Upon T cell stimulation, NFAT is dephosphorylated by calcineurin, leading to nuclear translocation via NFAT–importin β interaction. Whereas the process of NFAT dephosphorylation has been well researched, the molecular mechanism of NFAT–importin β interaction remains unknown. In contrast to NF-κB and STAT, no importin α family members have been reported as adaptor proteins for NFAT. Our study shows that tubulin α, but not tubulin β, binds to the N-terminal region of NFAT containing the regulatory and Rel homology domains. Importin β interacts with the NFAT–tubulin α complex rather than NFAT or tubulin α alone, resulting in cotranslocation of NFAT and tubulin α into the nucleus. Furthermore, the interaction is suppressed by acetate-induced tubulin α acetylation at lysine 40. In conclusion, tubulin α functions as an adaptor in NFAT–importin β interaction, and this function is regulated by acetate-induced acetylation.

The NFAT family of transcription factors comprises five members (1). Of these, NFAT1 (NFATc2), NFAT2 (NFATc1), and NFAT4 (NFATc3) are expressed in T cells and critically involved in the production of cytokines such as IL-2 (2). Therefore, NFAT is being targeted for modulating immune responses. NFAT resides in the cytoplasm of resting T cells, and it is dephosphorylated by calcineurin following TCR engagement or PMA plus ionomycin stimulation, which bypasses TCR engagement (1, 2). Subsequently, it interacts with importin β, a nuclear transport factor, resulting in nuclear translocation (1, 3). The mechanism by which NFAT interacts with importin β is still unknown. Understanding the molecular mechanism of NFAT–importin β interaction may provide a target for modulating NFAT nuclear translocation.

Clinical immunomodulators such as cyclosporin A (CsA) impair NFAT dephosphorylation by inhibiting calcineurin activity, leading to suppression of NFAT nuclear translocation (2). In contrast, in T cell leukemia-derived Jurkat cells and primary T cells, acetate suppresses NFAT nuclear translocation by inhibiting NFAT–importin β interaction without affecting NFAT dephosphorylation (3). We hypothesized that acetate could exert the inhibitory effect by inducing acetylation of a protein that regulates NFAT–importin β interaction, similar to the manner in which butyrate affects gene expression by inducing acetylation of histones that remodel chromatin structure and regulate gene expression (4). Acetate is the principal short fatty acid produced by bacterial fermentation in the colon (5), and its concentration is estimated to be 30–75 mM in human feces (4). It is also a major product of some Lactobacillus species and Bifidobacterium, both of which are used as probiotics in the treatment of inflammatory bowel diseases (6).

The present study identifies an endogenous protein that is acetylated following acetate treatment, and then assesses the relationship between acetylation of the identified protein and inhibition of NFAT nuclear translocation. We elucidate the critical role of the identified protein for NFAT–importin β interaction.

We purchased sodium acetate, butyrate, and trichostatin A (TSA) from Wako Pure Chemical (Osaka, Japan), and PMA and ionomycin from Sigma-Aldrich (St. Louis, MO).

We purchased anti-acetylated lysine Ab from StressMarq (Victoria, Canada); anti-tubulin α, anti-tubulin β, and anti-retinoblastoma Abs from Thermo Scientific (Fremont, CA); anti-tubulin α (acetyl K40), anti-importin β, and anti-Flag Abs from Sigma-Aldrich; anti-histone H3 (acetyl K9) Ab from GeneTex (Irvine, CA); anti-NFAT1 Ab from BD Biosciences (San Jose, CA); anti-p65 Ab from Cell Signaling Technology (Danvers, MA); anti-importin α1 Ab from Bethyl Laboratories (Montgomery, TX); anti-importin α3 and anti-importin α5 Abs from Abnova (Walnut, CA); anti-importin α6 Ab from Protein Tech Group (Chicago, IL); peroxidase-conjugated anti-rabbit IgG L chain Ab from Jackson ImmunoResearch Laboratories (West Grove, PA); peroxidase-conjugated anti-mouse IgG Ab from DakoCytomation (Glostrup, Denmark); and TrueBlot anti-mouse Ig HRP, which detects native Ab, from eBioscience (San Diego, CA).

Recombinant human histone deacetylace 6 (HDAC6) and the components for the HDAC6 activity assay were purchased from Enzo Life Sciences (Plymouth Meeting, PA). HDAC6 activity was determined in the presence of sodium acetate, according to the manufacturer's instructions.

