Tetraspanin Cd9b and Cxcl12a/Cxcr4b have a synergistic effect on the control of collective cell migration.

Collective cell migration is essential for embryonic development and homeostatic processes. During zebrafish development, the posterior lateral line primordium (pLLP) navigates along the embryo flank by collective cell migration. The chemokine receptors, Cxcr4b and Cxcr7b, as well as their cognate ligand, Cxcl12a, are essential for this process. We corroborate that knockdown of the zebrafish cd9 tetraspanin orthologue, cd9b, results in mild pLL abnormalities. Through generation of CRISPR and TALEN mutants, we show that cd9a and cd9b function partially redundantly in pLLP migration, which is delayed in the cd9b single and cd9a; cd9b double mutants. This delay led to a transient reduction in neuromast numbers. Loss of both Cd9a and Cd9b sensitized embryos to reduced Cxcr4b and Cxcl12a levels. Together these results provide evidence that Cd9 modulates collective cell migration of the pLLP during zebrafish development. One interpretation of these observations is that Cd9 contributes to more effective chemokine signalling.

Introduction

Cells can migrate individually or in groups, the latter is known as collective cell migration. During this process, cells exhibit coordinated behaviour, group polarisation and maintain cell-cell contacts [1]. This mode of migration is employed during embryonic development in the morphogenesis of multiple organ systems and is also important for effective immune responses. This mechanism contributes to several diseases including metastatic cancer and rheumatoid arthritis [2].

The zebrafish lateral line consists of a series of mechanosensory organs (neuromasts), which are distributed along the lateral surface of the zebrafish body and connected to the central nervous system by afferent axons [3]. This arrangement of innervated neuromasts is achieved by the collective migration of a placodal primordium of approximately 100 cells, arising just posterior to the otic placode [4]. This posterior lateral line primordium (pLLP), then migrates along the horizontal myoseptum to the tip of the tail. During migration, clusters of cells are deposited from the trailing end of the pLLP, which ultimately differentiate into neuromasts.

Primordium migration is directed by expression of the chemokine Cxcl12a along the horizontal myoseptum, which is received by two chemokine receptors expressed within the primordium [5-7]. Cxcr4b is expressed in the leading two thirds of the primordium, where ligand binding induces Gβ1 signalling and actin polymerisation to promote a migratory phenotype [8]. Cxcr7 is expressed in the anterior-most third and functions as a ligand sink, allowing the formation of a local Cxcl12a gradient. This results in directed migration of the primordium along the Cxcl12a pathway towards the caudal fin [9, 10]. There is evidence that the membrane environment of chemokine receptors, including Cxcr4 and Cxcr7, strongly influences their signalling properties. This includes homodimerisation, oligomerisation and heteromerisation with other membrane receptors [11]. In addition, chemokine receptors have been linked to membrane microdomains where the lipid environment strongly modulates function [12].

Tetraspanins are a large family of small integral membrane proteins, which have been shown to organise neighbouring membrane proteins into complexes called tetraspanin enriched microdomains. This is often referred to as the tetraspanin web, as different interactions build to form dynamic signalling networks that often induce similar functional outcomes [13]. Thus, tetraspanins are associated with a wide variety of cellular functions including signalling and cell migration [14]. For example, the tetraspanin CD9 is downregulated in many human cancers including lung, breast and ovarian, and reduced CD9 expression is related to colon cancer metastasis [15]. In particular, CD9 has been shown to regulate, and be regulated by, CXCL12-CXCR4 signalling [16-18]. CD9 was found in close proximity to CXCR4 on the membrane of B acute lymphoblastic leukaemia cells in vitro, and enhanced their CXCL12 dependent migration [18]. If regulation of CXCL12/CXCR4 signalling by CD9 is conserved, we hypothesised Cd9 would be strongly expressed in the zebrafish pLLP, where it might modulate Cxcr4 signalling and thus pLLP migration. Indeed microarray experiments have localised the zebrafish cd9b paralogue to the migrating primordium, whilst a morpholino (MO) experiment indicated loss of cd9b altered primordium structure at 36 hours post fertilisation (hpf) and reduced neuromast number at 48 hpf [2].

In this study we aimed to investigate further the role of zebrafish CD9 orthologues in Cxcr4b-mediated pLLP migration, through use of genetic knockouts. We show expression of both cd9a and cd9b paralogues in the primordium and confirmed the abnormal pLL phenotype seen previously with cd9b knockdown. The cd9b mutants showed delayed pLLP migration but did not fully replicate cd9b morphant pLL phenotypes. We therefore generated cd9a; cd9b double knockouts (cd9 dKOs). These showed both reduced migration of the pLLP and increased sensitivity to reduced levels of Cxcr4b and Cxcl12a, supporting a role for CD9 in regulation of Cxcr4 signalling.

Methods

Zebrafish maintenance

Adult wild-type zebrafish (WT), Tg(-8.0cldnb:LY-EGFP) and cd9a/b mutants were housed and bred in a regulated 14:10 hour light: dark cycle under UK Home Office project licence 40/3459 in the Bateson Centre aquaria at the University of Sheffield or in the Singapore IMCB zebrafish facility under the Biological Resource Centre oversight with project license IACUC 140924. Zebrafish were raised under the standard conditions at 28°C [19]. Ages are expressed as hours (hpf) or days (dpf) post fertilisation.

Morpholino injection

Antisense morpholino oligonucleotides (MO) were obtained from GeneTools LLC and re-suspended in MilliQ H2O to give a stock concentration of 1 mM and injected in one‐cell stage embryos. A Flaming/Brown micropipette puller was used to create micro-injection needles from borosilicate glass capillary tubes (0.5 mm inner diameter, Sutter). The PV800 Pneumatic PicoPump, as part of the micro-injection jig, was set up to release the required amount of injection material by adjusting the air pressure and air expulsion time. For the knockdown of cd9b two MOs were designed, a translation blocker with the following sequence (cd9b MO1): 5’-tttatgaggagaaacccaagactga-3’ and a splice site blocker (cd9b i2e3) with the following sequence: 5’-aacccctgaacacagagaaacaaca-3’, whilst the published mismatch MO was used 5’- tttccctgctgcttatacagcgatg -3’ [20]. For knockdown of cxcr4b and cxcl12a, the following sequences were used respectively, 5’-aatgatgctatcgtaaaattccat-3’and 5’-ttgagatccatgtttgcagtgtgaa-3’ [21].

Zebrafish mutant production

cd9b mutants were created from WT embryos using transcription activator-like effector nucleases (TALEN) and maintained on an WT background. TALEN (ZGene Biotech Inc., Taiwan) were provided in a pZGB4L vector, targeting the cd9b sequence 5’-ttgctctttatcttca-3’. Injected embryos were outcrossed and sequenced to identify mutations. A frameshifting deletion mutation (c.42_49del) was selected that caused premature termination within the first transmembrane domain (cd9ballele). CRISPR-Cas9 was used to create cd9a mutants using the gRNA sequence 5’-gagtgtatatcctcattgcgg-3’, which targeted the 3rd exon encoding part of the second transmembrane domain. cd9 dKO mutants were created by injecting the above cd9a gRNA and Cas9 RNA into cd9b embryos. These fish were screened for germline transmission by sequencing and backcrossed to cd9b mutants. An indel mutation deleting 4bp and inserting 8bp (c.180_187delinsTCGCTATTGTAT; cd9a) generated a frameshift mutation resulting in a premature stop codon in exon 3, which was predicted to truncate the protein before the large extracellular domain. Heterozygous fish of the same genotype were incrossed and adult F2 fish were genotyped to identify homozygous cd9b; cd9a (cd9 dKO).

