Vol. 183, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
March 16. 1992
Pages
THE HUMAN
SERUM
AMYLOID
A LOCUS
362-366
SAA4
IS A PSEUDOGENE George
H.
Sack,
Jr.
and
C. Conover
Talbot,
Jr.
JohnsHopkins University School of Medicine, Departmentsof Medicine, Biological Chemistry and Pediatrics and the Kennedy Institute, Baltimore, MD
Received
January
14,
1992
\Ve have isolated the human genomic DNA clone GSAA4 from a size-selected Egf II library by hybridization to a probe derived from the human serum amyloid A gene CSAAl. Sequencing the 5’ end of this clone revealed a region similar to the first exon of gene GSAAl but with significant nuclcotide differences and mutation of the 3’ splice site. The restriction map of the GS,$AJ clone corresponds to that for the locus “SAA4” recently reported by others. Sequence and hybridization details indicate that the locus in clone GSAA4 is a member of the human serum amyloid A gene family and contains a pseudogene. Isolating GSLiAql conrp1cte.sthe colkction of clones needed to account for all bands found in blot hybridizations of human DNA using serum amyloid A gene probes. J 1992AcademicrJres*,Inc.
The serum amyloid A’ proteins are prominent in acute phase responsesera of humans and many other vertebrates. The function(s) of these proteins is(are) unknown although SAA may function as an apolipoprotein of high-density lipoproteins (1). One member of the SAA protein family appearsto be an autocrine inducer of collagenasein inflamed synovial cells (2, 3). Both the proteins (4) and basic gene structure (5-7) of SAA family membersare well conserved. The murine and human SAA genesare clustered in small, homologousregions on chromosomes7 and 1lp, respectively (6, 8, 9). The four membersof the murine SAA gene family have been isolated and sequenced(10); one is a pseudogene.Three members of the human SAA
gene family have now been sequenced(3, 5, 7); eachcontains a transcribable and
translatable sequence.Becauseour earlier hybridization studiesof human DNA (6) revealed a presumptive SAA locus that could not be accounted for on the basisof known genes we sought to isolate the correspondinggenomic clone. As our characterization of the clone for this locus
’ “Serum amyloid A” is hereafter cited as “SAA:
Vol.
183,
No.
BIOCHEMICAL
2, 1992
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
proceeded, we were heartened by the work of Betts et al. (7) describing an additional locus “SAA4”
- with restriction
sites similar to those in our new clone. We report here studies of
this new clone (which we have called “GSAA4”)
indicating that it contains the “SAA4”
locus
which is a pseudogene.
,\IETHODS Libmry prepar&on and screening To isolate the genomic locus corresponding to the 9 kb Bgl JI fragment identified in our earlier grnomic hybridizations (6) we constructed a sizeselected genomic library using the 4 - 10 kb BgZ II fragments of DNA from a Caucasian male. The fr-agment ends were partially filled-in using GTP, ATP and the Klenow fragment of DNA polymerase I. The cos sites of the vector, X ZAP 11 (Stratagene) were ligated and the resulting DNA was linearized by XJzo I digestion. The vector ends were partially filled-in with CTP, TIP and the Klenow enzyme. Vector and fragments were ligated and packaged with Gigapack Plus II (Stratagene). The recombinant phage were titsred and plated in E. coli PLK-F’ for screening by the method of Benton and Davis (11) and hybridization to 3’P-labcled probe derived from a 1.2 kb Eco Rl fragment containing the 3rd and 4th exons of genomic cIone GSAAl (3, 6). Reactive plaques were purified and subjected to phagemid rescue (M13-K07). Basic restriction maps were constructed for each clone and Pvu II and Xba I fragments were subcloned into pIBI vectors and grown in strain NM-522 for subsequent characterization. PCR Srrrrlies Oligonucleotides were prepared using an Applied Biosystems synthesizer. PCR reactions were performed using a Cetus thermal cycler and Tuq DNA polymerase with 10 min at 94”C, 30 cycles of 30 set at 94”C, 30 set at WC, 30 set at 72°C and 6 min at 72°C. Fragments were resolved by electrophoresis in a 2% NuSeive 11% SeaKern gel. DLVA Squencing Initial reactions were performed using the Sequenase kit (US Biochemical Carp) to orient the clones and establish basic landmarks. An Applied Biosystems Xlodel 373A sequencer also was used. Sequence analysis was performed using the program Microgenie (Beckman).