Jurkat cells were incubated at 37°C with 5 mM sodium acetate, 10 nM TSA, or 5 μM CsA for 0.5 or 2 h, and then stimulated with 20 ng/ml PMA and 0.5 μM ionomycin for an additional 0.5 h. Nuclear extract was obtained for Western blot analysis, as described previously (3).

The NFAT-dependent Photinus luciferase reporter construct contains three copies of the NFAT binding site, and the NF-κB–dependent reporter construct contains eight copies of the NF-κB binding site (3). Jurkat cells (5 × 106) were electroporated (250 V, 950 μF) in 250 μl culture media with the reporter construct (8 μg NFAT dependent or 3 μg NF-κB dependent) and 10 ng phRL-TK (Promega, Madison, WI) expressing Renilla luciferase. They were then resuspended in 2.5 ml culture media. Twenty-four hours later, the cells were incubated with 5 mM sodium acetate or 10 nM TSA for 0.5 or 2 h, stimulated with 20 ng/ml PMA and 0.5 μM ionomycin for an additional 4 h, and then lysed to determine their luciferase activities using a dual-luciferase reporter assay system (Promega). Reporter activity was assessed by normalization of Photinus luciferase activity to Renilla luciferase activity.

GST (GenScript, Piscataway, NJ) or GST–human NFAT2 (SignalChem, British Columbia, Canada), 1 μg, was mixed with 6× His–human tubulin α (Protein Tech Group), 1 μg; 6× His–human importin β (Sigma-Aldrich), 4 μg; or both in 500 μl PBS containing 0.1% Triton X-100, and then incubated with glutathione-agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA) at 4°C for 4 h. After washing the beads, collected proteins were eluted in SDS buffer.

The 6× His–importin β, 4 μg, and 6× His–tubulin α, 0.5 μg, were incubated with GST or GST-NFAT2, 1 μg, in 500 μl PBS containing 0.1% Triton X-100 at 4°C overnight. The samples were mixed with 2 μg anti-tubulin α Ab and incubated for an additional 1.5 h. The immune complex was then collected for 0.5 h with protein G–Sepharose beads (GE Healthcare, Piscataway, NJ).

Jurkat cells (3 × 106) were lysed in 500 μl TBS containing 1% Triton X-100. The lysates were incubated with anti-NFAT1 Ab or mouse IgG1 (Imgenex, San Diego, CA), 2 μg, at 4°C for 3 h. The immune complex was collected with protein G–Sepharose beads.

Mouse NFAT1 cDNA was obtained from Open Biosystems (Huntsville, AL) and subcloned into the pCMV5-Flag vector. HEK-293 cells (American Type Culture Collection; 2 × 105) were transfected in 24-well plates with 2 μg Flag-NFAT1 constructs using 1.5 μl TransFectin (Bio-Rad, Hercules, CA). Following 20 h of incubation, the cells were lysed in 500 μl TBS containing 1% Triton X-100 for immunoprecipitation assay with 3 μg anti-Flag Ab.

Control small interfering RNA (siRNA) and two different tubulin α siRNA (No. 1, catalog sc-29188; No. 2, catalog sc-44242) were purchased from Santa Cruz Biotechnology. Jurkat cells were electroporated with 50 pmol siRNA. Forty hours later, the cells were lysed for Western blot analysis. Alternatively, the cells were stimulated with PMA plus ionomycin for nuclear translocation assay.

Data are presented as means ± SD. We used Student's t test and considered a p value <0.05 to be statistically significant.

To investigate the induction of acetylation of endogenous proteins, Jurkat cells were incubated with sodium acetate for 0.5 h and then lysed for Western blot analysis using anti-acetylated lysine Ab. Acetate induced the acetylation of a protein of the same molecular size as tubulin α (Fig. 1). One of the posttranslational modifications of tubulin α is acetylation at lysine 40 (7). Using the Ab for tubulin α acetylated at lysine 40, we confirmed that tubulin α acetylation was induced by acetate in a concentration-dependent manner (Fig. 1). We also observed acetate-induced tubulin α acetylation in primary CD4+ T cells (Supplemental Fig. 1A). Moreover, the acetate treatment suppressed IL-2 production following the stimulation with anti-CD3 and anti-CD28 Abs (Supplemental Fig. 1B), suggesting the association between acetylation of tubulin α and suppression of IL-2 production. In contrast to acetate, butyrate did not induce tubulin α acetylation (Fig. 1), as reported previously (8), thus indicating the specific effect of acetate on tubulin α acetylation.