In situ hybridisation

Embryos were raised at 28°C in petri dishes containing E3 solution. The E3 was changed daily, and any dead embryos removed. At 30–32 hpf, embryos were anaesthetised using tricaine and dechorionated before returned to fresh E3 solution. At the relevant timepoint embryos were fixed overnight at 4°C using 4% (w/v) paraformaldehyde (Sigma-Aldrich, UK) in phosphate buffered saline (PBS). Embryos were washed in PBS/0.05% (v/v) Tween 20 (PBST), then put through a methanol/PBS series using 30%, 60% and 100% (v/v) methanol before being stored in 100% methanol (Sigma-Aldrich) at -20°C. In situ hybridisation (ISH) was carried out as described by [22], except for the embryo digestion with proteinase K, for which 30–32 hpf embryos were digested with 10 mg/ml proteinase K at 20°C for 22 min. Primers used for PCR generation of the in situ probes are given in Table 1 below. The protocol was conducted with the embryos in 1.5 ml microfuge tubes for the first two days, after which they were held in 12-well plates for staining before transferring back to microfuge tubes for storage. Stained embryos were stored in the dark in 80% (v/v) glycerol.

Table 1

Primers for making ISH probes.

Probe	Primer name	Sequence (5’-3)
claudin b	Claudinb F	aaacgaaaaagcatggcatc
claudin b	Claudinb R	gaggctgtttcaaacgtggt
cd9a	CD9a F	gtcatattcgcggtcgaagt
cd9a	CD9a R	ctgcgagaacaacaaagcaa
cd9b	CD9b F	gttcgccacaagtgcctgat
cd9b	CD9b R	tacatgttactttctctccaaacaat

Time lapse imaging

Time-lapse recording was performed using an inverted Zeiss LSM700 Confocal microscope. All larvae were anaesthetised using 0.02% tricaine and then embedded on their side in 1% low melting agarose (Lonza) on a glass bottom dish (MalTEK Corporation) and covered with E3 supplemented with 0.02% tricaine. Separate Z-stack images covering the depth of the horizontal myoseptum were taken at specified intervals over a specified period and assembled into a final movie at a specified frame rate.

Statistics

Data distribution was first assessed for normality using a D’Agostino-Pearson omnibus K2 normality test on the experimental residuals, as well as creating a histogram of residuals. For normally distributed data, an ANOVA with Dunnet’s or Holms-Sidak multiple comparisons tests were used. For non-normally distributed data non-parametric tests, the Mann-Whitney U test or Kruskal-Wallis with Dunn’s multiple comparisons test, were used.

Results

Both cd9 zebrafish paralogues are expressed in the lateral line

Zebrafish possess two Cd9 paralogues, Cd9a [NP_997784] and Cd9b [NP_998593] which show 60% and 59% identity to human CD9 respectively and 63% identity to each other using multiple sequence comparison by log‐expectation (MUSCLE) [23]. Whole mount in situ hybridisation (WISH) using probes against cd9a and cd9b demonstrated expression of both paralogues in the migrating primordium at 36 hpf (Fig 1A–1D). Expression of both cd9a and cd9b was also observed in recently deposited neuromasts and was retained in neuromasts until at least 5 dpf (Fig 1G–1L).

Fig 1

cd9a and cd9b are expressed in the pLLP during zebrafish development.

A-F: Micrographs of WISH of cd9a and cd9b at 36 hpf with sense and anti-sense probes. (a-b) Overview shows staining in pLL; (c-d) higher resolution images show expression in the primordium; (e-f) sense probes show no staining. G-J: WISH of cd9a and cd9b at 48 hpf; (g-h) overview shows staining in pLL; (i-j) higher resolution images show expression in a neuromast. K-L: WISH of cd9a and cd9b at 5 dpf in a neuromast. Arrows indicate primordium (black) and neuromasts (red). Scale bar indicates 50 μm (white).

cd9b morphant phenotype

To evaluate if there was a role for either paralogue in pLLP migration, we initially focused on cd9b as its role in the primordium had been previously reported [2]. Two different MOs were designed to target cd9b; cd9b MO1 was a translation blocking MO which targeted the 5’ UTR of cd9b RNA and cd9b i2e3 was a splice blocking MO designed against the intron 2—exon 3 splice site. These MOs were injected independently into 1-cell stage embryos to ensure that they produced the same phenotype. Embryos injected with either cd9b MO1 or cd9b i2e3, but not the mismatch or uninjected embryos, showed a significant decrease in neuromasts deposited (Fig 2). For embryos injected with cd9b MO1, there was a decrease in the percentage of trunk length between the first and last neuromasts (Fig 2I). This suggests that the primordium is stalling prematurely or migrating more slowly. This recapitulates and expands on the work by Gallardo et al., 2010 which suggested a role for Cd9b in lateral line development [2].

Fig 2

cd9b MO caused a decrease in neuromasts deposited at 52 hpf.

A: Intron exon diagram of the cd9b gene with MO target sites. B-G: Representative images of WISH of claudin b at 52 hpf of (b) uninjected embryos, (c) embryos with 0.5 μM Mismatch, (d) 0.1 μM cd9b MO1, (e) 0.1 μM cd9b i2e3 MO1, (f) 0.2 μM cd9b MO1, (g) 0.2 μM cd9b i2e3 MO. H: Graph showing the number of neuromasts deposited at 52 hpf is significantly decreased after injection with both cd9b MO1 and cd9b i2e3 MO, with as low as 0.1 μM MO. I: Graph showing the percentage of trunk between the first and last deposited neuromast is decreased in cd9b MO1 morphants but not cd9b i2e3 morphants. ANOVA with Dunnett’s multiple comparisons test was used. n = minimum 12 lateral lines analysed per condition. ****p = <0.05.

cd9b mutant phenotype

To confirm this MO result, we created a TALEN mediated knockout of cd9b. A cd9b TALEN pair was designed by ZgeneBio using the program “TEL Effector Nucleotide Targeter 2.0”. The TALEN pair was designed to target Exon 1 and predicted to cut in the 1st transmembrane domain (Fig 3A). The TALEN pair was injected and after 72 hpf a proportion of embryos were genotyped. Injected embryos showed clear mosaicism around the TALEN cut site so embryos were raised to create mosaic adult F0s. F0s transmitting a mutation to offspring were out-crossed with WT fish and the resulting F1 offspring were raised. To create a cd9b homozygous mutant line with a single mutant allele, F1s with a c.42_49del mutation were selected, incrossed and the resulting offspring raised to adulthood. This allele (cd9b) was used as it had the largest deletion and caused the earliest nonsense stop codon. The 8bp deletion in the cd9b allele leads to a frameshift in exon 1, changing codon 15 from TTT (Phe) to CAA (Glu), then 22 aberrant amino acids followed by a stop codon (p.Phe15GlufsTer22) (Fig 3B). cd9b homozygous mutants are viable and showed normal development. Loss of cd9b in situ signal in the cd9b mutants suggested strong nonsense mediated decay (NMD) of the mutated allele (S1 Fig). Due to the phenotypes seen in cd9b morphants, it was expected that a lateral line phenotype would be seen in cd9b mutants with fewer neuromasts deposited. However, no significant difference was found in the number of neuromasts at 52 hpf (Fig 3). As cd9b is expressed throughout pLL development, migration of the primordium as well as lateral line structure was assessed at 36 hpf. Although the pLL shows no structural abnormalities in cd9b mutants at 52 hpf, it is possible that cd9b mutants show a lateral line phenotype earlier in development and have recovered by 52 hpf. At 36 hpf, cd9b mutants show delayed primordium migration, with the percentage of trunk traversed by the primordium reduced in mutants (Fig 3).

