High levels of the type III inorganic phosphate transporter PiT1 (SLC20A1) can confer faster cell adhesion

The inorganic phosphate transporter PiT1 (SLC20A1) is ubiquitously expressed in mammalian cells. We recently showed that overexpression of human PiT1 was sufficient to increase proliferation of two strict density-inhibited cell lines, murine fibroblastic NIH3T3 and pre-osteoblastic MC3T3-E1 cells, and allowed the cultures to grow to higher cell densities. In addition, upon transformation NIH3T3 cells showed increased ability to form colonies in soft agar. The cellular regulation of PiT1 expression supports that cells utilize the PiT1 levels to control proliferation, with non-proliferating cells showing the lowest PiT1 mRNA levels. The mechanism behind the role of PiT1 in increased cell proliferation is not known. We, however, found that compared to control cells, cultures of NIH3T3 cells overexpressing PiT1 upon seeding showed increased cell number after 24 h and had shifted more cells from G0/G1 to SþG2/M within 12 h, suggesting that an early event may play a role. We here show that expression of human PiT1 in NIH3T3 cells led to faster cell adhesion; this effect was not cell type specific in that it was also observed when expressing human PiT1 in MC3T3-E1 cells. We also show for NIH3T3 that PiT1 overexpression led to faster cell spreading. The final total numbers of attached cells did, however, not differ between cultures of PiT1 overexpressing cells and control cells of neither cell type. We suggest that the PiT1-mediated fast adhesion potentials allow the cells to go faster out of G0/G1 and thereby contribute to their proliferative advantage within the first 24 h after seeding. & 2014 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/3.0/). lsevier Inc. This is an open access article under the CC BY license linical Medicine, Aarhus University, Department of Hematology, Aarhus University Hospital, , Aarhus University, C. F. Møllers Allé 3, building 1130, 8000 Aarhus C, Denmark.

PiT1 and PiT2, however, also have non-overlapping functions [24][25][26][27][28][29], and P i -transport-independent functions of PiT1 have been discovered [27,29,30]. Knockout of PiT1 in mice retards growth of embryos, slows proliferation of liver cells, and impairs erythroid and early B-cell development [24][25][26][27]. The effects of knockout of PiT1 in mice on erythroid and early B-cell development have been associated with defects in cell cycle progression [26,27]. Knockdown of PiT1 impairs proliferation of the transformed cell lines HeLa and HepG2 and tumorigenesis of HeLa cells in nude mice [29], and proliferation of the murine density-inhibited cell line MC3T3-E1 [23]. Thus, a certain level of PiT1 is important for cell proliferation. Moreover, overexpression of PiT1 is sufficient to increase proliferation of the murine density-inhibited cell lines NIH3T3 and MC3T3-E1 [23]. In agreement with that PiT1 overexpression leads to a general proliferative advantage, cultures of these cells grew to higher cell densities, but they remained density-inhibited [23]. However, when transformed, NIH3T3 cells overexpressing PiT1 formed more colonies in soft agar than control cells [23]. The cellular regulation of the endogenous PiT1 expression in NIH3T3 and MC3T3-E1 cells supports that cells utilize their PiT1 levels to control proliferation, with nonproliferating cells showing the lowest PiT1 mRNA levels [23].
There is direct and indirect evidence that the role of PiT1 in cell proliferation is not dependent on its P i transport function [23,24,29]. Thus although mouse embryonic fibroblasts from PiT1 knockout mice [24] and MC3T3-E1 cells with knocked down PiT1 expression [23] exhibited increased PiT2 expression, they still showed impaired proliferation. In addition, overexpression of PiT2 leading to increased P i uptake, did not rescue impaired proliferation of HeLa cells caused by reduced PiT1 expression, while expression of a PiT1 transport knockout mutant did [29]. Moreover, overexpression of PiT1 in NIH3T3 and MC3T3-E1 cells in general does not lead to regulation of PiT2, and while it does lead to increased proliferation of both cell types, only NIH3T3 cells show increased ability to import P i [23]. Together these results suggest that PiT1's role in regulation of cell proliferation is independent on its P i -uptake ability.
PiT1 overexpression was found to confer increased proliferation upon NIH3T3 and MC3T3-E1 cells under standard cultivation conditions, i.e., in standard growth media containing 10% bovine serum and tissue-culture treated plastic ware, all negatively charged (when wetted) tissue-culture treated polystyrene (TCPS) 3 [23]. When cells are seeded in serum-containing cell medium onto TCPS, the proteins of the serum will very quickly adsorb to the negative charges of the polystyrene. Depending on the quality of the TCPS (i.e., charge density and topology of the polystyrene) adsorption of more or less adhesionpromoting proteins will occur. The seeded cells adhere to the adhesion-promoting proteins, which predominantly are vitronectin (spreading factor) and plasma fibronectin, or via adhesion-promoting protein already attached to the cells. Specific contact with the adhesion-promoting proteins through integrins allows the cells to spread out [31][32][33][34][35][36] (reviewed in Elbert et al. [37] and Wilson et al. [38]). It is presently not known why cultures of cells with increased expression of PiT1 show enhanced proliferation. We have, however, previously observed that after just one day in culture, fibroblastic NIH3T3 cells overexpressing PiT1 had proliferated faster, and already 12 h after seeding, a lower percentage of the cells were in G0/G1 in the PiT1 overexpressing cultures compared to cultures of control cells [23]. We have here addressed how a high expression of PiT1 allows the cells to increase proliferation within the first day in culture. Using the same conditions as in the proliferation experiments, i.e., standard cultivation conditions, we found that overexpression of PiT1 in NIH3T3 cells leads to faster adhesion and spreading compared to control cells. The adhesion advantage of PiT1 overexpressing NIH3T3 cells compared to control cells was, moreover, found to be independent of serum concentrations during cultivation prior to seeding and during seeding, and of the quality of the TCPS, which the cells had been cultivated on prior to seeding. In addition, adhesion of MC3T3-E1 cells was also investigated under standard cultivation conditions, and overexpression of PiT1 was also found to confer faster adhesion on this pre-osteoblastic cell line.

