Garneau AP et al. (2015), Aquaporins Mediate Silicon Transport in Humans.

XB-ART-51180

PLoS One 2015 Aug 13;108:e0136149. doi: 10.1371/journal.pone.0136149.

Show Gene links Show Anatomy links

Aquaporins Mediate Silicon Transport in Humans.

Garneau AP , Carpentier GA , Marcoux AA , Frenette-Cotton R , Simard CF , Rémus-Borel W , Caron L , Jacob-Wagner M , Noël M , Powell JJ , Bélanger R , Côté F , Isenring P .

???displayArticle.abstract???
In animals, silicon is an abundant and differentially distributed trace element that is believed to play important biological functions. One would thus expect silicon concentrations in body fluids to be regulated by silicon transporters at the surface of many cell types. Curiously, however, and even though they exist in plants and algae, no such transporters have been identified to date in vertebrates. Here, we show for the first time that the human aquaglyceroporins, i.e., AQP3, AQP7, AQP9 and AQP10 can act as silicon transporters in both Xenopus laevis oocytes and HEK-293 cells. In particular, heterologously expressed AQP7, AQP9 and AQP10 are all able to induce robust, saturable, phloretin-sensitive silicon transport activity in the range that was observed for low silicon rice 1 (lsi1), a silicon transporter in plant. Furthermore, we show that the aquaglyceroporins appear as relevant silicon permeation pathways in both mice and humans based on 1) the kinetics of substrate transport, 2) their presence in tissues where silicon is presumed to play key roles and 3) their transcriptional responses to changes in dietary silicon. Taken together, our data provide new evidence that silicon is a potentially important biological element in animals and that its body distribution is regulated. They should open up original areas of investigations aimed at deciphering the true physiological role of silicon in vertebrates.

???displayArticle.pubmedLink??? 26313002
???displayArticle.pmcLink??? PMC4551902
???displayArticle.link??? PLoS One
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: aqp1 aqp10 aqp11 aqp2 aqp3 aqp7 aqp8 aqp9 gapdh myc nhs

???attribute.lit??? ???displayArticles.show???