RESVLTS
AiVD DISCUSSION
Fig. 1 presents
the
restriction
map of clone GSAA4, emphasizing cleavage sites useful
for comparison mapping. The pattern is very similar to the regional map for the “SAA4”
locus
described by Betts et al. (7). Tmportantly, the latter locus is flanked by BgZ II sites - 9 kb apart which should yield the appropriate sized fragment that we identified in earlier blot hybridization studies (6). We concentrated our initial sequence studies on the leftward end of the clone as shown. The 707 bp region sequenced first showed (A + T) = 65% and contained a region very similar (o the 1st exon of the GSAAl
gene which reported earlier (3). However,
there are several nucleotide changes in the exon region and reading AG/GC
3’ donor splice site is aberrant,
instead of AG/GT.
To eliminate the possibility oligonucleotide
the
as shown in Fig. 2,
that
these nucleotide changes represented cloning artifacts,
primers were synthesized
fragment following
PCR amplification
as shown
in Fig. 2. These should give a 237 bp
of genomic DNA 363
if the sequence represents a true
Vol.
183,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
X
E
P
P
I
I
I
I ID
XHE
M
RESEARCH
X
PNH
S
I I
Ill
I
H
500
COMMUNICATIONS
bp
Figure 1. Restriction map of clone GSAA4. X = Xba-I, E = Eco Rl, P = Pst I, H = Hin dII1, M = S?naI, N = Nco I, S = SstI. Arrow indicatesregionof DNA sequence shownin Fig. 2.
genomic fragment. Fig. 3 shows that the PCR reaction synthesized the appropriate 237 bp fragment from DNA of 4 individuals of different ethnic backgrounds. Sequencingthe 237 bp
PCR fragment confirmed the sequence shown in Fig. 2 (data not shown). Interestingly,
the
reaction also synthesizeda fragment of approximately 610 bp. We cannot account for the larger band - the primers and stringenciesusedshould not recognize GSAAl (seeFig. 2B) or the other published SAA sequences(3, 5, 7). Thus, while GSAA4 - the clone described here - contains the “SAA4” locus, the larger PCR band may be derived from a related but yet unidentified locus. Interestingly, genomic hybridization studies using as probes either the 3’ end of the
CT
4 1
m
:::I:jQ/I::~[Q~
c
u
2. A. 707 nucleotidesof GSAA4 clone beginningat left sideof map shownin Fig. 1. Exon 1 shownunderlinedin bold face. PCRprimers(seetext for description)are underlined.B. Comparisonof sequences of GSAA4 and GSAAI clones;3’ splicesite indicatedby brackets.Identicalnucleotidesboxed. Heavyarrow = Exxon1; lighter linesshowPCR primers.Note that theseprimers cannotbe usedto amplify GSAAl. Figure
364
Vol.
183,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
3. PCR amplificationproductsusing primers shown in Fig. 2 and separatedby gel electrophoresis asdescribed.Standardsizesnotedin bp. Lanes l-4 correspondto amplificationsof DNA from Caucasian,Polish, Iraqi and SephardicJewishmales,rsspcctively;lane5 usedthe GSAA4 clone.The larger band(A) appearsat about30% of the densityof band(B) corresponding to the GSAA4 sequence. Figure
GSA,41 gene (6) or SAA cDNA (7) have not identified sequencescorresponding to the larger PCR fragment. Several featuresestablishthe nature of the locus in clone GSAA4. First, the locus falls \J,ithin a 9 kb BgZII fragment as we described (6) and asconfirmed by Betts ef al. (7). Second, the restriction enzyme cleavage map - particularly the characteristic cluster of Nco IIHin dIIII.%na I - corresponds to that reported for the SAA4 locus (7). Third, no sequences corresponding to a secondSAA exon were found in this region by Betts et al. (7) and we have found none in sequencingfurther 3’ in this clone. Fourth, a region very similar to the first exon of GSAAl
(and quite different from the first exon of the other human SAA genes [3]) is
present, although with several basechanges.