Constructs
The vector pLXSN [39] was modified. The original HindIII site in the plasmid was removed by site-directed mutagenesis and HindIII and NotI sites introduced in the multiple cloning region; the modified pLXSN vector is referred to as pLXSN-ΔHindIIIþHindIIIþNotI and the retroviral vector as LXSNþHindIIIþNotI. The PiT1 encoding sequence was cloned as a HindIII-XhoI fragment from pOJ75 [16] into the HindIII -XhoI sites of pLXSN-ΔHindIIIþHindIIIþNotI resulting in the plasmid pLPiT1SN-ΔHindIIIþHindIIIþNotI; the retroviral vector is referred to as LPiT1SNþHindIIIþNotI.

Malachite green based assay for P i determination
To determine the contribution of P i from 10% newborn calf serum (NCS) (Gibco BRL) and 10% fetal bovine serum (FBS) (Gibco BRL) supplemented to Dulbecco's Modified Eagles Medium (DMEM) and Minimum Essential Medium alpha (α-MEM) (Gibco BRL), respectively, the total P i contents of P i -free DMEM containing 10% NCS or 10% FBS were determined using a malachite green based method basically as described by Baykov et al. [40] with the exception that 50 mL samples were added to 100 mL assay solutions in a 96-well plate and the developed malachite green-phosphomolybdate complex was measured at 650 nm.

Cell culture
The murine fibroblastic cell line NIH3T3 (ATCC CRL-1658) and derivatives of this were cultivated in DMEM (0.916 mM P i ) supplemented with 10% NCS and 1% penicillin and streptomycin (PS) (Gibco BRL) (DMEM-NCS-PS). The P i content of DMEM-NCS-PS was 0.92mM, which is the P i level NIH3T3 cells have been adapted to. NIH3T3 cells were grown according to the 3T3 scheme and even though NIH3T3 cells overexpressing PiT1 grow denser than control NIH3T3 cells, they maintain strict culture-density inhibited proliferation when cultivated in NCS. The murine preosteoblastic MC3T3-E1 cells [41] (a kind gift from Dr. H. Kodama) and derivatives of this were cultured in α-MEM (1.014 mM P i ) supplemented with 10% FBS and 1% PS (α-MEM-FBS-PS). The P i content of α-MEM-FBS-PS was 0.94mM, which is the P i concentration undifferentiated MC3T3-E1 cells were adapted to grow in. MC3T3-E1 cells and derivatives below the 5th passage were used for experiments.
Unless otherwise indicated, the cells were cultivated in their standard growth medium. Before seeding, an aliquot of the cells was mixed with Trypan Blue Solution (0.4%) (Gibco BRL) and live cells were counted in a counting chamber. No cell death was observed, and the seeded cell numbers always represent live cells.

