Julius B. Lucks/Bibliography/Simpson-TrendsBiotechnol-19-2001

Notes on [1]

• complex tasks performed by cells: sensing, navigation, communication, cooperation, nano-fabrication
• review on mimicing man-made information processing inside cells
• to accomplish tasks, need memory, sensing, feedback, communication
• E. coli - 2 micrometer^2 cross section - 4.6 x 10^6 bp chromosome - equiv 9.2 megabit memory
  • how arrive at this number?
  • 4300 diff polypeptides produced from several hundred diff promoters
• compare to semiconductors International Technology Roadmap for Semiconductors
  • by 2014, memory density 24.5 Gbits/cm^2 and logic transistor density 664 M/cm^2
    • assuming 4 transistors per logic function (how get this?), 2 micrometer^2 Si could contain 490-bit memory with 3 simple logic gates
• bacterial cells, viewed as 'devices' - tolerate a wide array of conditions
• can integrate them in 3D structures
• biofilm formation an alternative to lithography and other manufacturing techniques neccesary for Si integrated circuits
• BBIC - bioluminescent bioreporter integrated circuit (4-8)
• present 3 examples of sensing/information processing/actuation that occurs in natural cells
  1. directed motility (chemotaxis, phototaxis, magnetotaxis (12-14))
  2. symbiotic collonization as a communications system, esp. to initially establish the relationship
     • Hawaiian squid Euprymna scolopes with lumincescent Vibrio fisheri (24)
       • Upon colonization V. fischeri looses flagella, reduces cell size, decreases growth rate, enhances luminescence (25,26)
  3. group formation (biofilms) - members have some protection from phage, biocides, antibiotics
• 'silicon mimetic' approach to engineering cell information processing
  • engineered genetic regulatory functions emulate the functionality of silicon semiconductor devices
  • silicon paradigm - three terminal device in which transport between 2 of the terminals controlled by a signal at the third (FET)
    • biochemical analogy - substrate-enzyme-product controlled by some effector acting on the enzyme (bipolar junction transistor) (32,33)
      • hardware interconnects not needed
    • Fig 2 shows the lux system in V. fischerei used in the BBIC that implements this design - control of light production by O2 and FMNH2 modulation of luciferase activity
  • realizing logic gates with genetic machinery
    • the key to logic gates is interconnectivity
    • constructing gene transcription modules for biochemical devices that are logic circuits (Knight's group - 9,10)
    • this group working on AND, OR and XOR
    • implementing OR
      1. use 2 promoters that behave same as transcription factors, but are affected by 2 diff effectors
      2. use 1 promoter that respots same way to 2 diff inhibitors - tod-lux fusion in P. putida TVA8 (induced by trichloroethylene and toluene)
    • combinations of AND, OR and XOR can implement any combinatorial logic function - but CANNOT be used for sequential circuits that require memory of past logic states and clock signal synchronization
    • Gardner, Cantor, Collins - one-bit memory using double repressor toggle switch (uses lamba cI and part of Lac operon - ('Construction of a Genetic toggle switch in E. coli' 35) (see also [2])
    • 'challenge - develop genetic circuit that makes more eff use of cells DNA memory capacity
    • Elowitz and Liebler repressilator can be used as clock (36) (why not also cell cycle?)
  • interconnecting
    • needed for even moderate complexity
    • suggest using cell-to-cell signals to connect - isolating single gate inside a cell
    • quorum sensing via N-acyl-homoseine lactones (AHLs) - diffusable cell-to-cell signals present in many Gram-negative bacteria (37)
    • V. fischeri - luxI encodes AHL synthetase, luxR encodes AHL-dependant transcriptional activator
    • primary consideration is number of molecules in circuits and the amoun of cross-talk
      • (39) - P. aeruginosa and V. fischerei AHL systems don't interact
  • Input/Output
    • review on EMF and electric current pulses in living cells (43) - little data on how effects gene expression - need to know this for hybrid systems - Fig 6 describes their experiment towars this investigation
  • noise
    • finite period of time necc for an effector to reach operational concentration, certain time needed for it to decay below operational concentration as well
    • outcomes of genetic networks not deterministic (44,45)
      • problem for non-linear circuits with bifurcations since can cause non-boolean, probabilistic responses
    • simulating such circuits - Gillespi algorithm - see Adam's paper on Bio-SPICE (48)

References

1. Simpson ML, Sayler GS, Fleming JT, and Applegate B. Whole-cell biocomputing. Trends Biotechnol. 2001 Aug;19(8):317-23. DOI:10.1016/s0167-7799(01)01691-2 |

   doi:10.1016/S0167-7799(01)01691-2 Notes

2. Kobayashi H, Kaern M, Araki M, Chung K, Gardner TS, Cantor CR, and Collins JJ. Programmable cells: interfacing natural and engineered gene networks. Proc Natl Acad Sci U S A. 2004 Jun 1;101(22):8414-9. DOI:10.1073/pnas.0402940101 |

   Notes