Fig 3

cd9b mutant does not recapitulate morphant phenotype, although primordium migration is delayed at 36 hpf.

A: Nature of the cd9b mutant allele showing TALEN site location within the intron-exon structure of the gene. B: The TALEN target sequence in exon 1 is shown in blue; the 8bp deletion in the cd9b allele is indicated under the WT sequence as dashes. This leads to a frameshift changing codon 15 from TTT (Phe) to CAA (Glu), then 22 aberrant amino acids (red lettering) followed by a stop codon (*). C: Schematic of the Cd9b protein with location of premature stop codon given by red arrow. The disulfide bonds between the conserved CCG motif and conserved cysteines are indicated by the dashed lines. EC1/2 = Extracellular domain 1/2, aa = amino acid. D-E: Sequence chromatograms of genomic DNA from (d) WT and (e) cd9b alleles. Deleted base pairs are underlined in red. F-I: Representative images of WISH of claudin b ISH on (f-g) WT and (h-i) cd9b mutants at time shown. J-K: Graphs quantifying pLLP measurements in WT and cd9b KOs; (j) the migration of the cd9b KO primordium at 36 hpf is significantly delayed compared to WT. (k) There is no significance in number of neuromasts deposited at 52 hpf. Unpaired T test on untransformed data. n = minimum 20. **** = p = <0.05.

Generation of a cd9a mutant

To ascertain a role for Cd9a, we targeted the cd9a gene using CRISPR/Cas9. A gRNA was designed to target cd9a in exon 3 (Fig 4A) and injected along with RNA encoding Cas9 into WT embryos to generate cd9a KOs. After screening for mosaicism, the CRISPR/Cas9 injected embryos were raised to maturity and screened for germline transmission of the cd9a mutation. F0s were outcrossed with WT fish to generate F1 cd9a single heterozygous mutants carrying an indel mutant allele (c.183_186delinsAT; cd9a). This indel led to a frameshift mutation, changing codon 62 from ATT (Iso) to TGC (Cys), then generating 54 aberrant amino acids followed by a premature stop codon before the EC2 domain (p.Iso62Cysfs54Ter) (Fig 4B and 4C). Heterozygous cd9a adults were in-crossed to produce the F2 generation. Adult F2s were viable and fertile and were genotyped to identify homozygous cd9a KOs. Loss of cd9a in situ signal in the cd9a mutants suggested strong NMD of the mutated allele (S2 Fig). The effect of cd9a KO on pLL development was investigated but overall, there were no major defects. No significant difference was found in the number of neuromasts deposited at 48 hpf or primordium migration at 36 hpf (Fig 4F–4K).

Fig 4

Generation of cd9a mutant does not show any abnormal pLLP development.

A: Nature of the cd9a mutant allele showing CRISPR site location within the intron-exon structure of the gene. B: The CRISPR target sequence in exon 3 is shown in blue; the 2bp deletion in the cd9a allele is indicated under the WT sequence as dashes. This leads to a frameshift changing codon 62 from ATT (Iso) to TGC (Cys), then 54 aberrant amino acids (red lettering) followed by a stop codon. C: Schematic of the Cd9a protein with location of premature stop codon, indicated by red arrow, at a predicted 115 aa. EC1/2 = Extracellular domain 1/2 D-E: Sequence chromatograms of genomic DNA from (d) WT and (e) cd9a allele. Deleted base pairs are underlined in red. F-G: claudin b expression in WT at (f) 30 hpf and (g) 48 hpf. H-I: claudin b expression in cd9a mutants at (h) 30 hpf and (i) 48 hpf. J-K: Graphs quantifying pLLP measurements in WT and cd9a KOs; (j) migration of the cd9a KO primordium at 36 hpf is similar to WT. (k) There is no significance in number of neuromasts deposited. Significance was assessed using an unpaired t test. N = minimum 7. Bars show mean +/- SD.

Generation of a cd9a; cd9b double mutants

Due to the redundant nature of tetraspanins, and cd9a having a similar mRNA expression pattern to cd9b, it was speculated that functional redundancy between Cd9a and Cd9b might be masking stronger phenotypes in the single mutants. Hence, we generated cd9a; cd9b double mutants. To do this, the same cd9a gRNA as above (Fig 4) and cas9 mRNA were injected into cd9b embryos. These fish were screened for mutation of cd9a as above and then backcrossed to cd9b mutants. An indel mutation in exon 3 was detected (c.180_184delinsTCGCTATTGTAT) that was predicted to lead to a frameshift mutation and an early stop codon, which truncated the protein before the EC2 domain (p.Leu61Alafs9Ter) (Fig 5A–5D). Transheterozygous fish were crossed and adult F2s were genotyped to identify doubly homozygous cd9b; cd9a individuals, from now on referred to as cd9 dKOs. Adult cd9 dKOs were viable and fertile and loss of cd9a and cd9b in situ signal suggested strong NMD of the mutated alleles (S3 Fig). It was expected that with both paralogues knocked out, the lateral line phenotype would now be as severe as cd9b morphants with fewer neuromasts deposited. However, no significant difference was found in the number of neuromasts at 48 hpf (Fig 5). Primordium migration was significantly delayed in cd9 dKO embryos at 36 hpf (Fig 5). This matches the phenotype shown by the cd9b single KOs. To see if the knockout of cd9a had an additional effect on the primordium delay, embryos from cd9b KOs were compared to those from cd9 dKOs (S4 Fig). The distance the primordium had migrated at 36 hpf was measured and no difference was seen between cd9b single KO and cd9b dKO.

Fig 5

Generation of cd9a mutant in the cd9b KO background to create a double cd9 KO mutant.

A: cd9a mutant allele sequence from exon 3, showing CRISPR target sequence in blue. The 4bp deletion (dashes) and 12bp insertion (lowercase) results in an 8bp insertion as indicated under the WT sequence. This leads to a frameshift changing codon 61 from CTC (Leu) to GCG (Ala), then 9 aberrant amino acids (red lettering) followed by a stop codon. B: Schematic of the Cd9 protein with location of premature stop codon for Cd9a (red arrow) and Cd9b (green arrow) at a predicted 70 aa and 36 aa respectively. EC1/2 = Extracellular domain 1/2. C-D: Sequence chromatograms of genomic DNA from (c) WT and (d) cd9a allele. Deleted base pairs are underlined in red and inserted base pairs in blue. E-F: claudin b expression in WT at (e) 30 hpf and (f) 48 hpf. G-H: claudin b expression in cd9 dKO mutants at (g) 30 hpf and (h) 48 hpf. I-J: Graphs showing pLLP measurements in WT and cd9 dKOs; (i) migration of the cd9 dKO primordium at 36 hpf is significantly delayed compared to WT. (j) There is no significance in number of neuromasts deposited. Significance was assessed using an unpaired t test. N = minimum 9, p = <0.001, n1 = neuromast 1 etc. Bars show mean +/- SD.

