Experiment 4

This is similar to Exp 3, except we are running 6 chemostats simulaneously, where MC1 (called MC1-C12) is the control with no 13C labels and the next five chemostats (MC2,..., MC6) each have one of the five substrates (Methanol, MC2-Met; Ethanol, MC3-Eth; Acetate, MC4-Ace; Glucose, MC5-Glu; Xylose, MC6-Xyl) replaced with its 13C equivalent. Chemostat operation has also changed, in that we will start the experiment with 13C-labeled substrates (one substrate labeled in each MC) and follow 13C inclusion into the food web during batch phase (no media flow). Once the batch phase has finished (about 1 week), chemostat operation will begin, but the media will contain no 13C-labeled substrates so that the heavy isotope will be washed out. After approximately 10 days, we will switch to the media with individually 13C-labeled substrates and follow their inclusion into the food web under chemostat conditions at a dilution rate of 1 d-1. This single experiment will allows us to track how the food web functions under both transient (batch) and steady state (chemostat) conditions.

As bacteria consume the substrates with heavy isotopes, as well as their predators (such as protists, viruses and other bacteria) that consume the enriched bacteria, their RNA will become enriched with 13C and detectable using stable isotope probing techniques (also see Fortunato and Huber 2016).   Experimental data collected during Exp 4 will be used to determine the extent of interconnectivity within microbial communities.

Experimental clock time started at 00:00 1-Apr-2022; however, all six chemostats were filled with 2.3 L of artificial media containing 13C-labeled substrates and inoculated with 600 mL of Siders Pond water at 13:00. Siders Pond surface water was collected on 31-Mar-2022 at 9:30 and passed through a 425 µm stainless steel sieve (#40). Water salinity was 3.8 and was acclimated from 8.39ºC to 20ºC over 36 hrs while being bubbled with air. The pH after acclimation was 7.7. All times are in EDT.

Note: Only MC2 and MC5 have DO and pH probes installed. Furthermore, the DO and pH data below are from the analog output of the probes, which can be offset from true values or missing. The true digital DO and pH data collected over ModBus is given in the link below, and the graph of these data can be found on the JS Graphics page.

On-line variables associated with MC1-12C Control (last 36 points):

Gas concentrations in output
Plot of O2, CO2 over all time.   Interactive graphics are also available under the JavaScript Graphics menu item above. 

On-line variables associated with MC2-13C Methanol (Last 36 points):

Concentration of off gases in MC2DO and pH associated with MC2
Plot of O2, CO2 all time.  Plot of pH and DO over all time. 

On-line variables associated with MC3-13C Ethanol (Last 36 points):

Concentration of off gases in MC3
Plot of O2, CO2 all time. 

On-line variables associated with MC4-13C Acetate (Last 36 points):

Concentration of off gases in MC4
Plot of O2, CO2 all time. 

On-line variables associated with MC5-13C Glucose (Last 36 points):

Concentration of off gases in MC5DO and pH associated with MC5
Plot of O2, CO2 all time.  Plot of pH and DO over all time. 

On-line variables associated with MC6-13C Xylose (Last 36 points):

Concentration of off gases in MC6
Plot of O2, CO2 all time. 

Instrument Drift (between analyzer calibrations), Last 36 points

Instrument Drift
Instrument Drift over all time.

Experimental Events:

  1. At 9:30 EDT 31-Mar-2022, a 20 L surface water sample was collected from the surface of Siders Pond, Falmouth MA (41.551115°N, -70.619438°E) and passed through a 425 µm stainless steel sieve (#40) to remove large organisms and particles. At time of sampling, water salinity was 3.8 PSU and temperature was 8.39ºC. The 20 L carboy was placed in growth chamber and allowed to acclimate to 20ºC while being bubbled with air.
  2. All 6 microcosms where inoculated with 600 mL of acclimated Siders Pond water at 13:00 1-Apr-2021 (or t = 0.54 d). This marks the start of Exp. 4.
  3. Due to sample withdrawals, and batch operation, the media level in the bioreactors was below the level of the DO probes from day 1.53 until they were lowered back into the culture media at day 2.49.
  4. pH Probe in MC 5 was exposed to air from t = 4.55 d (13:28 5-Apr-2022) until t = 5.37 d (9:08 6-Apr-2022) due to drop in medium volume from sampling chemostat.
  5. 6-Apr-2022 10:13 (t = 5.43 d): Microcosms have been sample each day following inoculation on 1-Apr-2022. Current volume of reactors is approximately 1 to 1.4 L.
  6. Chemostat operation started at t = 5.737 d (17:41 6-Apr-2022) at a dilution rate of 1 d-1 (3 L d-1).
  7. 11:30 14-Apr-2022 (t = 13.47 d). Air entrainment into the ASW pump tubing connector was identified, which resulted in lower flow and dilution rate to the chemostats than desired. This may have started around day 7.5, but timing is uncertain. This was repaired on 13-Apr-2022 (t = 12.5). Also, a gas leak was identified in the headspace gas sampling loop. Some air may have been entrained into the headspace of the bioreactors during sampling resulting in slight dilution of the headspace gas concentrations. This was repaired today (t = 13.50 d). It is uncertain when this began. Washout of the batch phase 13C labels continues.
  8. 13:51 19-Apr-2022 (t = 18.577 d) Chemostat 13C labeling phase of the experiment started. Analysis of PO13C during the washout phase (started at t = 5.737 d) shows del13C effectively back to background levels. For all chemostats except the 12C control (MC1), one of the 5 carbon sources have been replaced with their 13C equivalent. Like the batch phase, we will track the isotopes' incorporation into bacteria, protists and viruses (hopefully).
  9. 15:17 20-Apr-2022 (t = 19.637 d) Chemostats returned to 12C substrate feed due to pump failure and pulse of C to system. That pump has been removed from service and will be sledged hammered. Chemostats will again be flushed of 13C labels, and phase III will be attempted again in approximately 5 days.
  10. 15:49 24-Apr-2022 (t = 23.6596 d) Pump problem on MC2 where the Carbon + TE media began dripping into MC2 faster than desired. This resulted in vastly increased respiration for that MC. Situation was not fixed until 8:45 25-Apr-2022. (t = 24.3646 d).
  11. 11:42 26-Apr-2022 (t = 25.4875 d). Both C+TE and ASW media pumps were stopped for ~2 hours to address pump issues and replace pump tubing. Flow was resumed at 13:49 (t = 25.57597 d) following pump calibrations. MC2 still recovering for high C dose.
  12. 12:53 28-Apr-2022 (t = 27.53712 d) To insure community diversity between chemostats following several pump perturbations in the C+TE media feed, 180 mL from each chemostat was transfer to the other 5 chemostats, so that a total of 900 mL was removed from each MC and distributed to the other MCs.
  13. 14:00 28-Apr-2022 (t = 27.58333 d) In case there was loss of diversity in MCs do to perturbations, 600 mL of Siders Pond water was added to each MC. The Siders Pond water was collected yesterday morning (9:00 27-Apr-2022), passed through a 425 µm stainless steel sieve (#40), placed in the growth chamber at 20ºC, and bubbled with air to acclimated the community prior to re-inoculation of each MC.
  14. 10:58 4-May-2022 (t = 33.45715 d) Chemostat 13C labeling phase of the experiment has restarted. The chemostats have been flushed for 13.8 days with 12C substrates prior to change over to individually labeled 13C substrates, except for the control (MC1) that still has only 12C substrates. May the 4th be with us.
  15. 12:26 18-May-2022 (t = 47.51825 d) The last samples have been taken from Exp. 4 and media pumps have been switch off. Head space gas sampling and probes will remain on-line for a few days during the terminal batch phase of the experiment.
  16. 11:30 23-May-2022 (t = 52.458 d) All data collection has stopped and chemostats are being emptied.

