**3. Results**

## *3.1. Encrichment Cultivation*

Enrichment cultures for diazotrophic photoautotrophs carried out at di fferent temperatures using inoculum from four di fferent biocrusts were very revealing. Only heterocystous cyanobacteria were enriched for in our medium free of nitrate and ammonium, and all 994 colonies examined belonged to one of the three major clades known from biocrusts: *Nostoc* spp., *Tolypothrix* spp., and *Scytonema* spp. [19], as determined by microscopic inspection. The relative proportions obtained, however, were strongly dependent on the temperature of incubation (Figure 1). The composition of the enrichments at 4 ◦C was significantly di fferent from those growing at 25 ◦C (PERMANOVA pseudo-F: 6.22 df: 22 *p* ≥ 0.001) and 30 ◦C (PERMANOVA pseudo-F: 9.36 df: 22 *p* ≥ 0.001); the same was true for the comparison of 25 and 30 ◦C (PERMANOVA pseudo-F: 6.43 df: 22 *p* ≥0.001). *Scytonema* spp. made up the majority of the colonies at 30 ◦C, whereas *Tolypothrix* spp. was preferentially selected for at 4 ◦C. *Nostoc* spp. had a slight advantage at lower temperatures as well. This was so regardless of the origin of the crusts used for inoculation, in that there was no significant e ffect on outcomes by location (PERMANOVA, *p* ≤ 0.2; full dataset presented in Table S2).

**Figure 1.** Relative proportion of colonies assignable to *Scytonema* spp., *Nostoc* spp., and *Tolypothrix* spp. in enriched cultures obtained on medium without combined nitrogen as a function of incubation temperature.

#### *3.2. Temperature Range for Growth (or Survival) of Isolated Strains*

All cyanobacterial strains (tested in medium without combined N) showed robust growth at 15 and 25 ◦C, while none grew at 45 ◦C (Figure 2), the lower limit of moderate thermophilly. Formally then, all these strains were mesophiles with respect to temperature. At 4 ◦C, all *Tolypothrix* spp. strains grew well, while only one *Scytonema* sp. strain did. At this temperature, three *Nostoc* spp. strains

did not grow, while five strains were in apparent stasis (they neither grew nor show signs of cellular degradation). At 30 ◦C four out of eight *Tolypothrix* spp., nine out of ten *Nostoc* spp., and eleven out of twelve *Scytonema* spp. strains grew well. At 35 and 40 ◦C, no *Nostoc* spp. or *Tolypothrix* spp. strains grew, while eleven out of twelve *Scytonema* spp. did.

**Figure 2.** Temperature range at which the studied cyanobacterial strains can grow or survive under diazotrophic conditions. Colored rectangles indicate positive growth; hatched rectangles indicate stasis (no growth, but no obvious deterioration).

#### *3.3. Upper Temperature Limit for Growth and N2-Fixation*

We looked at growth (and survival) responses more in detail as a function of nitrogen source (N2-fixing vs. non N2-fixing conditions) in the upper range of temperature (35 and 40 ◦C) in an effort to infer if N2-fixation was the most sensitive cellular process determining the observed outcomes. Figure 3 shows the biomass yield of the 30 cyanobacterial strains after 30 days of growth cultivated in medium without combined nitrogen (nitrogen-free) and nitrogen replete media.

Our results show that providing a source of fixed nitrogen expanded the range for growth in many of them to 35 ◦C (*Scytonema* spp. JS003; *Nostoc* spp. HS002, HS094, HS013, *Tolypothrix* spp., HSN032, HSN033, HSN034) and in some cases, strains survived at 40 ◦C (*Nostoc* spp. HSN008, HS020, HS002, HS096, FB23, FB26; *Tolypothrix* sp. HSN042). The last column in Figure 3. shows the biomass yield in nitrogen replete minus that attained in medium without combined nitrogen at 35 ◦C, indicating a generalized positive effect on growth under nitrogen-replete conditions. For sixteen out of thirty strains this difference in growth was significant. This gives support to the contention that the upper temperature for growth may be determined by the sensitivity of N2-fixation in *Nostoc* spp. and *Tolypothrix* spp.; whereas it is not nearly as determinant for *Scytonema* spp.

**Figure 3.** Growth yield of N2-fixing cyanobacterial strains in the upper range of temperature for growth in nitrogen free (◦) vs. nitrogen replete media (•). Rhombuses indicate the mean and error bars indicate ± 1 SE, with *n* = 3. Vertical dashed lines indicate the amount of inoculum provided. At 40 ◦C, only observational data were recorded: colored rectangles indicate survival and white rectangles indicate death. \* Denotes statistically significant differences between growing conditions according to Wilcox's test.