In order to further investigate the migration and organisation of the primordium in cd9 dKO embryos they were crossed into the cldnb:gfp transgenic line. Under the claudin b promoter GFP is expressed in the lateral line primordium and newly deposited neuromasts [5]. This allowed observation of the primordium migration in real time. Primordium migration appeared normal in the cd9 dKO embryos (Fig 6, S1 and S2 Videos). At the leading edge of the primordium filopodia could be seen as well as proliferating cells. Within the migrating primordium rearrangements occurred and rosettes were formed, increasing the length of the primordium. As the trailing cells decelerated and deposited cells as neuromasts, the primordium reduced in size. Quantification of various aspects of the primordium during deposition revealed similar results in WT and mutants (S5 Fig).

Fig 6

Primordium organisation appears normal in cd9 dKO(cldnb:gfp) embryos from 30 hpf.

A-B: Still images from a time-lapse recording of a migrating primordium in (a) WT and (b) cd9 dKOs. 0 minutes shows initial deposition as a proneuromast becomes distinct from the primordium and then 2 sequential images show (a’-b’) 20 minutes and (a’’-b’’) 60 minutes later. In the primordium of both WT and cd9 dKOs, filopodia can be seen at the leading edge (white arrow) and the formation of rosettes in the trailing edge (white dashed circle). Representative images from two videos that included two depositions each. Scale bar: 20 μm.

Cd9 interacts with the Cxcr4b/Cxcl12a pathway in the migrating zebrafish lateral line primordium

To investigate if cd9 had an effect on cxcr4b expression in the primordium, cxcr4b WISH was performed but no perturbation of cxcr4b expression was seen in the cd9b KOs (S6 Fig). Quantification showed no significant difference between the expression of cxcr4b in WT and cd9b KOs. To investigate if Cd9 was affecting migration in the zebrafish primordium through the Cxcr4b/Cxcl12a pathway, an experiment using MOs was conducted. Cxcr4b and cxcl12a MOs induce premature stalling of the primordium at concentrations of 1.5 mM and 0.5 mM respectively [21]. 100 μM of cxcr4b MO was found to be the highest concentration injected into WT embryos that did not induce a phenotype. However, this concentration of MO resulted in a further delay of the primordium in the cd9 dKOs (Fig 7A). For cxcl12a the highest concentration of MO that did not induce a phenotype in WT embryos was determined to be 12.5 μM but this had no further effect on the primordium migration in cd9 dKOs (S7 Fig). If the cxcl12a MO concentration was raised to 25 μM then the primordium in both WT and cd9 dKOs was delayed although the primordium in cd9 dKOs was significantly more delayed (Fig 7B).

Fig 7

cxcr4b and cxcl12a MOs further delay primordium migration at 36 hpf in cd9 dKO embryos.

A-B: Distance migrated by the primordium (labelled by claudin b ISH) were recorded for (a) injection of 100 μM cxcr4b MO, and (b) 25 μM cxcl12a MO. Significance was assessed using one-way ANOVA, * = p<0.05, ** = p<0.01, *** = p<0.001 **** = p<0.0001. N = minimum 15. Box extends from the 25th to 75th percentile and whiskers from 10th to 90th.

Discussion

In this study we investigated the role of Cd9 in the migration of the pLLP in zebrafish development. We first verified previous data showing cd9b expression in the primordium at 30 hpf [2]. We also noticed cd9b was expressed throughout recently deposited neuromasts at 48 hpf and 5 dpf. Our MO knockdown recapitulated previous results from Gallardo et al., 2010. cd9b morphants showed fewer neuromasts deposited, which suggested a role for Cd9b in the development of the lateral line. A homozygous cd9b mutant was created using TALENs but surprisingly, the morphant phenotype was not recapitulated in cd9b homozygous mutants. This lack of phenotype could be due to several reasons. Firstly, the phenotype seen in cd9b morphants could be due to off target effects of the MOs and not due to cd9b knockdown. However, the fact that two different MOs and researchers induced the same phenotype suggests it was not the lack of specificity [2]. Secondly, the Cd9b truncated protein could have some residual function, although ISH results showed a downregulation of mRNA suggesting NMD was occurring. Also, the cd9b mutation aborts the normal sequence after aa15, before the EC2 domain, so any translated protein would be expected to have minimal function. Finally, the cd9b morphant phenotypes may be specific but tetraspanin redundancy could rescue the phenotype in cd9b mutants. Tetraspanins are well known for their redundancy within the tetraspanin family and mouse knockouts of single tetraspanins often appear healthy and viable with mild phenotypes, whereas double tetraspanin knockout mice often show increased numbers and severity of phenotypes [24-31]. In zebrafish, this redundancy may be amplified due to the occurrence of a fish-specific whole genome duplication in teleost fish. This means that many tetraspanins, for which there is one mammalian ortholog, have two paralogs in zebrafish [32-36]. This is true for CD9, as mammals have a single CD9 whereas zebrafish have Cd9a and Cd9b. We demonstrated cd9a to also be expressed in the pLL in a similar pattern to cd9b. Furthermore, NMD of mutated genes, as we have seen in both our mutants, has been shown to invoke an upregulation of closely related genes as a genetic compensation mechanism [37].

Following this, tetraspanin cd9a was targeted for knockout using CRISPR technology. Cd9a homozygous mutants and cd9b cd9a double homozygous mutants were generated. Cd9a KO embryos displayed a normal lateral line phenotype with normal primordium migration and neuromast deposition. Initial analysis of cd9 dKO embryos using the same measurements revealed that the number of neuromasts deposited and spacing were normal yet the distance of primordium migration at 36 hpf was significantly delayed. It was expected that knockout of cd9a would result in a stronger pLL phenotype but cd9 dKOs showed a similar phenotype to the cd9b KOs. Furthermore, when primordium migration of cd9b single KO was compared to that of the cd9 dKOs there was no difference. These results suggest that the different phenotypes seen between the cd9b morphants and mutants are not due to compensation by cd9a. It remains possible that there is compensation by another tetraspanin [38]. Close relations to CD9 are the tetraspanins CD81 and CD63, which have been shown previously to substitute for CD9 in some circumstances. Overexpression of CD81 partially rescued the infertility phenotype seen in CD9 KO mice [39] and it was recently demonstrated that CD9 deletion in human melanoma cells was quickly compensated by CD63 expression upregulation [40]. It would be interesting to perform RNA-seq or SWATH-MS on the cd9 dKOs to identify upregulated RNA and proteins respectively [41].

