**2. Materials and Methods**

Samples of healthy and diseased *L. baikalensis* and *B. intermedia* were collected by SCUBA in Southern (Bolshie Koty 51◦53 54.4" N, 105◦04 15.3" E and Chertov most 51◦54 41.9" N, 105◦13 29.2" E) and Northern (Severobykalsk 55◦60 93.4" N, 109◦34 80.3" E) basins of Lake Baikal in 2015 and 2006 at 10m depth. The description of the samplesare shown in Table 1. Sponge samples were frozen in liquid nitrogen for molecular analysis and were fixed in 70% ethanol for morphological examination. Spicule and skeleton preparation were performed as previously described [27] and were examined using an Olympus CX22 microscope. DNA extraction was performed using the RIBO-sorb RNA/DNAextraction kit (InterLabService, Moscow, Russia). DNA quality was assessed by running a subsample on a 1% agarose gel and the quantity of DNA was measured using a NanoVue (GE Healthcare).

**Table 1.** A detailed experimental determination of the effect of healthy and bleached freshwater sponges *Lubomirskiabaikalensis* and *Baikalospongia intermedia* collected from Baikal Lake, Russia.


The V3-V4 region of the 16S rRNA gene was amplified with primers 341F-806R [45] using Illumina MiSeq 250 bp chemistry. The PCR conditions were as follows: initial denaturation at 95 ◦C for 2min; followed by 25 cycles: denaturation at 95C for 30 s, anneal at 55C for 30 s, extension at 72C for 30 s; the final extension at 72C for 8 min. The complete dataset of seven paired-end healthy (LBH1, LBH2, LBH3, BIH1) and bleached (LBB1, LBB2, BIB1) freshwater sponge samples were generated by a fastQ file format. The primary analysis of NGS sequencing data, the removal of short and chimeric sequences, clustering in OTUs (operational taxonomic units), an assessment of biodiversity by calculating ACE, Chao1, and Shannon indices were carried out using the Mothur v.1.22.0 program (http: //www.mothur.org, accessed on 1 November 2021). The Pyrosequencing pipeline program (http://pyro.cme.msu.edu, accessed on 1 November 2021) was used to determine species diversity and taxonomic composition and to compare communities. Data cluster analysis was performed using the Complete Linkage Clustering program, which is part of the Pyrosequencing pipeline.

The metagenomic analysis was performed using the MG-RASTserverV4.0.3 software package [46]. Sequence similarity search of the 16s rRNA gene was performed against the available biological database SILVA. The best hits were classified based on the percentage of identity and sequence query coverage. The rarefaction curve plot was used to annotate the species richness among the samples based on Rarefaction Curve Analysis [47–50]. A genus-level relative abundance of an operational taxonomic unit (OTUs) was computed and compared for a better understanding of the composition of a microbial community among the selected samples.

The relative abundance of OTUs at genus level comparison was performed using the Morpheus online tool (https://software.broadinstitute.org/morpheus/, accessed on 1 November 2021). We carried out the clustering heat map with a complete linkage method using the Euclidian distance metric [51], and One minus Pearson correlation metric [52] implemented in the Morpheus tool. The genus identified in OTUs from seven selected samples was further compared with five-set and two-set interactive Venn diagrams using the InteractiVenn tool (http://www.interactivenn.net/, accessed on 1 November 2021).

#### **3. Results**