	Fig 2. Control Si transport studies in AQP-expressing Xenopus laevis oocytes.(A) 68Ge influx. Oocytes incubated for 90 min in B1 medium (see Table 2) + 2 mM H4GeO4 with 1 μCi/mL H468GeO4 in or + 2 mM H4SiO4 were assayed for 68Ge content. Data correspond to background-subtracted influx values and are presented as means ± S.E. of 12 oocytes among 3 experiments. They were significantly different relative to background, between channels (for a given substrate) and between substrates (for lsi1). The image inserted on the top is to show the chemical structure of H4GeO4. (B) Si influx in AQP7G264V-expressing oocytes. Channels were tagged at the N-terminus with the epitope c-Myc. Conditions were as described for Fig 1A except that 2 mM sulfo-NHS was added to B1 medium in certain studies. Data are presented as means ± S.E. of 3 measurements among 6–7 experiments using * to indicate that they are significantly different compared to the controls. The image inserted on the right shows that the abundance and migration pattern of cell surface AQP7G264V by Western blot analysis are similar to those of wild type AQP7. (C) Dependence of Si influx on external osmolality (or on net water transport). Oocytes incubated for 90 min in a modified B1 medium (B2a, B2b or B2c) were assayed for Si content. Data are presented as means ± S.E. of 3 measurements among 6 experiments. Influx values among conditions are not significantly different. Under isotonic conditions, cell volumes after 20 s were also similar (885 ± 72 vs. 791 ± 97 nL). Abbreviation: WT, wild type.
	Fig 3. Characteristics of Si transport by AQP-expressing Xenopus laevis oocytes.(A) Dependence of Si influx on [H4SiO4]. Oocytes incubated for 30 min (AQP9), 60 min (AQP10) or 90 min (AQP7 and lsi1) in B1 medium (see Table 2) + 0 to 2 mM H4SiO4 were assayed for Si content. Data are expressed as background-subtracted influx values and are presented as means ± S.E. of 3 measurements among 3–5 experiments. (B) Dependence of Si influx on intracellular pH and on extracellular pH, [Na+], [K+], and [Cl−]. Oocytes incubated for 90 min in a modified B1 medium (called medium B3a, B3b, B3c, B3d or B3e) + 2 mM H4SiO4 were assayed for Si content. Data are expressed as n-fold differences and presented as means ± S.E. of 3 measurements among 6 experiments. None of the data are significantly different from 1. (C) Effect of phloretin on Si influx and water transport. Left scale: Oocytes incubated for 30 min in B2a medium + 2 mM H4SiO4 ± 0.1 mM phloretin were assayed for Si content. Data are expressed as background-subtracted influx values and presented as means ± S.E. of 3 measurements among 6 experiments. Right scale: Oocytes were assayed for cell volume during a 20-s incubation in ~10 mM sucrose ± 0.1 mM phloretin. Data are expressed as n-fold increases in cell volumes (V) relative to initial cell volumes (V0) and presented as means ± S.E. of 5 oocytes among 4 experiments. * indicates that the values are significantly different between − P and + P. Abbreviations: i, intracellular; o, extracellular; + P, phloretin.
	Fig 4. AQP expression and Si transport in HEK-293 cells.(A) Effect of anti-AQP3, AQP7, AQP9 and AQP10 siRNAs on AQPG expression and Si efflux. HEK-293 cells incubated for ~48 h in R medium (see Table 2) + 2 mM H4SiO4 ± 5 different siRNAs and for 5 additional min in R medium alone were assayed for AQP expression by qPCR and for Si content. Data are expressed as % changes between conditions “scrambled siRNAs” and “anti-AQP siRNAs” and are presented as means ± S.E. of 2–8 measurements among 3 experiments. All of the changes are significantly different from 0. Note that threshold cycles in the absence of siRNAs were of ~30 for all the AQGPs. (B) Effect of anti-AQP7 siRNAs on AQPG expression. Protocols used and data expression are as described for panel A. Except for AQP10, all of the changes are significantly different from 0. (C) Total Si and 68Ge influx. AQGP-transfected HEK-293 cells incubated for 5 min in R medium + 2 mM H4GeO4 with 1 μCi/mL H468GeO4 or in R medium + 2 mM H4SiO4 were assayed for both 68Ge or Si content, respectively. Data are expressed as mean influx values ± S.E. of 4–8 measurements among 3–5 experiments. Compared to the controls, all of the values are significantly different. (D) Background-subtracted 68Ge and Si influx. Influx values measured in pCDNA-transfected HEK-293 cells were subtracted from the influx values measured in AQGP-transfected HEK-293 cells. All values are significantly different from 0.
	Fig 5. Effect of high- and low-Si diets on AQP transcription in selected mouse tissues.Animals were subjected to a Si-rich diet (4% elementary Si; n = 9) or a Si-poor diet (0.4% elementary Si; n = 8 or 9) for 3 weeks and sacrificed afterwards to carry out qPCR studies using total RNA from kidney (panel A), small intestine (panel B) or calvarium (panel C) as templates. Oligonucleotides used are shown in Table 1. Data are expressed as AQP-specific DNA copy numbers normalized to GADPH-specific DNA copy numbers and presented as means ± S.E. of 8–9 mice between two experiments using * to indicate that they are significantly different compared to each other. Of notice, relative AQP1 expression levels were lower than expected based on the data of Table 3 so that conditions used to amplify this isoform might have been suboptimal. Under the Si-rich diet, in addition, blood and urine [Si] were over 10-fold higher compared to the Si-poor diet.
	Fig 1. Standard Si transport studies in Xenopus laevis oocytes.(A) Si influx. Oocytes incubated for 30 min (AQP9), 60 min (AQP10) or 90 min (lsi1 and other AQPs) in B1 medium (see Table 2) + 2 mM H4SiO4 were assayed for Si content. Data are presented as means ± S.E. of 3 measurements among 4 experiments using * to indicate that they are significantly different compared to the controls. The image inserted on the top is to show the chemical structure of H4SiO4. (B) Si uptake and loss vs. time. Oocytes incubated for 0 to 36 h in B1 medium (± 2 mM H4SiO4 from 0 to 24 h and no H4SiO4 from 24 to 36 h) were assayed for Si content. Data are presented as means ± S.E. of 3 measurements among 5 or more experiments. Compared to the controls, data for AQP7 and AQP9 are all significantly different beyond the zero point, and data for AQP3 are all significantly different from 6 to 36 h. (C) Water transport. Oocytes were assayed for cell volume measurements during a 20-s incubation step in distilled water added with ~10 mM sucrose. Data are expressed as n-fold increases in cell volumes (V) relative to initial cell volumes (V0) and are presented as means ± S.E. of 5 oocytes among 3 experiments. They are all significantly different compared to the controls at 20 s.

References [+] :

Aharon, Involvement of aquaporin 9 in osteoclast differentiation. 2006, Pubmed