Data Files:

Primary gas and probe data: Exp4.dat
External variable values: Exp4.oc
Experimental info file: Exp4.info
Compiled high resolution data from DO and pH probes: Exp4_DOpH.csv

Higher resolution (10 min) probe data from Hamilton's RS485 ModBus (files upload here every 1 hr):
(Note, these data are now plotted on the JS Graphics page and the data from all trace files are collated and available here: Exp4_DOpH.csv)
MC2 pH: Trace_2022-04-01_12h38_pH_243632-2353-1178.txt
MC2 DO: Trace_2022-04-01_12h38_DO_242453-02-210052.txt
MC5 pH: Trace_2022-04-01_12h38_pH_243632-2353-1176.txt
MC5 DO: Trace_2022-04-01_12h38_DO_242453-02-210053.txt

Experimental Equipment:

MicrocosmsInstruement Rack
Image (left) shows the six  Bellco Glass 3 L  microcosms (MCs) and associated Masterflex Pumps on 21-Mar-2022 just prior to start of Exp 4. Instrument rack and environmental chamber housing the MCs (image on right) is equipped with MKS mass flow controllers, Oxigraf O2 and CO2 laser diode absorption spectrometer, Nafion drier and a computer for monitoring and experimental control via Valco selector valve, ASCO solenoid valves, Weeder Technologies WTDOT and WTAIN digital IO boards and Comtrol RocketPort serial card.  Monitoring and control software is, still, written in Intel Fortran (gigidy) and all com is via RS232 except ModBus is also used for probes.

Reactor dissolved oxygen (DO) and pH are monitored with Hamilton VisiFerm DO Arc 325 and EasyFerm Plus Arc 325 probes, respectively.  Probes are controlled and calibrated via Hamilton Device Manager software via a USB-ModBus RS 485 Converter.  Data from probes are logged from 4-20 mA signal outputs that are converted to 0-10 V with Wayjun Signal Isolators and digitized with a Weeder Tech WTAIN A2D board (data is also logged via ModBus at higher frequency). 

The microcosms are sparged with air at 10 sccm (10.7 mL min-1 at 20ºC) in a Conviron PGR15 environmental chamber that is currently operated at 20 ºC.  A closed gas sampling loop regulated at a flow rate of 270 mL min-1 by a Hargraves (now Parker) mini pump (B.1F15E4.A12VDC) delivers reactor headspace gas to analyzer.Sample loop has an approximate volume of ~50 mL. Reactor gas headspace is approximately 4.7 L. A Masterflex L/S Digital drive connected to Easy-Load 3 and cartridge peristaltic pump heads provides media flow to and from bioreactors using various sizes of L/S Tygon A-60-F tubing in the pump heads and Nylochem nylon-12 1/8" tubing for gas an liquid transfers.

Software

Program MonitorMCP is used for monitoring instruments and gas control while program datFile2CVS is used to convert *.dat files to *.csv files used by HighCharts. Data is collected every hour, backed up to a server, converted to csv files and copied to web server for near real-time display here. Note, MonitorMCP makes use of Intel Visual Fortran's serial IO library for RS232 communication:
Graphics are generated by Tecplot and HighCharts

Photos

Delay of game. Well, we were going to start Exp 4 today (29-Mar-2022), but the refrigeration unit on the growth chamber had other plans. MBL's crack HVAC team is working on it.
Refrigeration unit under repair
(31-Mar-2022: The growth chamber is back on-line. Thanks Ben!)

Dr. Ahern collecting water sample from Siders Pond on 31-Mar-2022 for inoculation of bioreactors. It was a bit windy.
Olivia Ahern collecting water sample from Siders PondSiders Pond water sample collection location