Videos generated of cd9 dKOs with fluorescent claudin b showed primordium migration and internal organisation were normal. Measurements of the primordium after DAPI stain also revealed the primordium were similar sizes between mutants and WT at 36 hpf. Overall, it seemed the delay in migration at 36 hpf in the cd9 dKO embryos was not due to alterations in the primordium organisation. Therefore, the cd9 dKO phenotype appeared to be a migratory defect that could either be due to a change in the onset of migration, a change in the speed of migration or a combination of the two. CD9 has previously been reported to promote CXCR4b/CXCL12a signalling in mammalian cells so we decided to investigate a potential interference with this chemokine signalling pathway [16-18]. Results from the WISH analysis and quantification showed no obvious perturbation of cxcr4b expression in the cd9 KOs. To postulate if there was some interaction between Cd9 and the Cxcr4b/Cxcl12a signalling pathway a MO experiment was performed. It was theorised that a primordium lacking Cd9 would be more susceptible to disruption by sub-functional doses of either cxcr4b or cxcl12a MO than WT. Injection with a sub-functional concentration of 100 μM cxcr4b MO demonstrated that the cd9 dKO larvae were more sensitive to MO treatment compared to WT with significant retardation in primordium migration in the dKO. This suggests that there is some interaction between Cd9 and Cxcr4b within the primordium that promotes migration. Injection with cxcl12a MO was more ambiguous as at a sub-functional concentration of 12.5 μM neither the WT nor the cd9 dKO primordium migration was perturbed. When the dose of cxcl12a MO was increased to 25 μM a migratory delay was induced in both genotypes but the delay seen in the cd9 dKO primordium was significantly worse. Despite disruption in the presence of the MO, the primordium was still able to eventually migrate the full length in the cd9 dKO embryos, suggesting that Cd9 is not essential for maintaining Cxcr4b/Cxcl12a signalling but may play a regulatory or buffering role. A further experiment to check for interaction between Cd9 and Cxcl12a/Cxcr4b in the primordium would be to use immunohistochemistry to investigate co-localisation. Unfortunately, there are no tetraspanin, Cxcl12a or Cxcr4b antibodies available in zebrafish. However, two plasmids are available that encode cxcr4b-egfp or cxcl12a-venus constructs [42]. An interesting experiment would be to inject embryos with these plasmids and visualise the distribution of fluorescent versions of the proteins in the cd9 dKOs.

We mentioned that the migratory defect could be due to a change in the onset of migration. This could be further investigated by performing claudin b ISHs at the onset and earliest stages of development of the posterior lateral line. On the other hand, to try and determine if Cd9b has a role in the speed of migration, ISHs could also be performed at various time points throughout primordium migration and speed could be calculated in a similar method to previous studies [43, 44].

We focused on a connection between Cd9 and Cxcr4b/Cxcl12 signalling in this paper because of previous research linking these proteins. However, Cd9, like other tetraspanins, contributes to cellular processes by organising molecules within the plasma membrane and there are many other signalling pathways involved in primordium development [45, 46]. The fundamental four are FGF, Wnt, Notch and chemokine signalling pathways and their specific and coordinated expression ensures the regular morphogenesis and migration of the pLL [47]. Even though the organisation of the primordium at high resolution appears normal, analysis of the expression of these crucial signalling molecules, through WISH or immunohistochemistry, during primordium development could reveal subtle changes in their organisation that may be affecting migration. Whilst Fgf signalling functions to determine rosette formation and deposition, Wnt signalling drives proliferation in the leading zone [46]. Proliferation analysis with BrdU incorporation could show if there are any changes in cell proliferation within the primordium, specifically in the leading edge. CD9 has also been demonstrated to downregulate expression of Wnt signalling pathways in cell culture [48, 49]. Another important component of collective cell migration is cell adhesion so migratory cells can pull insensitive cells in the same direction. Within the primordium cadherins mediate cell-cell adhesion between primordium cells and are required for directed and robust migration [50]. Cadherin-dependent intercellular adhesion regulated by tetraspanins plays an important role in tumour invasion and metastasis [51, 52]. CD9 has been associated with keratinocyte motility by regulating E-cadherin-mediated cell-cell contacts [53].

The use of computational models has been a valued tool to integrate analysis of the lateral line primordium acquired through microscopic, cellular, molecular and genetic analysis [47, 54]. These models allow researchers to edit parameters and observe their influences on the migrating primordium. This could help us choose the most appropriate avenue to explore next by checking what modifications are able to regenerate the phenotype we observe in the cd9 dKO embryos.

In conclusion, we propose that within the primordium Cd9b is functioning to compartmentalise signalling molecules, like Cxcr4b, to orchestrate and amplify signalling upon ligand binding but is not essential. This would explain why only subtle differences are seen between WT and cd9 dKOs.

cd9b is significantly decreased in cd9b KO embryos.

A-B: Representative images of cd9b WISH at 36 hpf in (a) WT and (b) cd9b homozygous embryos.

(TIF)

Click here for additional data file.

cd9a is significantly decreased in cd9a KO embryos.

A-D: Representative images of cd9a WISH on (a-b) WT and (c-d) cd9a homozygous embryos at time shown.

(TIF)

Click here for additional data file.

cd9a and cd9b are both significantly decreased in cd9 dKO embryos.

A-B: Representative images of cd9a ISH on (a) WT and (b) cd9 dKO mutants at time shown. C-D: Representative images of cd9b ISH on (c) WT and (d) cd9 dKO mutants at time shown.

(TIF)

Click here for additional data file.

cd9a is not compensating for cd9b in the primordium.

Graph showing the distance migrated by the primordium at 36 hpf is the same in cd9b KO and cd9 dKO embryos. Significance was assessed using an unpaired t test, N = minimum 17. Bars show mean +/- SD.

(TIF)

Click here for additional data file.

Primordium shape is normal in the cd9 dKO embryos at 36 hpf.

A: Representative image of a DAPI-stained primordium from a WT zebrafish at 36 hpf with red outline to show measured area. B-E: Graphs showing measurements of the primordium (b) height, (c) width, (d) width to height ratio, and (e) area in WT and cd9 KO embryos. The height was measured three times at equal points along the primordium and then averaged. The width of primordium was measured from the two furthest points along the middle of the primordium. Width was then divided by height to generate a ratio. The area was circled using the freehand selection and measured. Cells were counted using the multipoint tool on Image J software. Significance was assessed using an unpaired t test, p = <0.05, N = minimum 11. Bars show mean +/- SD.

(TIF)

Click here for additional data file.

cxcr4b expression is not altered in the pLLP of cd9 dKO embryos at indicated stages.

A-D: Representative images of cxcr4b ISH in the primordium of (a,c) WT and (b,d) cd9 dKO mutants at time shown. Scale bar: 50 μm. E-F: Graphs showing measurements of (e) length and (f) height of cxcr4b expression in the primordium of WT and cd9 KO embryos at indicated stages. Length measurements were taken along the middle of the embryo between the two furthest points of expression within the primordium. Height measurements were taken between the two highest points of expression within the primordium. Significance was assessed using an unpaired T test. N = minimum 13 for 24 hpf and 11 for 30 hpf. Bars show mean +/- SD.

(TIF)

Click here for additional data file.

12.5 μM of cxcl12a MO does not further delay primordium migration at 36 hpf in the cd9 dKO embryos.

Distance migrated by the primordium (labelled by claudin b ISH) was recorded for injection of 12.5 μM cxcl12a MO. Significance was assessed using one-way ANOVA. N = minimum 13. Box extends from the 25th to 75th percentile and whiskers from 10th to 90th.

(TIF)

Click here for additional data file.

WT primordium migration.

WT(cldnb:gfp) from 30–38 hpf. Images taken every 1 minute with a 20x objective. Movie length: 420 min.

(MP4)

Click here for additional data file.

cd9 dKO primordium migration.

cd9 dKO(cldnb:gfp) from 30–36 hpf. Images taken every 1 min with a 20x objective. Movie length: 360 min.