FIGURE 1.

Tubulin α acetylation induced by acetate. Jurkat cells were incubated at 37°C with sodium acetate, sodium chloride, or sodium butyrate for 0.5 h, and then lysed for Western blot analysis. Reproducible results were obtained from three independent experiments.

FIGURE 1.

Tubulin α acetylation induced by acetate. Jurkat cells were incubated at 37°C with sodium acetate, sodium chloride, or sodium butyrate for 0.5 h, and then lysed for Western blot analysis. Reproducible results were obtained from three independent experiments.

Close modal

Although the acetylation enzyme has not been identified, HDAC6 is involved in the deacetylation of tubulin α (7, 8). TSA, an HDAC6 inhibitor, induced tubulin α acetylation in Jurkat cells (Fig. 2A). Acetate inhibited HDAC6 activity in a concentration-dependent manner (Fig. 2B). These findings indicate that acetate induces tubulin α acetylation, at least partially, through enhanced accumulation of acetylated tubulin α by inhibiting HDAC6 activity.

FIGURE 2.

Tubulin α acetylation induced by acetate or TSA. A, Jurkat cells were incubated with sodium acetate or TSA for 0.5 h and then lysed for Western blot analysis. B, HDAC6 activity was determined in the presence of sodium acetate (n = 3).

FIGURE 2.

Tubulin α acetylation induced by acetate or TSA. A, Jurkat cells were incubated with sodium acetate or TSA for 0.5 h and then lysed for Western blot analysis. B, HDAC6 activity was determined in the presence of sodium acetate (n = 3).

Close modal

After 0.5 h of incubation, 5 mM acetate and 10 nM TSA treatments induced tubulin α acetylation to a similar extent (Figs. 2A, 3A). However, after 2 h of incubation, the TSA treatment induced far less tubulin α acetylation than the acetate treatment (Fig. 3A). To determine the relationship between acetylation of tubulin α and inhibition of NFAT nuclear translocation, Jurkat cells were stimulated with PMA plus ionomycin after 0.5 or 2 h of incubation with acetate or TSA. After 0.5 h of incubation, both the acetate and TSA treatments inhibited NFAT nuclear translocation (Fig. 3B) without affecting NFAT dephosphorylation (Supplemental Fig. 2). Neither the acetate nor TSA treatment impaired nuclear translocation of p65, a subunit of NF-κB (Fig. 3B). These findings are consistent with our previous report (3). After 2 h of incubation, acetate treatment continued to inhibit NFAT nuclear translocation, but TSA treatment did not (Fig. 3C). NFAT-dependent reporter assay also showed that in contrast to the effect with acetate treatment, NFAT activation was inhibited after 0.5 h, but not 2 h of incubation with TSA (Fig. 3D, 3E). Neither the acetate nor TSA treatment impaired NF-κB–dependent reporter activation (Fig. 3F). These data indicate that acetylation of tubulin α leads to inhibition of NFAT nuclear translocation.

FIGURE 3.

Relationship between acetylation of tubulin α and inhibition of NFAT nuclear translocation. A, Jurkat cells were incubated with 5 mM sodium acetate or 10 nM TSA for 0.5–2 h and then lysed for Western blot analysis. B and C, Jurkat cells were incubated with 5 mM sodium acetate or 10 nM TSA for 0.5 h (B) or 2 h (C), and then stimulated with PMA plus ionomycin for nuclear translocation assay. The intensity of the NFAT1 bands was normalized to the intensity of the retinoblastoma bands. *p < 0.05. D and E, Jurkat cells were incubated with 5 mM sodium acetate or 10 nM TSA for 0.5 h (D) or 2 h (E) and then stimulated with PMA plus ionomycin for NFAT-dependent reporter assay. F, Jurkat cells were incubated with 5 mM sodium acetate or 10 nM TSA for 0.5 h and then stimulated with PMA plus ionomycin for NF-κB–dependent reporter assay.

FIGURE 3.