Quantitative reverse transcription PCR (qRT-PCR) analysis
Cells seeded in 4-well plates (NUNC, Nunclon TM Δ surface) were lysed on the plates and RNA was purified using Ambions RNAqueouss-4PCR kit (Applied Biosystems) as described by the manufacturer. RNA was reversely transcribed to cDNA immediately after purification using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems). The qRT-PCRs were performed using the following TaqMans Gene Expression Assays (Applied Biosystems) as described by the manufacturer: human PiT1-FAM (Hs00965596_m1), and as an endogenous control, mouse β-2-microglobulin (β2M)-VIC (Mm00437 762_m1). For each sample, three technical replicates were made.
The individual qRT-PCR reactions contained: 10 mL TaqMans Universal Fast PCR Master Mix (Applied Biosystems), 1 mL Hs009 65596_m1, 1 mL Mm00437762_m1, and 8 mL cDNA (approximately 10 ng). The PCR cycles employed were: 95 1C for 10 min, 40 cycles of 95 1C for 1 s, and 60 1C for 20 s. The efficiencies of each set of primers were determined on dilution series of cDNA and were used in calculations of relative gene expression as described [42].

Evaluation of cell adhesion
Cells were cultivated using NUNC (Nunclon TM Δ surface) plastic wares (TCPS) and seeded in 6-well plates (NUNC, Nunclon TM Δ surface) in their standard growth medium. At different time points after seeding, the un-attached cells were removed by washing the cells in phosphate-buffered saline (PBS). The adhered cells were fixed in glutaraldehyde, and stained in 0.1% crystal violet (MERCK). The cells were photographed and 12 randomly picked fields per data-point were counted. Alternatively, NIH3T3-LPiT1SN and NIH3T3-LXSN cells were seeded in 4-well plates (NUNC, Nunclon TM Δ surface), un-attached cells removed by washing the cells in PBS, and adhered cells lysed in 0.5% Triton X-100 in ddH 2 O. As a measure of the number of adhered cells, the protein content of the wells were determined using BCA Protein Assay Reagent kit (Pierce) as described by the manufacturer. The average amount of protein per cell was determined on known numbers of pelleted cells; the cells were counted as described above. NIH3T3-LPiT1SN and NIH3T3-LXSN contained the same average amount of protein per cell, and the protein contents of the wells were used as a relative measure of the number of adhered cells.
The effects of serum concentration and TCPS quality on adhesion were analyzed as follows. NIH3T3-PiT1 and NIH3T3-control cells were cultivated at equal densities using cell culture flasks of low quality TCPS (Sarstedt Red Cap tissue culture flasks) or of high quality TCPS (NUNC, Nunclon TM Δ surface) in DMEM-PS supplemented with varying serum concentrations (10%, 5%, or 0.5% of NCS) for 48 h before seeding. The cells were seeded at 5,000 cells/ cm 2 in 4-well plates (NUNC, Nunclon TM Δ surface) in DMEM-PS containing serum concentrations corresponding to their treatment before seeding. Three hours after seeding, the protein contents of the wells were determined as described above. The average amount of protein per cell was determined as described above. NIH3T3-PiT1 and NIH3T3-control contained the same average amount of protein per cell, and the protein contents of the wells were used as a relative measure of the number of adhered cells.