(MP4)

Click here for additional data file.

9 Jul 2021

PONE-D-21-17584

CD9 tetraspanins convey robustness to CXCR4b signalling during collective cell migration

PLOS ONE

Dear Dr. Roehl,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Your manuscript has been seen by two expert reviewers in the field. One reviewer highlights a shortcoming of the figures uploaded to the PLoS One server. Please check that the uploaded figures represent the final versions and correspond to the respective manuscript text. Please also carefully take into consideration the comments made by Reviewer 2, in terms of interpretation of your findings and the contribution of FGF signaling.

Please submit your revised manuscript by Aug 22 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Christoph Winkler, Dr.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

3. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript contains plenty of errors and most of the figures do not show all images/panels described in the text. At this stage, the manuscript is not competent for publication and should be prepared again with more attention.

Reviewer #2: The paper by Marsay et al revisits the role of the tetraspanin CD9 in the primordium. A previous microarray analysis study in 2010 had shown that CD9 is expressed in the primordium and that that CD9 morpholino injected embryos have delayed migration and fewer neuromasts. This study examines expression of cd9b and the closely related cd9a with whole mount in situ hybridization, confirming expression of these cd9 paralogues in the primordium. It recapitulates previous results showing that cd9 knockdown delays migration and reduces the number of neuromasts. It goes on to show that a TALEN generated mutant, predicted to have loss of CD9b function, does not recapitulate the morphant phenotype and though migration is delayed the number of neuromasts is not altered. The authors go on to determine if the difference between the morphant and mutant phenotype is related to the redundant role of cd9a by generating a CRISPR based mutant expected to have of function. However, the phenotype of the double mutant is not significantly different from that of cd9b alone. Finally, the authors explore the potential role of CD9 in chemokine dependent migration in the primordium by showing that a subcritical knock down of cxcr4b or cxcl12a results in slowing of migration only in the background of a double mutant loss of cd9b and cd9a.

Overall, though the study documents their results effectively, the observations only add modestly to what was previously known about cd9b function. Beyond showing that, unlike morphants, mutants do not have a reduction in the number of neuromasts, they also show that the difference between cd9b morphant and mutant phenotypes cannot be accounted for by redundant function of cd9a.

The authors demonstrate a reduction in primordium migration specifically in double cd9b/cd9a mutants but not wild-type embryos following partial knock down of cxcr4b or cxcl12a. They interpret this to suggest cd9 functions to make cxcr4b function more robust. While strictly speaking it may be correct to say the function is less “robust”, the analysis is minimal and it is remains unclear what the role of cd9 in chemokine might be or whether the exaggeration of deficits seen specifically in the double mutants necessarily result from a role of cd9 in determining efficacy of chemokine signaling. The deficits seen in cd9 double mutants are not typical of those seen when chemokine signaling alone is compromised as it is not even clear whether primordium migration is slower or simply delayed in its initiation. A deficit in chemokine signaling alone might be expected to slow migration and reduce spacing between deposited neuromasts, however, this is not what is seen in the mutants. Furthermore, as collective migration of the primordium is not just dependent on chemokine signaling but also, at a minimum, on Fgf signaling, it is not clear if the observed deficits in migration seen specifically in cd9 mutants might also be seen when Fgf signaling dependent migration is compromised. The authors should test if compromised migration, observed in cd9 double mutants in the background of compromised cxcr4b/cxcl12a function, is also seen when migration is slowed with subcritical doses of an Fgf signaling inhibitor. Are synergistic effects on migration only seen following manipulation of chemokine dependent migration and not following subcritical interference with Fgf-dependent migration? The literature suggests quite a broad range of potential roles for Cd9 and it would be premature to conclude that changes following loss of cd9b function demonstrated in this study are necessarily related to a role for CD9 in facilitating chemokine signaling.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: PlosOne Review_Phan Quang Tien.docx

Click here for additional data file.

2 Oct 2021

Christoph Winkler, Dr.

Academic Editor

PLOS ONE

Manuscript no. PONE-D-21-17584

Manuscript title: Augmentation of chemokine signalling by CD9 tetraspanins facilitates collective cell migration

Dear Dr Winkler,

We were glad to learn that the reviewers found our work interesting and that, pending revisions, the manuscript is acceptable for publication in PLOS ONE. We modified our manuscript according to the comments of the reviewers as detailed in the point-by-point response below. The modified manuscript has highlighted modified text in red. We thank the reviewers for their constructive criticisms and hope that you will find the manuscript suitable for publication in PLOS ONE.

Sincerely,

Dr. Katherine Marsay

Detailed response to comments:

We would like to thank the reviewers for their constructive comments that helped us to improve the manuscript. As detailed below we modified the manuscript according to these comments.

Journal Requirements

Comment: 1. Please ensure that your manuscript meets PLOS ONE's style requirements.

Answer: High attention was made to ensure all PLOS ONE style requirements were met including: correct formatting in corresponding author and joint author information, correct acknowledgements, correct reference format.

Comment: 2. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016.

Answer: HR has validated his ORCID iD.

Comment: 3. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements.

Answer: Previous results referred to as “data not shown” have been added to Supplementary figures S5 and S6.

Reviewer #1:

Comment: The manuscript contains plenty of errors and most of the figures do not show all images/panels described in the text.

Answer: All figures were evaluated in detail and corrected to match figure legends. Gene and protein labels were corrected.

Comment: Throughout the manuscript, the use of upper and lower case letters to label protein symbols for fish and for human should be corrected according to standard gene/protein nomenclature.

Answer: All fish and human gene and protein labels have been corrected according to standard gene/protein nomenclature.

Comment: reference format e.g. line 34

Answer: All references were checked and reformatted correctly.

Comment: There were a few comments regarding figure 1 so I will address them together:

Image in A needs a different scale bar as it appears that the image in D is of different magnification than image in A. One image of the sense probes of cd9a and cd9b should be added. The embryo images in B and E appear younger than 48 hpf. It would be more demonstrative if authors can show alternative images for this time point. Although the image in A shows the labelling of primordium structure, a colocalization of cd9 in situ with a known marker of neuromast (such as ath1 or cldnb) would strongly support the author’s statement that cd9 genes are expressed in deposited neuromasts.

Answer: Figure 1 was redrawn to best illustrate cd9a and cd9b expression in the lateral line. For 36 hpf and 48 hpf a low and high magnification image is included so the expression can be seen in the posterior lateral line as a whole, as well as in detail in the primordium or neuromasts. The images of the neuromasts at high magnification show more clearly how cd9 is specifically expressed in the cell membranes of most cells in a deposited neuromast and supports the statement that both cd9 paralogues are expressed in deposited neuromasts. Scale bars are consistently 50 µM throughout the figure. As requested, one image of the sense probes of cd9a and cd9b were added.

Reviewer #2:

Comment: While strictly speaking it may be correct to say the function is less “robust”, the analysis is minimal and it is remains unclear what the role of cd9 in chemokine might be or whether the exaggeration of deficits seen specifically in the double mutants necessarily result from a role of cd9 in determining efficacy of chemokine signaling.

Answer: Our morpholino results show that a lack of Cd9 exacerbates cxcr4b knockdown and primordium delay. The reviewer is correct that we cannot assign a specific role of Cd9 to Cxcr4b in this pathway. We changed some wording throughout our manuscript to support our results while keeping the potential mechanisms of Cd9 broad. e.g.

