Fig. S1. Characterization of fspA KO cells. fspA KO cells obtained from the random mutagenesis screen had the REMI-mutagenic vector (pSC) inserted 98 nucleotides downstream from the start codon of DDB_G0277237 (A). To create by homologous recombination new fspA KO cells, we isolated the REMI vector with the genomic flanking sequences (obtained by digesting the genome with ClaI) and re-transfected it in WT cells. In (B), a schematic representation of the recovered KO plasmid cleaved at the ClaI site is shown: the vector comprises 2066 bp upstream from the BamHI insertion site (98 bp of fspA gene – depicted by a black box, plus 1968 bp upstream to fspA start codon – represented by a grey box) and 656 bp downstream; the pSC vector, containing the blasticidin resistance cassette, is in white.

Screen for KO cells was done by PCR using a combination of distinct pairs of oligonucleotides (C), allowing us to verify both gain and loss of signal in KO cells (D, E). Arrows in A and B indicate the position of the screening oligos.

Fig. S2. The organization of the endocytic pathway is unaltered in fspA KO cells.

A. Endosomal pH was determined as the fluorescence ratio of two internalized probes (pH-insensitive and pH-sensitive dextrans) at different chase times compared with a calibration curve. No significant difference was seen in acidification and neutralization of endosomal compartments between WT and fspA KO cells.

B. Recycling of fluid phase was measured in cells fed with Alexa-647 for 30 min, washed and chased for up to 180 min. Fluid phase remaining inside endosomal compartments was assessed in aliquots taken every 30 min. Internalized fluorescence at time 0 was defined as 100%, and no significant difference was observed between WT and fspA KO cells.

C. Phagosomal proteolysis was determined as the fluorescence ratio of DQG and Alexa-594 attached to the surface of latex beads. DQG is quenched when linked to the beads by BSA; degradation of BSA in the proteolytic phagosomal environment promotes the release and the de-quenching of DQG fluorescence. Alexa-594 shows no changes in fluorescence during the transit inside the phagosome. The fluorescence ratio in WT cells after 120 min was defined as 100%, and no significant difference was seen in phagosome proteolysis between WT and fspA KO cells.

D. The enzymatic activity of lysosomal N-acetyl glucosaminidase was measured in the cellular lysate (intracellular) and in the cell culture medium (extracellular). Values were normalized to the total activity measured for WT cells; no significant difference was observed between WT and fspA KO cells.

E. Average speed of WT and kil2 KO cells in the presence of K. pneumoniae and folate. No significant difference was observed between WT and kil2 KO cells. **P < 0.01, to control condition (in phosphate buffer).

Fig. S3. Presence of surface and internal markers in surface biotinylated fractions. To assess the amount of different markers in the surface biotinylated samples, one surface membrane protein (SibA) was used as positive control, and two internal proteins (cytoplasmic actin and ER-resident PDI) were used as negative controls. The Western blot of a representative experiment is shown; all lanes are from the same gel and have the same exposure time. For actin and PDI, 20% of the total biotinylated fraction volume was loaded, and dilutions of the lysate (from 0.025% to 0.005% of the total lysate volume) were used to estimate the amount of biotinylated proteins. For SibA, 2% of the total biotinylated fraction volume and 1% to 0.025% dilutions of the total lysate volume were loaded.

Table S1. Primers for semiquantitative RT-PCR.

Table S2. List of bacterial strains.

Table S3. UniProt accession numbers of the orthologues used for phylogenetic reconstruction (Fig. 2B).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.