Cell cycle analysis
Cell cycle analysis was performed as previously described [23]. Cells were seeded at 20,000 cells/cm 2 in T25-flasks in triplicates. They were detached using trypsin, pelleted by centrifugation, and re-suspended in PBS. The cells were then cooled on ice, mixed 1:1 with ice-cold 99.9% ethanol, incubated on ice, and stored at 4 1C until staining. In the staining procedure, the cells were pelleted by centrifugation, re-suspended in 1 mL PBS containing 20 mg/mL RNase A, and stained by addition of 100 µL propidium iodide (1 mg/mL). The cells were analyzed by flow cytometry (FL2 channel) on a FACS Calibur flow cytometer (Becton Dickinson). CellQuest software was used for acquisition, followed by doublet subtraction, and data analysis employing the Watson Pragmatic model using FlowJo software.

P i -transport assay
Cells were seeded in 4-well plates (NUNC, Nunclon TM Δ surface), and 32 P i -transport assay was performed as previously described [23] using 100 mM total [P i ] in the uptake experiments. The protein contents in the cell lysates were determined using BCA Protein Assay Reagent kit (Pierce) as described by the manufacturer.

F-actin staining
Cells were seeded at 5,000 cells/cm 2 in 8-well chamber slides (TCPS, Sarstedt). At different time points after seeding, the adhered cells were fixed in 4% paraformaldehyde, washed, permeabilized using 0.1% Triton-X100, and incubated in the dark with phalloidin conjugated with Alexa Fluor 594 (Life Technologies) (1U/well). The cells were mounted with fluorescence mounting medium (Dako), studied under a Zeizz Axiovert 200M microscope, and photographed at 20X and 40X magnification (Photometrics CoolSNAP HQ camera).

Statistical analysis
All data are presented as means7standard deviation (SD). The hypothesis that two values were identical was tested using a twotailed Student's t test. In Fig. 4B, factorial ANOVA was employed. A value of pr 0.05 was considered statistically significant.

Results and discussion
High PiT1 expression levels increase the adhesion rates of fibroblastic NIH3T3 and pre-osteoblastic MC3T3-E1 cells We have previously shown that NIH3T3 cells expressing human PiT1 (NIH3T3-LPiT1SN) proliferate faster than NIH3T3 cells harboring the empty transfer vector (NIH3T3-LXSN) under standard cultivation conditions [23]. Specifically, analyses of the progression through the cell cycle showed a statistically higher percentage of cells in SþG2/M in PiT1 overexpressing cultures than in control cell cultures after just 12 h in culture [23], and we therefore studied an early event in cell culture: adhesion. To determine whether the level of PiT1 expression affects cell adhesion, we studied the adhesion rate of NIH3T3-LPiT1SN and control cells under standard cultivation conditions. The cells were seeded at 5,000 cells/cm 2 in 6-well plates, fixed, and counted after significantly different from control cells at the same day, po0.05. C) Cells were seeded in 4-well plates at 20,000 cells/cm 2 . 2 h later, the cells were harvested for protein measurement. Each column represents cell lysates from 6 wells and duplicate protein measurements. The two cell populations harbored equal average amount of protein per cell; the protein content per well is used as a measure for the number of cells adhered in the well after 2 h. Data are means7SD. * indicates statistically significantly different from control cells, po0.05. 0.5, 1, 2, and 6 h in culture. After 1 h in culture, 78% more NIH3T3-LPiT1SN cells than control cells had adhered, and after 2 h in culture, 73% more NIH3T3-LPiT1SN cells than control cells had adhered ( Fig. 1A and B). After 6 h in culture, an equal number of NIH3T3-LPiT1SN cells and control cells had adhered ( Fig. 1A and B). Thus, an increased level of PiT1 expression was sufficient for the NIH3T3 cells to adhere faster than control cells; however, the PiT1 level did not affect the total number of adhered cells.
In the experiment shown in Fig. 1A and B, we seeded the cells at 5,000 cells/cm 2 . In the previous studies analyzing cell cycle progression and cell proliferation, the cells were seeded at a density of 20,000 cells/cm 2 [23]. To address whether the difference in seeding density influenced the adhesion characteristics, we seeded the cells at 20,000 cells/cm 2 in the experiment shown in Fig. 1C. We determined the protein content of the cells prior to seeding and of each well 2 h after seeding. The two seeded cell populations had the same average amount of protein per cell (data not shown) and the protein contents of the wells were used a relative measure of the number of adhered cells. Almost 92% more NIH3T3-LPiT1SN than control cells had adhered during these 2 h (Fig. 1C).
We also addressed whether the observed faster cell adhesion mediated by increased PiT1 expression in fibroblastic NIH3T3 cells is cell type specific. Overexpression of human PiT1 in preosteoblastic MC3T3-E1 cells also results in increased proliferative potential under standard cultivation conditions [23]. We therefore investigated adhesion of MC3T3-E1 cells transduced with the human PiT1 expressing vector LPiT1SN (MC3T3-E1-LPiT1SN) or the transfer vector LXSN (MC3T3-E1-LXSN) under standard cultivation conditions. The cells were seeded at 5,000 cells/cm 2 in 6well plates and fixed and counted after 0.5, 1, 2, 6, and 24 h in culture (Fig. 2). Already 0.5 h after seeding, statistically significantly more PiT1 overexpressing cells than control cells had adhered (250% more) (Fig. 2). After 1 h in culture, 240% more MC3T3-E1-LPiT1SN cells than control cells had adhered (Fig. 2). At 2 h after seeding, the total number of adhering cells did not differ between the two cultures. Thus, as we found for NIH3T3 cells (Fig. 1A and B), the total number of cells adhering did not differ between PiT1 overexpressing MC3T3-E1 and control cells. However, at 6 and 24 h, statistically significantly more MC3T3-E1-LPiT1SN cells than control cells were present in agreement with increased cell division in PiT1 overexpressing cultures. Thus, PiT1 overexpression in the pre-osteoblastic MC3T3-E1 cells also led to faster adhesion.
Thus overexpression of PiT1 leads not only to increased proliferation of the NIH3T3-LPiT1SN and MC3T3-E1-LPiT1SN cells compared to their respective control cells [23] but also to faster adhesion under the same standard cultivation conditions (Figs. 1  and 2).