Relationship between acetylation of tubulin α and inhibition of NFAT nuclear translocation. A, Jurkat cells were incubated with 5 mM sodium acetate or 10 nM TSA for 0.5–2 h and then lysed for Western blot analysis. B and C, Jurkat cells were incubated with 5 mM sodium acetate or 10 nM TSA for 0.5 h (B) or 2 h (C), and then stimulated with PMA plus ionomycin for nuclear translocation assay. The intensity of the NFAT1 bands was normalized to the intensity of the retinoblastoma bands. *p < 0.05. D and E, Jurkat cells were incubated with 5 mM sodium acetate or 10 nM TSA for 0.5 h (D) or 2 h (E) and then stimulated with PMA plus ionomycin for NFAT-dependent reporter assay. F, Jurkat cells were incubated with 5 mM sodium acetate or 10 nM TSA for 0.5 h and then stimulated with PMA plus ionomycin for NF-κB–dependent reporter assay.

Close modal

Using recombinant proteins, we performed pull-down assay to investigate interaction among NFAT, tubulin α, and importin β. Recombinant tubulin α bound to NFAT both in the absence and presence of importin β (Fig. 4A, second row), whereas importin β bound to NFAT in the presence of tubulin α rather than in the absence (Fig. 4A, fourth row). To examine the binding between tubulin α and importin β, we also performed immunoprecipitation assay and found that importin β bound to tubulin α in the presence of NFAT (Fig. 4B). These findings indicate that importin β binds to the NFAT–tubulin α complex rather than to NFAT or tubulin α alone.

FIGURE 4.

Interaction among NFAT, tubulin α, and importin β. A, GST or GST–NFAT2 was mixed with 6× His–tubulin α, 6× His–importin β, or both (Input), and collected with glutathione–agarose beads (Pull). Coprecipitated tubulin α and importin β were detected by Western blot analysis. B, 6× His–importin β and 6× His–tubulin α were incubated in the presence of GST or GST–NFAT2. The samples were mixed with anti-tubulin α Ab, and the immune complex was collected with protein G–Sepharose beads (IP) for Western blot analysis. C, Lysates of Jurkat cells were incubated with IgG or anti-NFAT1 Ab, and the immune complex was collected. D, Jurkat cells were incubated with 5 mM acetate for 0.5 h, stimulated with PMA plus ionomycin for an additional 0.5 h, and lysed for immunoprecipitation assay with anti-NFAT1 Ab. E, HEK-293 cells were transfected with constructs of Flag–NFAT1 fragments. Twenty hours later, the cells were lysed for immunoprecipitation assay with anti-Flag Ab. *Nonspecific bands. F, Jurkat cells were incubated with 5 μM CsA or 5 mM acetate for 0.5 h, and stimulated with PMA plus ionomycin for an additional 0.5 h. Cytosolic and nuclear extracts were obtained for Western blot analysis. G, Jurkat cells were electroporated with control or tubulin α siRNA. Forty hours later, the cells were lysed for Western blot analysis. Alternatively, the cells were stimulated with PMA plus ionomycin for nuclear translocation assay.

FIGURE 4.

Interaction among NFAT, tubulin α, and importin β. A, GST or GST–NFAT2 was mixed with 6× His–tubulin α, 6× His–importin β, or both (Input), and collected with glutathione–agarose beads (Pull). Coprecipitated tubulin α and importin β were detected by Western blot analysis. B, 6× His–importin β and 6× His–tubulin α were incubated in the presence of GST or GST–NFAT2. The samples were mixed with anti-tubulin α Ab, and the immune complex was collected with protein G–Sepharose beads (IP) for Western blot analysis. C, Lysates of Jurkat cells were incubated with IgG or anti-NFAT1 Ab, and the immune complex was collected. D, Jurkat cells were incubated with 5 mM acetate for 0.5 h, stimulated with PMA plus ionomycin for an additional 0.5 h, and lysed for immunoprecipitation assay with anti-NFAT1 Ab. E, HEK-293 cells were transfected with constructs of Flag–NFAT1 fragments. Twenty hours later, the cells were lysed for immunoprecipitation assay with anti-Flag Ab. *Nonspecific bands. F, Jurkat cells were incubated with 5 μM CsA or 5 mM acetate for 0.5 h, and stimulated with PMA plus ionomycin for an additional 0.5 h. Cytosolic and nuclear extracts were obtained for Western blot analysis. G, Jurkat cells were electroporated with control or tubulin α siRNA. Forty hours later, the cells were lysed for Western blot analysis. Alternatively, the cells were stimulated with PMA plus ionomycin for nuclear translocation assay.

