Figure 1

(Color) Schematic diagram of a time-shared and holographic optical trapping apparatus. Two-dimensional (2D) trap arrays are formed using a Zeiss Achroplan high NA objective $(100\times ∕1.25\mathrm{NA})$ commercial Axiovert 200M optical microscope in conjunction with either one of two diffractive elements: i.e., two acousto-optic deflectors, or a spatial light modulator. The same microscope that is used to produce the cell traps is also used for viewing (via the yellow beam) and photopolymerizing the PEGDA hydrogel (via the blue beam). The inset in the upper right shows an example of a 2D $5\times 5$ array of E. coli formed using this apparatus and subsequently embedded in hydrogel. The distances are AODs: $\mathrm{L}1=400\mathrm{mm}$; L1 $\mathrm{L}2=800\mathrm{mm}$; L2 entrance $\text{pupil}=400\mathrm{mm}$; L3 $\mathrm{L}4=800\mathrm{mm}$; and L4 $\mathrm{OEA}=420\mathrm{mm}$, where the focal lengths for L1, L2, L3, L4 are $400\mathrm{mm}$.Reuse & Permissions

Figure 2

(Color) (a) The receiver plasmid, GFP-M1, is the focus of this work. It combines the lac operon with green fluorescent protein (GFP). Induction by IPTG initiates GFP production. The resulting fluorescence is used to indicate an active cell metabolism. (b) Threshold for induction of M1’s by IPTG at $25°\mathrm{C}$. Bacteria count measured using laser cytometry for 23 000 E. coli cultured at $25°\mathrm{C}$. A typical distribution as a function of the fluorescent intensity obtained for $1\mathrm{mM}$ IPTG is shown in the inset. The active cell count indicated by green fluorescence is shown in green, and the inactive is shown in red. The IPTG threshold estimated from a fit to the data is $24\mu \mathrm{M}$. Only about 80% of the cells are active even at high IPTG concentration.Reuse & Permissions

Figure 3

(Color) Genetically identical cells express GFP gene differently due to the stochastic nature of biochemical processes. (a) Time evolution of green fluorescence in a $5\times 5$ microarray of M1 bacteria following the induction using $10\mathrm{mM}$ IPTG at $t=0$. The bacterial array was assembled using time-shared optical traps formed using a beam with $P=140\mathrm{mW}$ at $\lambda =900\mathrm{nm}$. The array was held for about $8\min$ prior to gelling. We can follow the development of the fluorescence associated with each cell in the array as a function of time. (b) A summary of the time development of the fluorescence measured on five cells in the array. The fluorescence usually saturates after about $5\mathrm{h}$.Reuse & Permissions

Figure 4

(Color) Viability as a function of near IR wavelength, power for time-shared and static optical traps. (a) $5\times 5$ 2D arrays of E. coli bacteria incorporating the plasmid GFP-M1 are assembled using a time-shared optical trap with the specified wavelength and power. In each case the cells in the microarray are held for about $8\min$ prior to gelling. The peak power is indicated along with the corresponding time-averaged power in parentheses. The bar graph represents viable, active bacteria (green), inactive bacteria (gray), and dead bacteria (red) for each wavelength and power. Viability decreases nearly linearly with increasing power, and peaks at $\lambda =840$ and $900\mathrm{nm}$. (b) Similar to (a) but now using a CW beam to form the $5\times 5$ 2D arrays of E. coli. The static CW in optical traps ranges from about $5\text{to}20\mathrm{mW}$ at the specified wavelength. Again, in each case the cells in the microarray are held for about $8\min$ prior to gelling. The CW viability tracks that found for the time-shared trap at about the same time-averaged power. The right side of the corner shows control bacteria, untrapped but encapsulated in the hydrogel spot.Reuse & Permissions

Figure 5

Viability depends on energy. The percentage of the population that is viable (i.e., expressing GFP) after trapping is shown as a function of total energy delivered to each cell. The CW and time-shared data follow the same trend. The energy seems to be a good predictor of viability. The line fit to the data represented by the dotted line is a given by $\%V=-11.1$$E+106$. An energy of about $5\mathrm{J}$ is the lethal dose associated with 50% of the community. The inset shows viability as a function of the dwell time for time-shared optical traps. $5\times 5$ 2D arrays of E. coli bacteria incorporating the plasmid GFP-M1 are assembled using time-shared optical traps at $\lambda =900\mathrm{nm}$ with a power of $140\mathrm{mW}$ in the beam. In each case the cells in the microarray are held for about $8\min$ prior to gelling. The percentage of the population that is viable (i.e., expressing GFP) after trapping is shown as a function of dwell time for a 1:25 duty cycle scan. Since the duty cycle is the same, the time-averaged power is independent of the dwell time. Likewise, the viability is relatively independent of dwell time.Reuse & Permissions