Characterization of a second population of NIH3T3 cells overexpressing PiT1
To further validate that overexpression of PiT1 was sufficient to give NIH3T3 cells an adhesion advantage, we established other populations of human PiT1 expressing and control NIH3T3 cells. NIH3T3-PiT1 refer to NIH3T3 cells stably transduced with a slightly modified LXSN vector encoding human PiT1, and NIH3T3-control refer to NIH3T3 cells transduced with the empty transfer vector. The mRNA expression level of human PiT1 in NIH3T3-PiT1 cells is shown in Fig. 3A. Human PiT1 supports P i uptake in NIH3T3 cells, and the NIH3T3-PiT1 cells show increased P i uptake function compared to the NIH3T3-control cells (Fig. 3B), and at a similar level as NIH3T3-LPiT1SN cells (Fig. 3B) [23]. These results show that there is an increased expression of functional PiT1 protein at the cell surface of NIH3T3-PiT1 cells and at a level comparable to that of the NIH3T3-LPiT1SN cells. The NIH3T3-PiT1 cells also showed increased proliferation compared to NIH3T3-control cells (not shown). We performed cell cycle analysis to address whether the NIH3T3-PiT1 cells, as the NIH3T3-LPiT1SN cultures [23], showed different cell cycle profiles after just 12 h in culture. In agreement with previous results, overexpression of PiT1 led to statistically significantly lower percentage of cells in the G0/G1 phases and increased percentage of cells in SþG2/ M phases at 12 and 24 h after seeding (Fig. 3C). We also included a cell cycle analysis 6 h after seeding in the present study. Albeit the differences are small, already 6 h after seeding, the PiT1 overexpressing cultures and the control cultures showed statistically significant differences in their cell cycle profiles, with a higher percentage of PiT1 overexpressing cells in G0/G1 and a lower percentage in SþG2/M phases (Fig. 3C). We next investigated whether the Fig. 3 -Characterization of NIH3T3-PiT1 and NIH3T3-control cells. A) Analysis of the human PiT1 mRNA level in a second NIH3T3 cell population transduced with human PiT1 expressing vector. NIH3T3-PiT1 (human PiT1 (hPiT1) expressing) and NIH3T3-control (transduced with the transfer vector) cells were seeded at 20,000 cells/cm 2 in 4-well plates. After one day in culture, the human PiT1 mRNA level was determined using qRT-PCR. Results from cell lysates from three wells and triplicate qRT-PCR analyses of each cell lysates are shown. B) P i -transport assay of NIH3T3-PiT1 and NIH3T3-control cells. The cells were seeded at 20,000 cells/cm 2 in 4-well plates. The next day, 32 P i import in P i -free medium supplemented with 5 lM 32 P i and 95 lM P i was analyzed over 5 min.
Each column represents 32 P i import per mg protein per hour of four wells. Data are means7SD. * indicates statistically significantly different from control cells, po0.05. C) Cell cycle analysis of NIH3T3-PiT1 and NIH3T3-control cells. The cells were seeded at 20,000 cells/cm 2 in T25-flasks in triplicates. After 6, 12, and 24 h, the cells were fixed in ethanol. The cells were stained with propidium iodide and analyzed by flow cytometry. The percentages of cells in the respective phases of the cell cycle 6, 12, and 24 h after seeding obtained using the Watson Pragmatic model are shown. In each sample, 20,000 cells have been counted. Each column represents the mean of triplicate set upsþSD. * indicates statistically significantly different from control cells in the same phases of the cell cycle, po0.05.