Fifth, the 3’ donor splice site at the end of the
exon region is mutated to AG/GC which is not recognizable as a consensusdonor sequence (12). Sixth, PCR ampIification of genomic DNA confirms that the aberrant exon region is not a cloning artifact but is co~xcrved
in humansof different ethnic groups. These findings support
the notion that the SAA4 locus is a conserved pseudogenein the human SAA gene family. The murine !OCUS @AA lacks first and secondexon sequences but contains someregions similar to the third intron (10). Our sequenceof GSAA4 shows similarity to the first intron of GSAAl
(3; see Fig. 2B) but not to the other human SAA genes (5, 7). This latter
observation is consistentwith the considerablesequencedivergence at the 5’ end of the gene and the N-terminal region of the predicted protein distinguishing GSAAl from the other human SAA loci (3). It also raisesthe possibility that the GSAA4 locus, which is well conserved (see Fig. 3) was derived from GSAAl. We reported (6) that all sequenceshybridizing to human SAA gene probes are found within a 90 kb Nor I fragment and the observations of Betts et al. (7) are compatible with this. 365
Vol.
183,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Current data indicate that all members of the human SAA gene family are closely linked on chromosome 1 lp. These findings establish the identity of all members of the human SAA gene family which are revealed by hybridization to genomic blots (6, 7). Recent observations (2, 3) that at least one human SAA protein is likely involved in mediating synovial inflammation indicate that further studies of this important group of genes and proteins will be important both for elucidating basic aspects of human gene structure and for understanding the physiology of SAA genes and proteins.
This work was supported, in part, by the National Foundation, March of Dimes (6-559) and by generous gifts from Mr. Daniel M. Kelly and Mr. and Mrs. William M. Griffin. Mrs. Mary A. Mix provided excellent secretarial assistance. REFERENCES 1.
Benditt, E.P., Eriksen, N., and Hanson, R.H. (1979) Proc. Natf. Acad. Sci. USA 76, 4092-4096.
2. 3. 4. 5. 6. 7.
Brinckerhoff, C.E., Mitchell, T.I., Karmilowicz, M.J., Kluve-Beckerman, B., and Benson, M.D. (1989) Sci~rce 243, 655-657. Sack, G.H. ,Jr., and Talbot, C.C. ,Jr. (1989) Gene 84, 509-515. Kushner, I. (1989) In TewBook of ~Plreuntarology pp. 719-727, W.B. Saunders, Philadefphia. Woo, P., Sipe, J.D., Dinarello, C.A., and Colten, H.R. (1987) J. Biol. Chem. 262, 15790-15795. Sack, G.H. ,Jr., Talbot, C.C.,Jr., Seuanez, H., and O’Brien, S.J. (1989) &and. J. Zmn~iinol. 29, 1 13-119. Betts, J.C., Edbrooke, M.R., Thakker, R.V. and Woo, P. (1991) &and. J. I~nmunol., 34, 471-482.
8. 9.
10. 11. 12.
Taylor, B.A., and Rowe, L. (1984) Mol. Gen. Genet. 195, 491-499. Yamamoto, K., Goto, N., Kosata, J., Mitomo, K., Shiroo, M., Migita, S., Nakayama, S., and Nastuume-Sakai, S. (1988) In: Amyloid and Amyloidosis. pp. 293-297, Plenum Press, New York. Lowell, C.A., Potter, D.A., Stearman, R.S., and Morrow, J.F. (1986) J. Biol. Chem. 261, 8442-8452. Benton, W.D., and Davis, R.W. (1977) Science 196, 180-182. Mount, S.M. (1982) Nucl. Acids Res. 10, 459-472.
