendriks, J.H.A.J. Curfs, C.J. van der Linden, R.W. Sauerwein, "Differential Induction of Pro- and Anti-Inflammatory Cytokines in Whole Blood by Bacteria: Effects of Antibiotic Treatment," Antimicrob. Agents Chemother. 41(July 1997):1439-1443.

18. Robert A. Freitas Jr., "Nanopyrexia," Foresight Update No. 43, 30 December 2000, pp. 14-16. See at: http://www.imm.org/Reports/Rep022.html.

19. J. Hulkkonen, P. Laippala, M. Hurne, "A rare allele combination of the interleukin-1 gene complex is associated with high interleukin-1b plasma levels in healthy individuals," Eur. Cytokine Netw. 11(April 2000):251-255.

Copyright 2001 Robert A. Freitas Jr. All Rights Reserved. Used with permission.