Fast adhesion mediated by high PiT1 expression levels is independent of serum concentrations and growth on different TCPS qualities
Both cell division and adhesion can be affected by the concentration of different serum factors. When seeding cells in varying serum concentrations, the amount of vitronectin compared to fibronectin adsorbed onto the TCPS changes. However, even at as low as 0.1% serum, adhesion-promoting proteins from the serum are adsorbed to the TCPS prior to cell adhesion. Using human serum, the variation in adsorbed adhesion-promoting proteins from the serum has been seen to affect the speed of adhesion and spreading of baby hamster kidney (BHK) cells, with cells seeded in low serum concentrations (0.1-1%) adhering and spreading faster than cells seeded in higher serum concentrations (e.g., 10%) [32].
We therefore also addressed a possible influence of serum factors in the experiment addressing whether the NIH3T3-PiT1 and NIH3T3-control cells showed different adhesion potentials. The cells were grown in the presence of 5% or 0.5% NCS for 48 h before seeding in the same serum concentrations. The protein contents per well were determined 3 h after seeding, and independent on the serum concentration, statistically significantly more protein per well was present in cultures of PiT1 overexpressing cells compared to control cells (Fig. 4A). We also found that the serum concentration did not have an effect on the adhesion rates of control (compare two left most columns in Fig. 4A) or PiT1 overexpressing cells (compare two right most columns in Fig. 4A). Thus, the ability of PiT1 overexpressing cells to adhere faster than control cells was not affected by the nutritional state of the cells or by the concentration of external serum factors during seeding.
Differences in the modifications of the TCPS affect which and how serum proteins are adsorbed [37,38]. This could influence the cell-adhesion rate in that cells can adjust to growth on the TCPS surface they are cultivated on. We therefore tested whether the fast adhesion of NIH3T3-PiT1 cells was affected by the quality of the TCPS they had been cultivated on prior to seeding. The cells were, for 48 h before seeding, cultivated on two types of TCPS, Sarstedt Red Cap (low quality TCPS) and NUNC, Nunclon TM Δ surface (high quality TCPS), respectively. Both polystyrene surfaces are negatively charged when wetted, but flasks with the Nunclon TM Δ surface have an optimized and guaranteed uniform charge density allowing adsorption of more adhesion-promoting proteins from the serum than standard treated TCPS as Red Cap TCPS flasks from Sarstedt, which are designed for adhesion of robust and easily adhering cells. As seen in Fig. 4B, both control cells and PiT1 overexpressing cells adhered faster after growth on high quality TCPS when re-seeded on TCPS of high quality, than after growth on low quality TCPS followed by seeding on high quality TCPS. Thus, the abilities of the cells to adhere were influenced by the TCPS quality they had previously been cultivated on. PiT1 overexpression, however, still resulted in increased adhesion rate independent of the TCPS quality of the cultivation flasks.
Thus, the increased adhesion rate of PiT1 overexpressing cells was neither affected by the TCPS surface that the cells had been cultivated on nor by serum concentrations. Fig. 4 -Analyses of the effects of serum concentration and growth on high and low TCPS qualities on adhesion. NIH3T3-PiT1 and NIH3T3-control cells were grown for 48 h in 5% or 0.5% NCS and in culture flasks of different TCPS quality before seeding, as indicated. The cells were seeded in 4-well dishes (NUNC, Nunclon TM Δ surface) at 5,000 cells/cm 2 in serum concentrations corresponding to their growth media. The protein amounts per well were measured 3 h after seeding. Each column represents cell lysates from 8 wells and duplicate protein measurements. Data are means7SD. A) NIH3T3-PiT1 and NIH3T3-control cells grown, and seeded, in 0.5% and 5% NCS on NUNC, Nunclon TM Δ surface. * indicates statistically significant differences between NIH3T3-PiT1 and NIH3T3-control cells, po0.05. B) NIH3T3-PiT1 and NIH3T3-control cells grown in NUNC (Nunclon TM Δ surface) (high quality TCPS) or Sarstedt Red Cap (low quality TCPS) flasks and 0.5% NCS before seeding. * indicates independent statistically significant differences between NIH3T3-PiT1 and NIH3T3-control cells and between cultivation on the two different TCPS qualities, po0.05. Comparable results were obtained after growth in 5% NCS (data not shown).