New title: Augmentation of chemokine signalling by Cd9 tetraspanins facilitates collective cell migration.

Abstract: Together these results provide evidence that Cd9 modulates collective cell migration of the pLLP during zebrafish development, potentially through promoting Cxcxr4b signalling.

Comment: The deficits seen in cd9 double mutants are not typical of those seen when chemokine signaling alone is compromised as it is not even clear whether primordium migration is slower or simply delayed in its initiation. A deficit in chemokine signaling alone might be expected to slow migration and reduce spacing between deposited neuromasts, however, this is not what is seen in the mutants.

Answer: We did not wish to make the claim that Cd9 is acting solely on Cxcr4 signalling and apologize if this was not made clear. We have now been more explicit about this in the discussion and explained that we chose to investigate Cxcr4b/Cxcl12 signaling as CD9 has been previously reported to interact and regulate with this signaling pathway in the migration of mammalian cells.

Comment: Furthermore, as collective migration of the primordium is not just dependent on chemokine signaling but also, at a minimum, on Fgf signaling, it is not clear if the observed deficits in migration seen specifically in cd9 mutants might also be seen when Fgf signaling dependent migration is compromised. The authors should test if compromised migration, observed in cd9 double mutants in the background of compromised cxcr4b/cxcl12a function, is also seen when migration is slowed with subcritical doses of an Fgf signaling inhibitor. Are synergistic effects on migration only seen following manipulation of chemokine dependent migration and not following subcritical interference with Fgf-dependent migration?

Answer: The reviewer is correct that we failed to discuss other potential mechanisms of Cd9 inhibition that could result in a delayed migration. We expanded the discussion to review other signalling mechanisms that could be relevant to Cd9 such as the regulation of Wnt signalling pathways. Cd9 has been previously reported to downregulate this pathway in cell culture. Unfortunately, I am no longer in a position to perform experiments on the mutant fish but the reviewer provoked an interesting discussion for future investigation.

Comment: The literature suggests quite a broad range of potential roles for Cd9 and it would be premature to conclude that changes following loss of cd9b function demonstrated in this study are necessarily related to a role for CD9 in facilitating chemokine signaling.

Answer: The reviewer is correct that we failed to discuss other mechanisms aside from chemokine signalling that could be associated with tetraspanins such as Cd9. We expanded the discussion to review other tetraspanin related functions that could be affecting cell migration in the developing primordium such as Cadherin-dependent intercellular adhesion.

Submitted filename: Response to Reviewers v1.docx

Click here for additional data file.

3 Nov 2021

PONE-D-21-17584R1Augmentation of chemokine signalling by CD9 tetraspanins facilitates collective cell migrationPLOS ONE

Dear Dr. Roehl,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

As you will see from the reviewers' comments below, both reviewers agree that the manuscript has improved with the revisions made. On the other hand, they both still feel that title and parts of the abstract overstate the findings made in this study and request to rephrase your wording. I agree with both that the experimental evidence provided is not sufficient to directly link Cd9 function to expression or activity of Cxcr4b. Other interpretations of your findings, as outlined by reviewer 2, are possible. Both reviewers suggest alternative titles and text changes. I recommend that you consider their proposed changes. In your response, please also address the comment raised by reviewer 2 on the movies of LLP migration, whether a delayed start of the LLP is possible in mutants. In your response letter, please also describe the contributions of the newly added co-authors.

Please submit your revised manuscript by Dec 18 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Christoph Winkler, Dr.

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript aimed to investigate the function of Cd9 tetraspanins in pLLP migration and neuromast formation in zebrafish. First, the authors showed that knocking down of cd9b by morpholinos delayed pLLP migration and reduced the number of deposited neuromasts along the trunk. This observation confirmed the data previously shown by Gallardo et al. To confirm the observed phenotypes the authors went on to use cd9b mutant. Unexpectedly, cd9b mutant only partly recapitulated the phenotypes seen in morphants in which the pLLP migration was transiently delayed, while the number of neuromasts was not affected in the mutant. The authors also looked into the phenotypes of cd9a mutant and cd9a/cd9b double mutants. While no obvious phenotype related to pLLP migration and neuromast formation was seen in cd9a mutant, the double mutant showed similar phenotype as of cd9b single mutant. The different phenotypes of cd9b morphants and cd9b mutant were explained that there could have compensations by other members of the tetraspanin family.

In another attempt, the authors injected morpholinos blocking Cxcr4b and Cxcl12a which previously known to interfere with the primodium migration, into cd9 double mutants. The authors showed that even the low concentrations of morpholinos, which had been known to cause no effect on pLLP migration, could induce further delay of pLLP migration in cd9 double mutants.

Overall, although the new findings in this study is not abundant, the manuscript provides certain clues for future research on the potential interplays among members of tetraspanin family, and between cd9b with Cxcl12/Cxcr4 pathway in the control of primodium migration and neuromast formation.

Minor comments:

The title “Augmentation of chemokine signalling by Cd9 tetraspanins facilitates collective cell migration” is missleading as there is no direct data showing that Cd9 enhances Cxcl12/Cxcr4 expresion or activity which in turn facilitates collective cell migration. The authors may consider a more direct title, for instance “Cd9b tetraspanin and Cxcl12a/Cxcr4b convey synergistic effect on the control of collective cell migration”.

Line 311: typo “Cxcrb” change to “Cxcr4b”

Reviewer #2: Marsay et al have resubmitted a revised version of their original manuscript “CD9 tetraspanins convey robustness to CXCR4b signaling during collective cell migration” with a new title “Augmentation of chemokine signaling by CD9 tetraspanins facilitates collective cell migration”.

A primary issue raised with the original submission was that, with limited analysis of deficits following morpholino, TALEN and/or CRISPR manipulation of cd9a and cd9b function in the zebrafish, the authors were not in position to definitively link Cd9 function to chemokine signaling. No new experiments help to establish a stronger functional relationship between cd9 and chemokine signaling in the primordium and the authors have addressed this weakness of the paper by toning down the suggested link, which is an improvement. However, both the title of the paper and some of the text still in my mind would suggest to the casual reader of the paper a stronger link than has been established.

What is not in doubt is that partial knock down of the cxcr4b in a cd9 double KO has a significant effect on retarding the position of the primordium compared to a situation, while the same subcritical dose of morpholino has no obvious effect in a wild-type background. This confirms that while the deficit in cxcr4b caused by the morpholino alone is not adequate to slow migration, an additional deficit caused by the loss of cd9 function results in delayed migration. Such observations on their own are not adequate to further conclude that the exaggerated deficits in migration resulting from the combined interference with cxcr4b or cxcl12a with cd9 necessarily result from a role of cd9 in the chemokine signaling pathway when it is known that multiple signaling systems in combination with chemokines contribute to primordium migration. The authors do raise this issue in the discussion, however the title “Augmentation of chemokine signaling by CD9 tetraspanins…” still suggests a stronger link between cd9 and chemokine signaling than the authors have evidence for in the paper and I think both the title and the abstract should be further modified to prevent unintentionally perpetuating a potentially erroneous conclusion based on limited analysis.