366
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
March 16. 1992
Pages
THE HUMAN
SERUM
AMYLOID
A LOCUS
362-366
SAA4
IS A PSEUDOGENE George
H.
Sack,
Jr.
and
C. Conover
Talbot,
Jr.
JohnsHopkins University School of Medicine, Departmentsof Medicine, Biological Chemistry and Pediatrics and the Kennedy Institute, Baltimore, MD
Received
January
14,
1992
\Ve have isolated the human genomic DNA clone GSAA4 from a size-selected Egf II library by hybridization to a probe derived from the human serum amyloid A gene CSAAl. Sequencing the 5’ end of this clone revealed a region similar to the first exon of gene GSAAl but with significant nuclcotide differences and mutation of the 3’ splice site. The restriction map of the GS,$AJ clone corresponds to that for the locus “SAA4” recently reported by others. Sequence and hybridization details indicate that the locus in clone GSAA4 is a member of the human serum amyloid A gene family and contains a pseudogene. Isolating GSLiAql conrp1cte.sthe colkction of clones needed to account for all bands found in blot hybridizations of human DNA using serum amyloid A gene probes. J 1992AcademicrJres*,Inc.
The serum amyloid A’ proteins are prominent in acute phase responsesera of humans and many other vertebrates. The function(s) of these proteins is(are) unknown although SAA may function as an apolipoprotein of high-density lipoproteins (1). One member of the SAA protein family appearsto be an autocrine inducer of collagenasein inflamed synovial cells (2, 3). Both the proteins (4) and basic gene structure (5-7) of SAA family membersare well conserved. The murine and human SAA genesare clustered in small, homologousregions on chromosomes7 and 1lp, respectively (6, 8, 9). The four membersof the murine SAA gene family have been isolated and sequenced(10); one is a pseudogene.Three members of the human SAA
gene family have now been sequenced(3, 5, 7); eachcontains a transcribable and
translatable sequence.Becauseour earlier hybridization studiesof human DNA (6) revealed a presumptive SAA locus that could not be accounted for on the basisof known genes we sought to isolate the correspondinggenomic clone. As our characterization of the clone for this locus
’ “Serum amyloid A” is hereafter cited as “SAA:
Vol.
183,
No.
BIOCHEMICAL
2, 1992
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
proceeded, we were heartened by the work of Betts et al. (7) describing an additional locus “SAA4”
- with restriction
sites similar to those in our new clone. We report here studies of
this new clone (which we have called “GSAA4”)
indicating that it contains the “SAA4”
locus
which is a pseudogene.
,\IETHODS Libmry prepar&on and screening To isolate the genomic locus corresponding to the 9 kb Bgl JI fragment identified in our earlier grnomic hybridizations (6) we constructed a sizeselected genomic library using the 4 - 10 kb BgZ II fragments of DNA from a Caucasian male. The fr-agment ends were partially filled-in using GTP, ATP and the Klenow fragment of DNA polymerase I. The cos sites of the vector, X ZAP 11 (Stratagene) were ligated and the resulting DNA was linearized by XJzo I digestion. The vector ends were partially filled-in with CTP, TIP and the Klenow enzyme. Vector and fragments were ligated and packaged with Gigapack Plus II (Stratagene). The recombinant phage were titsred and plated in E. coli PLK-F’ for screening by the method of Benton and Davis (11) and hybridization to 3’P-labcled probe derived from a 1.2 kb Eco Rl fragment containing the 3rd and 4th exons of genomic cIone GSAAl (3, 6). Reactive plaques were purified and subjected to phagemid rescue (M13-K07). Basic restriction maps were constructed for each clone and Pvu II and Xba I fragments were subcloned into pIBI vectors and grown in strain NM-522 for subsequent characterization. PCR Srrrrlies Oligonucleotides were prepared using an Applied Biosystems synthesizer. PCR reactions were performed using a Cetus thermal cycler and Tuq DNA polymerase with 10 min at 94”C, 30 cycles of 30 set at 94”C, 30 set at WC, 30 set at 72°C and 6 min at 72°C. Fragments were resolved by electrophoresis in a 2% NuSeive 11% SeaKern gel. DLVA Squencing Initial reactions were performed using the Sequenase kit (US Biochemical Carp) to orient the clones and establish basic landmarks. An Applied Biosystems Xlodel 373A sequencer also was used. Sequence analysis was performed using the program Microgenie (Beckman).