NIH3T3 cells overexpressing PiT1 spread out faster after seeding compared to control cells
In order to initiate proliferation upon adhesion, anchoragedependent cells are dependent on spreading by interaction with extracellular adhesion-promoting proteins [43][44][45][46]. During cell adhesion and spreading, F-actin is rearranged, thus staining of the actin filaments can be used to follow the adhesion and spreading processes. To address whether the increased adhesion rate of PiT1 overexpressing cells compared to control cells leads to faster spreading of NIH3T3-PiT1 cells, we did an F-actin staining after 0.5, 1, 2, and 6 h in culture (Fig. 5). As seen in Fig. 5A, more NIH3T3-PiT1 cells had adhered after both 0.5, 1, and 2 h compared to NIH3T3-control cells in agreement with the results shown in Figs. 1 and 4. Furthermore, NIH3T3-PiT1 cells appear to spread out at earlier time points compared to NIH3T3-control cells (1 h and 2 h) (Fig. 5A). When studying the cells at 40X magnification (Fig. 5B), it was seen that NIH3T3-PiT1 cells had started spreading already after 30 min whereas NIH3T3-control cells only showed initial spreading at 1 h and only after 6 h had both NIH3T3-PiT1 and NIH3T3-control cells spread out. Thus overexpression of PiT1 not only led to faster adhesion but also to faster spreading of the anchorage-dependent NIH3T3 cells.
Our previous results showed that when cultivated under standard cultivation conditions, overexpression of PiT1 led to increased proliferation rates of MC3T3-E1 and NIH3T3 cells [23]. We have here shown that overexpression of PiT1, under standard cultivation conditions, also leads to faster adhesion of MC3T3-E1 and NIH3T3 cells and, at least for NIH3T3 cells, to faster spreading after seeding. Using NIH3T3 cells, we have moreover shown that the increased adhesion rate induced by overexpression of PiT1 was unaffected by changes in serum-factor concentrations and TCPS quality during cultivation.
In order to proliferate, anchorage-dependent cells are dependent on signals from growth-factors in G1 and signals from adhesion in G1 and during cytokinesis [44,47]. The observation that growth factors are only required until mid G1 phase, while adhesion also is required later in the G1-phase as well as during cytokinesis underscores adhesion as an important regulator of cell proliferation in anchorage-dependent cells [44][45][46][47]. The signals from adhesion necessary in G1 and during cytokinesis are dependent on interactions between integrins and extracellular adhesion-promoting proteins and, e.g., allow the cells to spread out [46,48,49]. Specifically, spreading in G1 is critical for progression to the S phase [43,46]. Faster adhesion/spreading could therefore explain at least initial increased proliferation of anchorage-dependent cells after seeding. We hypothesize that the observed faster adhesion, and for NIH3T3 also faster spreading, after seeding of cells overexpressing PiT1 compared to control cells are contributing to an initial faster cell cycle progression of the PiT1 overexpressing cells after seeding and thereby also to at least the initial increased proliferation after seeding.
As mentioned above and shown in Byskov et al. [23], the PiT1 overexpressing cells do not just show an initial faster cell cycle progression after seeding, they proliferate faster, and cultures of these grow to higher cell densities than cultures of control cells [23]. Cell adhesion, cell shape, and cell density have long been recognized as tightly coupled during cell division [50]. Studies in which the cell shape was controlled suggest that cell shape is involved in controlling density-inhibited proliferation, i.e., decreased cell flattening in dense cultures relates to decreased proliferation [43,51]. If the level of PiT1, which indeed is regulated by the cells themselves [23], in general, and not only after seeding, is involved in determining the shape of the cells, this could also explain the observed increase in culture density of PiT1 overexpressing cells.
