Previous   |    Next

doi:10.2204/iodp.proc.347.107.2015

Microbiology

Hole M0063E was drilled specifically for microbiology, interstitial water chemistry, and unstable geochemical parameters at Site M0063. Counts of microbial cells were made on board the ship by fluorescence microscopy using acridine orange direct counts (AODC) and by flow cytometry (FCM) using SYBR green I DNA stain. Additional AODC were also made during the Onshore Science Party (OSP). Further counts by fluorescence microscopy will be done after the OSP using both acridine orange and SYBR green I staining.

Microbial cells were enumerated at 67 sediment depths from independently taken samples for FCM (66 samples) and AODC (32 samples), both on the ship and during the OSP (Table T12).

The most striking observations for these data are the extremely high cell densities determined by AODC near the top of this hole and the fact that the two cell counting techniques do not produce similar results in the upper half of the hole (Fig. F23). The uppermost cell count at 1.22 mbsf, by FCM, was 3.72 × 10⁸ cells/cm³, whereas 6.53 × 10⁹ cells/cm³ was determined by AODC, a 17-fold difference. Maximum cell numbers in this hole were 1.40 × 10⁹ cells/cm³ at 15.44 mbsf for FCM counts and 8.77 × 10⁹ cells/cm³ at 2.65 mbsf for AODC. In the lower half of this hole (deeper than 43 mbsf), data from both counting techniques appeared similar. The minimum microbial cell numbers were determined as 6.31 × 10⁷ cells/cm³ at 83.42 mbsf by FCM and 3.24 × 10⁷ cells/cm³ at 71.52 mbsf by AODC.

Regression analyses of both data profiles indicated different trends shallower and deeper than 43 mbsf (Fig. F23). The decrease in cell numbers with depth was significantly steeper shallower than 43 mbsf compared to deeper than this depth, both for FCM (F = 8.14; degree of freedom [df] = 1.62; P < 0.01) and AODC (F = 14.12; df = 1.28; P < 0.001). In the upper 43 m, the change in cell numbers with depth was ~2.4 times steeper for AODC compared to FCM. Deeper than 43 mbsf, regression lines from the two depth profiles were not significantly different from each other (F = 0.818; df = 1.29 [not significant]), which was confirmed by a paired sample t-test (t = 0.008; df = 13 [not significant]).

The reasons for the split in the data at 43 mbsf do not seem to be directly related to interstitial water chemistry (Fig. F23), as there are no abrupt changes in either alkalinity or salinity at this specific depth. It is, however, interesting to note that the highest cell numbers are observed in the upper part of this hole (to 27 mbsf), where there are the highest levels of alkalinity, a product of microbial degradation of organic carbon. The profile break at ~43 mbsf seems to be related to stratigraphy with this depth being the approximate transition between laminated and well sorted clays of Unit III above and iron-sulfide laminated clays of Unit IV below (determined in “Lithostratigraphy” as being ~41 mbsf). The lithologic change at around 41 mbsf was observed in cores from Holes M0063A, M0063C, and M0063D, whereas cores from the microbiology hole (M0063E) were so heavily sampled for microbiology that the remaining core was too disturbed for good stratigraphic description.

It is striking that the two cell counting methods produce two such different depth profiles. All samples taken for AODC were counted from the surface to 5 mbsf and were also counted from 28.2 mbsf to the base of the hole. In the 23 m deep interval between these two depths, 21 samples were taken for counting by AODC, but only 5 of these were actually processed because of difficulties with preparing the other 16 samples for counting. After staining, cells were observed in large clumps of “fluff,” which made counting of individual cells impossible. An improved processing technique will need to be developed to deal with these samples. This fluff is likely to be bacterially derived exopolysaccharide causing cell clumping. It is of note that this depth interval of difficult-to-count samples coincides with both the largest differences between the two counting techniques and the presence of very organic rich sediments (Fig. F17B). Over this depth interval, all samples for FCM counting were processed and counted, yet there are two major observations that may explain the reduced counts. First, after sonication, the samples were filtered through a 40 µm filter and any bacterial clumps that had resisted sonication would be filtered out of the sample, thus significantly reducing the final count. Second, casual observation suggested that this depth interval in the sediment had large numbers of very small cells of ~0.1 µm diameter that FCM could not detect. Missing these small cells caused a reduction in the FCM count. Figure F24 shows samples stained by SYBR green I. The fluff was visualized but with too many cells to count. After sonication and HF treatment, both very small cells and clumps of cells can be seen. We estimated that >20% of total cells were 0.1 µm in diameter or perhaps even smaller.

As reported for Sites M0059, M0060, and M0061, cell numbers were very high and, with one exception, all cell counts exceeded the global regression (Fig. F23). The maximum deviation from the global regression was at 15.44 mbsf in the profiles from both techniques. The FCM count was 80-fold higher, and the AODC technique gave a result 265-fold higher.

When the data from both techniques are plotted against each other, results from deeper than 43 mbsf cluster around the line of x = y (Fig. F25). Data are too clustered for a regression line to be calculated and compared to the x = y line. Results from shallower than 43 mbsf clearly deviate from the x = y line.

PFC was above detection in the liner fluid and exteriors of all cores, indicating continuous PFC delivery into the borehole (Table T13). Liner fluid PFC concentrations fluctuated over 2–3 orders of magnitude (Fig. F26), indicating variations in the rate of PFC injection or mixing into the drilling fluid stream. Generally, the measured PFC concentrations were considerably below the target concentration of 1 mg PFC/L. Despite the variations, PFC was above detection in the vast majority of core halfway and interior sections. Interestingly, and contrasting with the other stations where contamination was most pronounced near the top of the hole, the uppermost cores, 347-M0063E-1H and 2H, have no detectable contamination in the interior. This might in part be due to low PFC concentrations in liner fluids in these cores, which would lower the PFC detectability in the cores compared to cores with high liner fluid PFC concentrations. Deeper than these top cores, and as at the other stations, there is no clear depth- or lithology-related trend in the level of contamination (Table T13; Fig. F23). The most highly contaminated cores are Cores 347-M0063E-3H, 4H, 30H, 31H, and 42H, which have potential contaminant cell densities of >10⁴ cells/cm³ in core interiors (Fig. F26).

Based on the PFC data and calculated contamination with microbes from drilling fluid, Cores 347-M0063E-1H, 2H, 16H, and 17H, in which PFC was below detection, are suitable for microbiological analyses. Moreover, the interiors of Cores 5H, 10H, 12H, 25H, and 41H, have calculated potential contaminant cell numbers <100 cells/cm³ sediment and are thus at most moderately contaminated (Table T13). The in situ communities of microbial cells, 10⁸–10⁹ cells/cm³, are more than a million-fold more abundant than this potential contamination level.

Top of page   |    Previous   |    Next