The title should be changed. A subtly different but perhaps less misleading title could be:

“Loss of Cd9 exaggerates deficits in collective migration caused by interference with chemokine signaling in the posterior Lateral Line primordium”

Similarly, the abstract should be edited to simply remove or edit the last line:

Loss of both Cd9a and Cd9b sensitized embryos to reduced Cxcr4b and Cxcl12a levels.

Together these results provide evidence that Cd9 modulates collective cell migration of the

pLLP during zebrafish development.”

OR

Loss of both Cd9a and Cd9b sensitized embryos to reduced Cxcr4b and Cxcl12a levels.

Together these results provide evidence that Cd9 modulates collective cell migration of the

pLLP during zebrafish development. One interpretation of these observations is that Cd9 contributes to more effective chemokine signaling.

One additional issue that requires comment. The video of pLLP migration does not show any obvious deficits in migration and migration speed was never assessed or reported. Is it possible that the difference in position of primordium observed by the authors results from a delay in the start of migration. Do the authors have evidence that argues against this? If not, this possibility should also be raised in the discussion.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

7 Nov 2021

Christoph Winkler, Dr.

Academic Editor

PLOS ONE

Manuscript no. PONE-D-21-17584

Manuscript title: Cd9b tetraspanin and Cxcl12a/Cxcr4b have a synergistic effect on the control of collective cell migration

Dear Dr Winkler,

We appreciate that you saw the merit in our research and we were happy to incorporate the minor revisions according to the comments of the reviewers. They are detailed in the point-by-point response below and the modified manuscript has highlighted modified text in red. Previous revisions required additional experiments that were performed by Harsha Mahabaleshwar and Charmaine Ho Min. The main authors who performed the majority of the research are now in positions where they are unable to access the facilities associated with this research. Harsha and Charmaine aided the communication and organisation of the revisions between the authors despite us all being in different continents. They were vital for the latest version of the manuscript and therefore, they were added as additional authors to this manuscript.

We thank you and the reviewers for your patience and for giving us another chance to meet the publication criteria of PLOS ONE. We hope that you will find the revised manuscript suitable for publication in PLOS ONE.

Sincerely,

Dr. Katherine Marsay

Detailed response to comments:

We would like to thank the reviewers for their constructive comments that helped us to improve the manuscript. As detailed below we modified the manuscript according to these comments.

Journal Requirements

Comment: Please review your reference list to ensure that it is complete and correct.

Answer: All references were reviewed thoroughly to ensure they were complete, correct and that none had been retracted since the first draft of this manuscript. Two references were added in response to a comment by reviewer 2 as detailed later. Reference 15 was noticed as an incorrect reference and replaced from Charrin et al., 2003 to Ovalle et al., 2007. Reference 42 was noticed to be a duplication of reference 16 and deleted. No other changes to the reference list were required.

Reviewer #1:

Comment: The title “Augmentation of chemokine signalling by Cd9 tetraspanins facilitates collective cell migration” is missleading as there is no direct data showing that Cd9 enhances Cxcl12/Cxcr4 expresion or activity which in turn facilitates collective cell migration. The authors may consider a more direct title, for instance “Cd9b tetraspanin and Cxcl12a/Cxcr4b convey synergistic effect on the control of collective cell migration”.

Answer: I agree that although our data may provide clues for future research, our current observations on their own are not adequate to suggest a conclusive interaction between Cd9b and the Cxcl12/Cxcr4 pathway in the control of primordium migration. I appreciate the suggested alternative title and would be pleased to use a slightly modified version of this as the new title for our manuscript.

Comment: Line 311: typo “Cxcrb” change to “Cxcr4b”

Answer: Changed.

Reviewer #2:

Comment: The title should be changed. A subtly different but perhaps less misleading title could be: “Loss of Cd9 exaggerates deficits in collective migration caused by interference with chemokine signaling in the posterior Lateral Line primordium”.

Answer: I agree that although our current hypothesis is that Cd9 modulates collective cell migration of the pLLP through promoting Cxcr4b signaling, we cannot conclusively prove this from our current results. I appreciate the suggested alternative title but have decided to use the following as the new title for our manuscript: “Cd9b tetraspanin and Cxcl12a/Cxcr4b have a synergistic effect on the control of collective cell migration”.

The loss of Cd9b causes a significant delay in collective migration of the posterior lateral line, and also exaggerates deficits in collective migration caused by interference with chemokine signaling. I believe the chosen title describes the phenotype of cd9b KO as being similar to that of loss of Cxcr4b and also highlights the interaction I hypothesized from our preliminary chemokine interactions. I hope you agree but I am happy to change the title again if it is still not clear.

Comment: Similarly, the abstract should be edited to simply remove or edit the last line: “Loss of both Cd9a and Cd9b sensitized embryos to reduced Cxcr4b and Cxcl12a levels. Together these results provide evidence that Cd9 modulates collective cell migration of the pLLP during zebrafish development.”

OR

“Loss of both Cd9a and Cd9b sensitized embryos to reduced Cxcr4b and Cxcl12a levels. Together these results provide evidence that Cd9 modulates collective cell migration of the pLLP during zebrafish development. One interpretation of these observations is that Cd9 contributes to more effective chemokine signaling.”

Answer: I appreciate the suggestions and agree that the last line of the abstract should be edited to make the conclusions clearer and prevent perpetuating overconfident interpretations of results. The abstract was edited in accordance to the latter suggestion as so: “Together these results provide evidence that Cd9 modulates collective cell migration of the pLLP during zebrafish development. One interpretation of these observations is that Cd9 contributes to more effective chemokine signaling.”

Comment: One additional issue that requires comment. The video of pLLP migration does not show any obvious deficits in migration and migration speed was never assessed or reported. Is it possible that the difference in position of primordium observed by the authors results from a delay in the start of migration. Do the authors have evidence that argues against this? If not, this possibility should also be raised in the discussion.

Answer: The reviewer raises an important point. The delay in primordium migration could either be due to a change in the onset of migration, a change in the speed of migration or a combination of the two. The idea that differences in the observed position of the primordium could be due to a delayed start in migration was added to the discussion. Further experiments that could elaborate on this were also added and involved the addition of two references (44 & 45 - Matsuda et al. 2013, Valdivia et al. 2011). Their publications provide methods that could be used to calculate the speed of primordium migration.

Inserted/edit text is shown below:

Line 387-389: Therefore, the cd9 dKO phenotype appeared to be a migratory defect that could either be due to a change in the onset of migration, a change in the speed of migration or a combination of the two.

Line 415: We mentioned that the migratory defect could be due to a change in the onset of migration. This could be further investigated by performing claudin b ISHs at the onset and earliest stages of development of the posterior lateral line. On the other hand, to try and determine if Cd9b has a role in the speed of migration, ISHs could also be performed at various time points throughout primordium migration and speed could be calculated in a similar method to previous studies [44,45].

Submitted filename: Response to Reviewers v1 05112021.docx

Click here for additional data file.

9 Nov 2021

Tetraspanin Cd9b and Cxcl12a/Cxcr4b have a synergistic effect on the control of collective cell migration

PONE-D-21-17584R2

Dear Dr. Roehl,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Christoph Winkler, Dr.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

17 Nov 2021

PONE-D-21-17584R2

Tetraspanin Cd9b and Cxcl12a/Cxcr4b have a synergistic effect on the control of collective cell migration

Dear Dr. Roehl:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Christoph Winkler

Academic Editor

PLOS ONE