RESVLTS
AiVD DISCUSSION
Fig. 1 presents
the
restriction
map of clone GSAA4, emphasizing cleavage sites useful
for comparison mapping. The pattern is very similar to the regional map for the “SAA4”
locus
described by Betts et al. (7). Tmportantly, the latter locus is flanked by BgZ II sites - 9 kb apart which should yield the appropriate sized fragment that we identified in earlier blot hybridization studies (6). We concentrated our initial sequence studies on the leftward end of the clone as shown. The 707 bp region sequenced first showed (A + T) = 65% and contained a region very similar (o the 1st exon of the GSAAl
gene which reported earlier (3). However,
there are several nucleotide changes in the exon region and reading AG/GC
3’ donor splice site is aberrant,
instead of AG/GT.
To eliminate the possibility oligonucleotide
the
as shown in Fig. 2,
that
these nucleotide changes represented cloning artifacts,
primers were synthesized
fragment following
PCR amplification
as shown
in Fig. 2. These should give a 237 bp
of genomic DNA 363
if the sequence represents a true
Vol.
183,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
X
E
P
P
I
I
I
I ID
XHE
M
RESEARCH
X
PNH
S
I I
Ill
I
H
500
COMMUNICATIONS
bp
Figure 1. Restriction map of clone GSAA4. X = Xba-I, E = Eco Rl, P = Pst I, H = Hin dII1, M = S?naI, N = Nco I, S = SstI. Arrow indicatesregionof DNA sequence shownin Fig. 2.
genomic fragment. Fig. 3 shows that the PCR reaction synthesized the appropriate 237 bp fragment from DNA of 4 individuals of different ethnic backgrounds. Sequencingthe 237 bp
PCR fragment confirmed the sequence shown in Fig. 2 (data not shown). Interestingly,
the
reaction also synthesizeda fragment of approximately 610 bp. We cannot account for the larger band - the primers and stringenciesusedshould not recognize GSAAl (seeFig. 2B) or the other published SAA sequences(3, 5, 7). Thus, while GSAA4 - the clone described here - contains the “SAA4” locus, the larger PCR band may be derived from a related but yet unidentified locus. Interestingly, genomic hybridization studies using as probes either the 3’ end of the
CT
4 1
m
:::I:jQ/I::~[Q~
c
u
2. A. 707 nucleotidesof GSAA4 clone beginningat left sideof map shownin Fig. 1. Exon 1 shownunderlinedin bold face. PCRprimers(seetext for description)are underlined.B. Comparisonof sequences of GSAA4 and GSAAI clones;3’ splicesite indicatedby brackets.Identicalnucleotidesboxed. Heavyarrow = Exxon1; lighter linesshowPCR primers.Note that theseprimers cannotbe usedto amplify GSAAl. Figure
364
Vol.
183,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
3. PCR amplificationproductsusing primers shown in Fig. 2 and separatedby gel electrophoresis asdescribed.Standardsizesnotedin bp. Lanes l-4 correspondto amplificationsof DNA from Caucasian,Polish, Iraqi and SephardicJewishmales,rsspcctively;lane5 usedthe GSAA4 clone.The larger band(A) appearsat about30% of the densityof band(B) corresponding to the GSAA4 sequence. Figure
GSA,41 gene (6) or SAA cDNA (7) have not identified sequencescorresponding to the larger PCR fragment. Several featuresestablishthe nature of the locus in clone GSAA4. First, the locus falls \J,ithin a 9 kb BgZII fragment as we described (6) and asconfirmed by Betts ef al. (7). Second, the restriction enzyme cleavage map - particularly the characteristic cluster of Nco IIHin dIIII.%na I - corresponds to that reported for the SAA4 locus (7). Third, no sequences corresponding to a secondSAA exon were found in this region by Betts et al. (7) and we have found none in sequencingfurther 3’ in this clone. Fourth, a region very similar to the first exon of GSAAl
(and quite different from the first exon of the other human SAA genes [3]) is
present, although with several basechanges.