Close modal

Using lysates of Jurkat cells, we performed immunoprecipitation assay to further investigate interaction among endogenous NFAT, tubulin α, and importin β. The results of the assay indicated that tubulin α, but not tubulin β, bound to NFAT (Fig. 4C). Neither acetate treatment nor PMA plus ionomycin stimulation affected binding between NFAT and tubulin α (Fig. 4D, first row). The results also showed that importin β bound to NFAT following PMA plus ionomycin stimulation and that binding was impaired by acetate treatment (Fig. 4D, fourth row). We confirmed that acetate treatment induced acetylation of tubulin α binding to NFAT (Fig. 4D, second row). These findings indicate that the NFAT–tubulin α complex interacts with importin β following NFAT dephosphorylation and that tubulin α acetylation impairs the interaction without affecting NFAT–tubulin α complex formation. NFAT dephosphorylation exposes the nuclear localization signal in the regulatory domain (2), and nuclear localization signals usually require adaptor proteins to interact with importin β (9). In contrast to NF-κB and STAT (1012), no importin α family members have been reported as adaptor proteins for NFAT. We did not observe binding of importin α1, α3, α5, or α6 to NFAT, even after PMA plus ionomycin stimulation (Supplemental Fig. 3).

Using Flag–NFAT fragments, we determined the NFAT region responsible for binding to tubulin α. Endogenous tubulin α was coprecipitated with full and N-terminal 1–571 fragments much more than with another N-terminal 1–393 fragment (Fig. 4E), indicating the critical involvement of the 394–571 region in the binding of NFAT to tubulin α. However, tubulin α was not coprecipitated with a 394–571 fragment (Fig. 4E). These findings reveal that the N-terminal region of NFAT containing the regulatory and Rel homology domains is responsible for binding to tubulin α.

Binding between tubulin α and NFAT suggested that tubulin α could be translocated into the nucleus together with NFAT. PMA plus ionomycin stimulation increased the amount of tubulin α, but not tubulin β, in the nuclear extract (Fig. 4F), indicating tubulin α nuclear translocation. The increased amount of tubulin α was suppressed with CsA and acetate treatments (Fig. 4F), both of which impair NFAT nuclear translocation. We also observed that knockdown of tubulin α suppressed NFAT nuclear translocation (Fig. 4G) and inhibited NFAT-dependent reporter activation (Supplemental Fig. 4). These findings show cotranslocation of tubulin α and NFAT into the nucleus.

Although a previous study has shown that the disruption of tubulin polymerization affects NFAT nuclear translocation in neural cells (13), the mode of the action remains unknown. Our present study identified the role of tubulin α as an adaptor in NFAT–importin β interaction. In addition to a novel role for tubulin α, we also elucidated a molecular mechanism for regulating T cell activation with acetate. Considering the abundance of acetate in human feces (4, 5), commensal bacteria might maintain low-level inflammation in the colon by producing acetate. In contrast, the reduced amount of acetate in the feces of patients with inflammatory bowel diseases (14, 15) might lead to immune hypersensitivity of the colon. Such pathological conditions can be improved by supplying acetate in the colon, because previous clinical trials have demonstrated the efficacy of enemas containing acetate against ulcerative colitis (16, 17). Our previous study also showed an anti-inflammatory effect of acetate in mouse models of colitis and dermatitis (3). We propose tubulin α acetylation not only as a target for modulating NFAT nuclear translocation, but also as a biomarker for determining the influence of acetate on T cell activation.

We thank Dr. Ramnik Xavier for critical comments on the manuscript.

This work was supported by grants from the Japan Society for the Promotion of Science and the Mishima Kaiun Memorial Foundation (to K.I.). The Department of Molecular Biology and Pathogenesis of Gastroenterology was endowed by Schering-Plough.

The online version of this article contains supplemental material.

Abbreviations used in this article:

CsA

cyclosporin A

HDAC6

histone deacetylase 6

siRNA

small interfering RNA

TSA

trichostatin A.