Beck et al. reported that knockdown of PiT1 in HeLa cells resulted in delayed progression through mitosis due to an extended metaphase stage. They also reported an impaired ability to proceed through anaphase and telophase/cytokinesis resulting in many large multinucleated cells [29]. Although the transformed status of the HeLa cells does make a direct comparison difficult, it is however interesting that knockdown of PiT1 expression impairs the ability of the cells to proceed through specific stages of the cell cycle, which, for attached cells, are dependent on changes in the ability of the cells to interact with extracellular adhesionpromoting proteins, e.g., rounding up and migration [47][48][49][50]. Thus interestingly, the effects of overexpression of PiT1 on NIH3T3 and MC3T3-E1 cells and the effects of knockdown of PiT1 expression in HeLa cells are all in agreement with a role of PiT1 in regulation of processes involving interaction with an extracellular matrix.
We have previously shown that the pre-osteoblastic MC3T3-E1 cells, unlike NIH3T3 cells, regulate the overexpressed PiT1 in a manner, so it does not support P i import into the cell [23], while increased PiT1 expression does lead to faster adhesion and proliferation of both cell types. Since PiT1 overexpression does lead to increased adhesion rate and proliferation of MC3T3-E1 cells, this could suggest that the increased adhesion and proliferation rates of PiT1 overexpressing cells are independent of the P i transport ability of PiT1. The reduced proliferation of HeLa cells observed with knocked down PiT1 expression could be rescued by expression of a P i -transport-defect PiT1 protein [29]. These results strongly suggest that P i transport per se is not involved in the effects that changes in the PiT1 expression level have on cell proliferation. Although a P i -sensor function of mammalian PiT1 has so far not been directly shown, it has been reported for the yeast P i transporter Pho84 [52]. As the P i -transport-defect PiT1 used by Beck and co-workers [29] potentially could still convey a P i signal or be captured in a signaling conformation, it is thus possible that the role of PiT1 in increased adhesion/spreading and proliferation rates relates to a P i -sensor function of PiT1.

Conclusion
The present study provides insight into how PiT1 overexpressing cells can proliferate faster than control cells. Thus we have here shown that elevated expression level of PiT1 can confer faster adhesion on MC3T3-E1 and NIH3T3 cells, and for NIH3T3 cells also faster spreading, under conditions which support increased cell proliferation and increased cell-density. Using NIH3T3 cells, we showed that the increased adhesion rate induced by overexpression of PiT1 was not affected by changes in serum concentrations or by the TCPS quality during cultivation. We suggest that the here described faster adhesion and cell spreading rates are contributing to, at least the initial increase in proliferation after seeding of cells overexpressing PiT1 compared to control cells. It is notable that a role of PiT1 in processes regulating cell adhesion and spreading is in agreement with the observed later proliferative advantage of PiT1 overexpressing cells and increased cell density of the cultures as well as with the by others reported delayed progression through mitosis and impaired cytokinesis induced by knocking down PiT1 in HeLa cells growing attached.