Fifth, the 3’ donor splice site at the end of the
exon region is mutated to AG/GC which is not recognizable as a consensusdonor sequence (12). Sixth, PCR ampIification of genomic DNA confirms that the aberrant exon region is not a cloning artifact but is co~xcrved
in humansof different ethnic groups. These findings support
the notion that the SAA4 locus is a conserved pseudogenein the human SAA gene family. The murine !OCUS @AA lacks first and secondexon sequences but contains someregions similar to the third intron (10). Our sequenceof GSAA4 shows similarity to the first intron of GSAAl
(3; see Fig. 2B) but not to the other human SAA genes (5, 7). This latter
observation is consistentwith the considerablesequencedivergence at the 5’ end of the gene and the N-terminal region of the predicted protein distinguishing GSAAl from the other human SAA loci (3). It also raisesthe possibility that the GSAA4 locus, which is well conserved (see Fig. 3) was derived from GSAAl. We reported (6) that all sequenceshybridizing to human SAA gene probes are found within a 90 kb Nor I fragment and the observations of Betts et al. (7) are compatible with this. 365
Vol.
183,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Current data indicate that all members of the human SAA gene family are closely linked on chromosome 1 lp. These findings establish the identity of all members of the human SAA gene family which are revealed by hybridization to genomic blots (6, 7). Recent observations (2, 3) that at least one human SAA protein is likely involved in mediating synovial inflammation indicate that further studies of this important group of genes and proteins will be important both for elucidating basic aspects of human gene structure and for understanding the physiology of SAA genes and proteins.
This work was supported, in part, by the National Foundation, March of Dimes (6-559) and by generous gifts from Mr. Daniel M. Kelly and Mr. and Mrs. William M. Griffin. Mrs. Mary A. Mix provided excellent secretarial assistance. REFERENCES 1.
Benditt, E.P., Eriksen, N., and Hanson, R.H. (1979) Proc. Natf. Acad. Sci. USA 76, 4092-4096.
2. 3. 4. 5. 6. 7.
Brinckerhoff, C.E., Mitchell, T.I., Karmilowicz, M.J., Kluve-Beckerman, B., and Benson, M.D. (1989) Sci~rce 243, 655-657. Sack, G.H. ,Jr., and Talbot, C.C. ,Jr. (1989) Gene 84, 509-515. Kushner, I. (1989) In TewBook of ~Plreuntarology pp. 719-727, W.B. Saunders, Philadefphia. Woo, P., Sipe, J.D., Dinarello, C.A., and Colten, H.R. (1987) J. Biol. Chem. 262, 15790-15795. Sack, G.H. ,Jr., Talbot, C.C.,Jr., Seuanez, H., and O’Brien, S.J. (1989) &and. J. Zmn~iinol. 29, 1 13-119. Betts, J.C., Edbrooke, M.R., Thakker, R.V. and Woo, P. (1991) &and. J. I~nmunol., 34, 471-482.
8. 9.
10. 11. 12.
Taylor, B.A., and Rowe, L. (1984) Mol. Gen. Genet. 195, 491-499. Yamamoto, K., Goto, N., Kosata, J., Mitomo, K., Shiroo, M., Migita, S., Nakayama, S., and Nastuume-Sakai, S. (1988) In: Amyloid and Amyloidosis. pp. 293-297, Plenum Press, New York. Lowell, C.A., Potter, D.A., Stearman, R.S., and Morrow, J.F. (1986) J. Biol. Chem. 261, 8442-8452. Benton, W.D., and Davis, R.W. (1977) Science 196, 180-182. Mount, S.M. (1982) Nucl. Acids Res. 10, 459-472.
366