1
Macian
F.
2005
.
NFAT proteins: key regulators of T-cell development and function.
Nat. Rev. Immunol.
5
:
472
484
.
2
Kiani
A.
,
Rao
A.
,
Aramburu
J.
.
2000
.
Manipulating immune responses with immunosuppressive agents that target NFAT.
Immunity
12
:
359
372
.
3
Ishiguro
K.
,
Ando
T.
,
Maeda
O.
,
Ohmiya
N.
,
Niwa
Y.
,
Goto
H.
.
2007
.
Acetate inhibits NFAT activation in T cells via importin beta1 interference.
Eur. J. Immunol.
37
:
2309
2316
.
4
Hamer
H. M.
,
Jonkers
D.
,
Venema
K.
,
Vanhoutvin
S.
,
Troost
F. J.
,
Brummer
R. J.
.
2008
.
Review article: the role of butyrate on colonic function.
Aliment. Pharmacol. Ther.
27
:
104
119
.
5
Wong
J. M.
,
de Souza
R.
,
Kendall
C. W.
,
Emam
A.
,
Jenkins
D. J.
.
2006
.
Colonic health: fermentation and short chain fatty acids.
J. Clin. Gastroenterol.
40
:
235
243
.
6
Preidis
G. A.
,
Versalovic
J.
.
2009
.
Targeting the human microbiome with antibiotics, probiotics, and prebiotics: gastroenterology enters the metagenomics era.
Gastroenterology
136
:
2015
2031
.
7
Hammond
J. W.
,
Cai
D.
,
Verhey
K. J.
.
2008
.
Tubulin modifications and their cellular functions.
Curr. Opin. Cell Biol.
20
:
71
76
.
8
Blagosklonny
M. V.
,
Robey
R.
,
Sackett
D. L.
,
Du
L.
,
Traganos
F.
,
Darzynkiewicz
Z.
,
Fojo
T.
,
Bates
S. E.
.
2002
.
Histone deacetylase inhibitors all induce p21 but differentially cause tubulin acetylation, mitotic arrest, and cytotoxicity.
Mol. Cancer Ther.
1
:
937
941
.
9
Görlich
D.
,
Kutay
U.
.
1999
.
Transport between the cell nucleus and the cytoplasm.
Annu. Rev. Cell Dev. Biol.
15
:
607
660
.
10
Fagerlund
R.
,
Kinnunen
L.
,
Köhler
M.
,
Julkunen
I.
,
Melén
K.
.
2005
.
NF-kappaB is transported into the nucleus by importin alpha3 and importin alpha4.
J. Biol. Chem.
280
:
15942
15951
.
11
McBride
K. M.
,
Banninger
G.
,
McDonald
C.
,
Reich
N. C.
.
2002
.
Regulated nuclear import of the STAT1 transcription factor by direct binding of importin-alpha.
EMBO J.
21
:
1754
1763
.
12
Ma
J.
,
Cao
X.
.
2006
.
Regulation of Stat3 nuclear import by importin alpha5 and importin alpha7 via two different functional sequence elements.
Cell. Signal.
18
:
1117
1126
.
13
Mackenzie
G. G.
,
Oteiza
P. I.
.
2007
.
Zinc and the cytoskeleton in the neuronal modulation of transcription factor NFAT.
J. Cell. Physiol.
210
:
246
256
.
14
Marchesi
J. R.
,
Holmes
E.
,
Khan
F.
,
Kochhar
S.
,
Scanlan
P.
,
Shanahan
F.
,
Wilson
I. D.
,
Wang
Y.
.
2007
.
Rapid and noninvasive metabonomic characterization of inflammatory bowel disease.
J. Proteome Res.
6
:
546
551
.
15
Treem
W. R.
,
Ahsan
N.
,
Shoup
M.
,
Hyams
J. S.
.
1994
.
Fecal short-chain fatty acids in children with inflammatory bowel disease.
J. Pediatr. Gastroenterol. Nutr.
18
:
159
164
.
16
Patz
J.
,
Jacobsohn
W. Z.
,
Gottschalk-Sabag
S.
,
Zeides
S.
,
Braverman
D. Z.
.
1996
.
Treatment of refractory distal ulcerative colitis with short chain fatty acid enemas.
Am. J. Gastroenterol.
91
:
731
734
.
17
Vernia
P.
,
Marcheggiano
A.
,
Caprilli
R.
,
Frieri
G.
,
Corrao
G.
,
Valpiani
D.
,
Di Paolo
M. C.
,
Paoluzi
P.
,
Torsoli
A.
.
1995
.
Short-chain fatty acid topical treatment in distal ulcerative colitis.
Aliment. Pharmacol. Ther.
9
:
309
313
.

The authors have no financial conflicts of interest.